当前位置:首页 >> 交通运输 >>

核电厂系统分析retran程序news


INSIDE
Featured Articles International RETRAN Meeting ............................ 1 Ken Moore Honored .......... 2 In Every Issue From the Editor................. R2 Trouble Reports ........... Tech Tip ............................ R3D Trouble Reports ........ Calendar............................ 3 3 4 5 6

Published by Computer Simulation & Analysis, Inc. for EPRI

International RETRAN Meeting
Garry C. Gose, CSA

The Tenth International RETRAN Meeting was held in Jackson Hole, Wyoming, during October 14-17, 2001. Fifty individuals representing 31 organizations from Canada, Japan, Korea, Switzerland, Spain, Taiwan and the United States participated in the meeting. The participants included members of electrical utilities, government research agencies, research laboratories, and independent consultants. One objective of these international meetings is to provide a forum for the exchange of information concerning the EPRI sponsored computer programs, RETRAN, CORETRAN and GOTHIC. Discussions of analysis results, the use of the programs for new applications, and transient analysis modeling techniques were elements of many of the presentations. This meeting included a panel discussion on the recent NRC review of RETRAN-3D and the interpretation of the Safety

Evaluation Report (SER). Following the opening remarks from Mr. Jack Haugh of EPRI, Mr. Lance Agee, the RETRAN Project Manager at EPRI, delivered the key address of the plenary session. His presentation summarized some of the significant EPRI activities in the nuclear safety analysis area. This presentation included information of the EPRI Strategic Bridge Plan to support the industry's vision of nuclear power expansion for 2002. A specific item to support this plan is the role of RETRAN-3D in the area of Risk Informed Regulation. He presented his vision of the future for applications, and identified important new application areas. The plenary session continued with an invited panel discussion, "RETRAN-3D SER Review Process". The panel was comprised of EPRI, Electric Utility, and code development individuals. The session involved participation by all

the attendees in the question and answer session that followed the panel discussion. The technical sessions for RETRAN included presentations of RETRAN-3D analysis results as well as some of the current development activities. Technical papers included descriptions of analyses performed in support of plant operations as well as for licensing, topical reports, and plant support activities. Papers were presented on the use of RETRAN-3D as an analytical engine for real time simulators, an exciting and significant new application area for the code.

T h e R E T R A N N e w s l e t ter – January 2002

1

International RETRAN Meeting (Cont'd)
The third day focused mainly on the CORETRAN and GOTHIC programs. One new application area for CORETRAN is to support the operational problem of Axial Offset Anomaly (AOA). Presentations by APS and CSA reported on the progress made in this effort in both the analysis and model development areas. The final session was devoted to the GOTHIC code and many presentations were made on the modeling capabilities and applications of this multidimensional flow analysis tool. The continuing interest in RETRAN, CORETRAN and GOTHIC, as demonstrated by the participation in the meeting and the variety of presentations, is encouraging to EPRI and CSA, the co-sponsors of the meeting. The Proceedings of this meeting, including complete papers of the presentations and a list of participants, will be distributed to those individuals who attended the meeting. Others may obtain a copy of the Proceedings from the Research Reports Center.

Ken Moore Honored at the Tenth International RETRAN Meeting
Garry C. Gose, CSA

A Pioneer in the development of Thermal-Hydraulics Analysis codes was recognized for his achievements during a special award ceremony at the Tenth International RETRAN Meeting. Ken Moore, who was a principal developer of the RELAP code and who later went on to the same role in the RETRAN code was honored for his pioneering work. A Certificate of Appreciation for his lifetime of dedication to nuclear safety analysis was presented to Ken by Jack Haugh and Lance Agee of EPRI. Following the award presentation, Ken gave a talk about the early days of code development.

Presented to

Ken Moore
For a lifetime of dedication to nuclear safety analysis and for pioneering the development of RELAP and RETRAN.

Jack Haugh Area Manager Nuclear Safety & Analysis

Lance J. Agee RETRAN Project Manager

2

The RETRAN Newsletter – January 2002

Summary of RETRAN-02 Trouble Reports
The following is a summary of RETRAN-02 Trouble Report/Code Maintenance Activity as of December 31, 2001. There are five outstanding trouble reports. A list of trouble reports and the status can be obtained directly from the EPSC (1-800-763-3772). Additional information is available from the RETRAN-02 Trouble Report Page at http://www.csai.com/retran/r02trpt/index.html. NO. TROUBLE REPORT TYPE OF PROBLEM COMMENTS 354 376 394 408 442 Large Step Change in PHIR Control Reactivity, No Motion Anomalous Heat Trans. Behavior OTSG Heat Transfer Problems Poor Diagnostics CORRECTION NO. IDENT *** *** *** *** *** ******* ******* ******* ******* *******

From the Editor
This has been a significant year for the RETRAN project. With the issue of the RETRAN-3D SER, many groups can now go forward with plans to migrate to the newer code from the earlier RETRAN-02 versions. This is a goal that we all have been working towards for some time. This issue of the newsletter contains a summery of the 10th international meeting which was another very successful gathering of the RETRAN community. It is always good to see colleagues and friends from all parts of the globe exchanging information and renewing relationships. Two items from the meeting particularly struck me. First was an inspiring after dinner presentation given by EPRI’s Jack Haugh on the significance and impact of what we nuclear safety analysts do. The idea that we are doing something important in the world is sometimes lost in the daily grind. His thoughts and ideas brought a big picture focus to what we are really accomplishing in this industry. The second item was a post presentation talk given by Ken Moore (see the related article in this issue) on the early days of code development. Ken was one of my mentors in my early days as a code jock and even then he had a worldwide reputation as one of the best. Listening to his recollection of how things were and remembering what it was like in the “good old days” was pretty emotional. Part of this was because I realized that I was involved in part of this ancient technology, a little unnerving I must admit. But on the heels of Jack Haugh’s remarks, I felt myself wondering where we would all be and what we would be doing 10 to 15 years from now.

T h e R E T R A N N e w s l e t ter – January 2002

3

Storing A Value With A RETRAN Control System
J. G. Shatford, CSA

One of the most powerful features of the RETRAN code is the control system. We use control systems for many purposes, from simple things like converting a flow to gallons per minute to sophisticated control systems such as complex feedwater controllers. Control systems are highly versatile, but occasionally there are limitations which seem to prevent us from performing the most basic of tasks. One such task is saving a value for later use. A recent project involved a control system model that was dependent upon the normalized prompt power. The control system was designed such that the initial value of prompt power needed to be saved for later use. Saving an initial value seems like a common item; many RETRAN models contain control systems that use initial values. But most involve user interaction by assuming initial value is already known which is ‘manually’ put in the input file. For these control systems, even small changes in the input model require the user to go back and reevaluate all of the initial conditions. There is an easier way to solve this problem in many cases, based on the trick we describe below. The description of a delay block (DLY) states that the output of the block is: y(t) = y0 for 0 < t < τ .

If a time constant, τ, is larger than the transient time is used, the output of the DLY is always the initial value. The initial value is saved and can be obtained as the output from the DLY block. But why stop there? This idea can be extended to save a value at any point in the transient. The time constant for a DLY block can be variable, defined by the output of another control block If the time constant is zero then there is no delay and the input is just passed through. On the other hand, if the time constant is suddenly increased to a very large value, the output of the block is frozen and the value at that point is saved. As an example, let us define Control Block -9 as a delay on the normalized power (Control Input 8) where the time constant, τ, is defined by Control Input 9 as shown in Figure 1. The number of samples for Control Block –9 is set to 1. Control Input 9 samples Trip ID 4, which is the reactor trip. Since the value of the time constant is zero until the reactor trip, initially the delay block is transparent and the output of Control Block -9 tracks Control Input 8. When the reactor trips, the large gain on Control Input 9 makes the time constant big and the output of Control Block -9 is frozen. Figure 2 demonstrates that this simple trick has saved the value of normalized power at the time of the reactor trip. So, with the simple addition of a DLY block we can use the output as a storing device for any control system variable.

STEAMLINE BREAK - RETRAN-3D MODEL -HALF CORE, 19 CHANNEL (19FILL BC) 15/12/01 16:56:32 RETRAN-3D/MOD003.1 28/08/01 EPRI
1.60

1.40

8 PNRM 0 9 TRIP 4

-9
1.20

DLY
OUTPUT

1.00

:CBLK

8

:CBLK -9
0.80

G = 1.E+5

τ

= COUT 9

0.60

0.40

0.20

0.00 0.0 5.0 10.0 15.0
SYSTEM ELAPSED TIME (SEC)

20.0

25.0

30.0

Figure 1. Flow Logic for Variable Time Constant Delay Block

Figure 2. Normalized Power Saved at Time of Reactor Trip

4

The RETRAN Newsletter – January 2002

Summary of RETRAN-3D Code Trouble Reports
A total of 255 trouble reports had been filed as of December 31, 2001. Of these, 224 reports have been resolved, while 31 remain unresolved. A summary of the unresolved trouble reports is shown below. Additional information for RETRAN-3D trouble reports is available at http://www.csai.com/retran/r3dtrpt/index.html.

NO. 30 40 48 52 54 60 70 81 116 122 152 168 170 181 200 201 202 203 212 226 240 246 247 248 249 250 251 252 253 254 255

TROUBLE REPORT TYPE OF PROBLEM 2-loop Oconee w/5-eq. fails in steady state Results do not agree with data Steady state fails after 6 iterations MOC does not return to the initial temp. MOC solution; no null transient for two-phase Anomalous countercurrent flooding Fails in subroutine DERIVS Steady-state failure at iteration #6 Fails in steady-state initialization Problems with EOS convergence Junct pressure lags vol pressure 1 time step Incorrect null trans w/3d Kin. mod ht & 5eq PARCS numerics will not hold a null transient No rod cusping treatment in 3D kinetics SS failure for NCG (WAT0 error maybe WAT17) SS failure when flow split option used Error when pcrit reached during tran – 5-Eq Pressurizer time step selectn when Przr solid Possible errors in dynamic flow regime model MOC error when flow reverses Junction property error after 92sec transient Control system floating point exception Scratch space overlaps with other FTB files Critical heat flux exceeded in SG secondary PNM floating parameter defaults not set on PC Index used before defined - Zolotar-Lellouche Slip vel. incorrect for TDV with multiple con. Volume description incorrectly printed Concentration instability in surgeline press. Model cannot hold a null transient Segmentation Fault in 3D Kinetics Case

CORRECTION NO. IDENT *** *** *** 006 *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ******* ******* ******* MOD001g ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* *******

COMMENTS

(partial fix)

(water packing) Model limitation Preliminary update

The RETRAN Newsletter – January 2002

5

About This Newsletter
RETRAN Maintenance Program
The RETRAN Maintenance Program is part of a program undertaken by EPRI to provide for the support of the software developed in the Nuclear Power Division. The main features of the Subscription Service include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN.

Newsletter Contributions
The RETRAN Newsletter is published for members of the Subscription Service program. We want to use the newsletter as a means of communication, not only from EPRI to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. The next newsletter is scheduled for May 2002 and we would like to include a brief summary of your RETRAN activities. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to the E-mail addresses below by May 10, 2002. Contributors of a feature article will receive a RETRAN polo shirt. We are looking forward to hearing from all RETRAN licensees. Mark Paulsen Garry Gose Pam Richardson paulsen@csai.com gcg@csai.com pam@csai.com (208) 529-1700

The RETRAN Maintenance Program now has 26 organizations participating in the program, including 22 U.S. utilities and 4 organizations from outside of the U.S. A Steering Committee, composed of representatives from the participating organizations, advises EPRI on various activities including possible enhancements for the code and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from Lance Agee EPRI P. O. Box 10412 Palo Alto, CA 94303 lagee@epri.com or (650) 855-2106

The RETRAN Web Page is located at http://www.csai.com/retran/index.html. Previous issues of the RETRAN Newsletter are available from the RETRAN Web Pages at http://www.csai.com/retran.

EPSC Contacts
EPSC 3412 Hillview Ave. Palo Alto, CA 94307-1395 Hours: 9 a.m. to 8 p.m. EST To Order EPSC Software: (800) 313-3774 EPSC Fax: (650) 855-1026 To Order RETRAN Products contact Colette Handy via email chandy@epri.com

Calendar of Events
RETRAN Training Course: June 2002

User Group Meeting: Fall 2002

Please supply us with technical tips for our section and you will receive a RETRAN mouse pad.
Your contributions are greatly appreciated. We, EPRI and CSA, encourage everyone to participate in this newsletter.

6

The RETRAN Newsletter – January 2002

INSIDE
New Organizations Join RVUG .......................... KEPRI Group Sponsors RETRAN Training ............. PC Version of VIPRE-01 MOD2.2 Available .............. Tech Tip .................................. Steering Committee Members ............................. About this Newsletter ............ Calendar of Events ................ 2 2 3 3 4 4 4

Published by Computer Simulation & Analysis, Inc.

November 2005 RETRAN and VIPRE User Group Meeting Held in Palo Alto
The fall 2005 RETRAN/VIPRE User Group (RUG) Meeting was hosted by EPRI in Palo Alto, California. Mr. Gregg Swindlehurst, Steering Committee Chairman, welcomed the attendees and opened the meeting, which was attended by EPRI, eight U.S. utilities, one U.S. commercial vendor, four international organizations, and CSA. The meeting was held during the same week as the EPRI-sponsored FALCON user group meeting resulting in a good representation for both code groups. The user group meeting was split between RETRAN and VIPRE project management issues and technical presentations. CSA presented a summary of the status of the RETRAN project for 2005, including the current membership roster, project revenue from membership fees, and year-todate expenses. The code maintenance work performed for the year was discussed along with work scope items for the remainder of the year. Preliminary budget, fee structure, and candidate work scope items for 2006 were identified. The 3% user group fee increase approved in the May 2005 RVUG meeting was translated to a funding formula that was approved by the membership. CSA distributed a handout and presented a six-year plan for RETRAN development tasks. A summary of the VIPRE project was also presented by CSA. It included revenues from maintenance fees and carryover from EPRI maintenance for ∑ prior years, year-todate expenses, and a summary of the work completed. CSA identified potential tasks for development work for 2006. Members were asked to provide additional work scope items of interest so they can be finalized. As part of a joint meeting with the FALCON user group, Suresh Yagnik of EPRI presented an overview of FALCON (Fuel Analysis and Licensing Code). Utilities made formal and summary presentations of RETRAN and VIPRE activities: ∑ Two presentations were made by staff members from Duke Energy. - The first presentation dealt with the UFSAR Chapter 15 Margin Recovery. - The second presentation discussed Simulation/ Evaluation of Oconee Nuclear Station Unit 3 8/31/05 Reactor Trip Event. A presentation was made by Iberdrola that addressed the RETRAN-3D analysis of Cofrentes NPP CRDA, Cold and Hot Conditions. TXU presented the Uncertainty of Initial RCS Temperature on Transient Analysis.

Gregg Swindlehurst, Duke Energy ∑ Westinghouse presented the RAVE methodology application for AP1000 plant design. .

Sama Bilbao y Leon was unanimously elected by the membership to replace Todd Flowers as a Steering Committee Member. Current Steering Committee members are: Gregg Swindlehurst, Duke (Chairman) Andres Gomez Navarro, Iberdrola Sama Bilbao y Leon, Dominion David Huegel, Westinghouse Adi Irani, ENN The spring RVUG meeting will be held in Ogden, Utah, in early May. Thanks to EPRI for their hospitality and willingness to host the RVUG Meeting.



The RETRAN/VIPRE Newsletter – January 2006

1

New Organizations Join RVUG
Two organizations have joined the RVUG in 2005. Genden Information System Company (GISC) is currently interested in using RETRAN-3D to support the Tokai 2 plant and will be very active in the near future. Mr. Hiroshi Kawai of GISC attended a RETRAN training session in Idaho Falls, Idaho, in 2005. Two individuals from GISC attended the November 2005 RVUG meeting in Palo Alto. Dr. Noburu Katayama and Mr. Takauki Negishi traveled from Tokyo, Japan, to attend the RVUG meeting and the associated FALCON meeting at Palo Alto. Arizona Public Service has joined both the RETRAN and VIPRE user groups in 2005. APS will use the codes to support the Palo Verde units.

Noburu Katayama and Takauki Negishi, GISC

FPL has joined the VUG for 2005. They plan to use VIPRE-01 for their Seabrook unit in support of their AOA evaluation effort.

KEPRI Group Sponsors RETRAN Training
A KEPRI RETRAN training session was held in Daejeon, Korea (Republic), on October 4-6, 2005. The sessions were held at the training facilities of KAERI International Nuclear Training and Education Center. The session was sponsored by KEPRI and 15 individuals from the KEPRI organization attended. KEPRI also invited representatives from the Korea Institute of Nuclear Safety (KINS), the Korean Nuclear Licensing Authority. The three-day session consisted of a combination of RETRAN theory lectures combined with sample problem sessions designed to give the individuals some practical "hands-on" experience with the RETRAN code. Garry Gose, CSA, Presenting Lecture at KEPRI Training Facilities

2

The RETRAN/VIPRE Newsletter – January 2006

PC Version of VIPRE-01 MOD2.2 Available
A new version of VIPRE-01 is available from CSA for VUG members using Windows XP. It contains Modifications 233 through 247, which include eight error corrections, modifications to remove platform dependencies and ease code maintenance, and six new code features. This version, designated as MOD2.2, satisfies the quality assurance requirements of 10CFR50 Appendix B. The new features include: ? ? ? ? ? ? ? an option to generate a file to interface with the EPRI BOA code, improved input processing that identifies input errors that previously resulted in run-time errors, an option to read a new VIPRE Boundary Condition (VBC) file generated by RETRAN-3D, an extension to the nuclear rod model input that provides for detailed pellet nodalization with variable mesh size and burnup dependent material properties, an MDNBR iteration option on inlet temperature/enthalpy where volumetric flow remains constant, an option to allow input in GPM, and clean-up so user-selected output files are only generated if requested (restart, plot, microfiche, CHF summary, auxiliary edit).

The PC transmittal is comprised of a PC executable file, a license file, a batch file to run the code, sample problem input and output files, an automated compare program to test installations against baseline results, and Revision 5.0 of the documentation in Acrobat PDF format. Automated installation procedures are provided to install the code.

New Control System Capability Saves Time and Allows More Elements
RETRAN-3D MOD004.1 has some control system features that will be of interest to all users. There is no longer an upper limit of 999 to the number of control block elements. And now users can manipulate large sums and limits without adding a lot of complexity to the input model. By using new control system input and new control block data records, 702XXXX and 703XXXX, up to 9999 elements can now be described. These can be used in combination with the original 702XXX and 703XXX data records so minimal impact on existing input models is seen. New control block elements are available in RETRAN-3D MOD004.1 that allow users to model more with less, so to speak. The new "Super" blocks allow the user to work with arrays of inputs to perform algebraic operations. These blocks, Super Sum, Super Minimum, and Super Maximum, are designated by the symbols SSM, SMN, and SMX, respectively. The blocks allow the user to describe a series of inputs and the output is the summation, the maximum, or the minimum of all the input data. The SSM, summation block simplifies input files because the user is no longer required to build a 'cascade' of summation blocks for multiple inputs as the original SUM of two inputs required. The SMN and SMX super minimum and maximum blocks allow new flexibility in processing data. For example one can now find the maximum metal temperature in a core, the maximum pressure on any number of variations with just a few simple input additions.
The RETRAN/VIPRE Newsletter – January 2006 3

About This Newsletter
RETRAN Maintenance Program
The RETRAN/VIPRE Maintenance Program is a program that provides for the support of software developed and maintained by CSA. The main features of the Subscription Service include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN and VIPRE.

Newsletter Contributions
The RETRAN/VIPRE Newsletter is published for members of the Subscription Service program. We want to use the newsletter as a means of communication, not only from CSA to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. The next newsletter is scheduled for July 2006 and we would like to include a brief summary of your RETRAN and VIPRE activities. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to the email addresses below by July 3, 2006. Contributors of a feature article will receive a RETRAN polo shirt. We are looking forward to hearing from all RETRAN and VIPRE licensees. Mark Paulsen Garry Gose Pam Richardson paulsen@csai.com gcg@csai.com pam@csai.com

The RETRAN Maintenance Program now has 19 organizations participating in the program, including 13 U.S. utilities and 6 organizations from outside of the U.S. Ten U.S. utilities are currently participating in the VIPRE maintenance program. A Steering Committee, composed of representatives from the participating organizations, advises CSA on various activities including possible enhancements for the code and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from Mark P. Paulsen Computer Simulation & Analysis, Inc. P. O. Box 51596 Idaho Falls, ID 83405 paulsen@csa.com or (208) 529-1700

The RETRAN web page is located at http://www.csai.com/retran/index.html. The VIPRE web page is located at http://www.csai.com/vipre/index.html Previous issues of the RETRAN/VIPRE Newsletter are available from the RETRAN or VIPRE web pages.

Steering Committee Members
Gregg Swindlehurst, Duke Energy (Chairman), gbswindl@duke-energy.com Andres Gomez Navarro, Iberdrola, agn@iberinco.com Sama Bilbao y Leon, Dominion, Sama_Bilbao@dom.com David Huegel, Westinghouse, huegelds@westinghouse.com Adi Irani, Entergy Nuclear Northeast, airani@entergy.com

Cal en d ar o f Even ts
User Group Meeting: May 2006 Ogden, Utah Details will be emailed to Maintenance Group Members

4

The RETRAN/VIPRE Newsletter – January 2006

Published by Computer Simulation & Analysis, Inc. for EPRI

The RETRAN Newsletter
Summary of Activities This issue of the RETRAN Newsletter contains articles on the NRC Review of RETRAN-3D, an article on the CPM-3/CORETRAN Workshop, and interesting articles from code users. Your contributions are greatly appreciated. We, EPRI and CSA, encourage everyone to participate in this newsletter. Previous issues of the RETRAN Newsletter are available from the RETRAN Web Pages at http://www.csai.com/retran.
In This Issue... NRC Review of RETRAN-3D Underway . . 1 Utility Support Expedites RETRAN-3D Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Duke Energy Updates to RETRAN-02 MOD005.2 . . . . . . . . . . . . . . . . . . . . . . . . . 3 Summary of RETRAN-3D Code Trouble Reports . . . . . . . . . . . . . . . . . . . . 4 Summary of RETRAN-02 Trouble Reports . . . . . . . . . . . . . . . . . . . . 5 RETRAN-3D Cross-Section Model to be Changed . . . . . . . . . . . . . . . . . . . . . . . . . . 6 CPM-3/CORETRAN Workshop . . . . . . . . . 7 Hurray! New RETRAN Graduates! . . . . . 8 INER/TPC RETRAN-3D Training Seminar 9 TechTips . . . . . . . . . . . . . . . . . . . . . . . . . . 10 About this Newsletter . . . . . . . . . . . . . . . . 11

January, 1999

NRC Review of RETRAN-3D Underway
Gregg Swindlehurst, Duke Energy Mark P. Paulsen, CSA A major milestone in the life cycle of the RETRAN-3D computer program was reached in November 1998 when the U.S. Nuclear Regulatory Commission started a formal review of the code. Initial meetings have been held with the NRC staff and more recently with the ACRS Subcommittee on Thermal/Hydraulic Phenomena. The current review schedule is aimed at issuing an SER in October 1999. In early July, Gregg Swindlehurst of Duke Energy and Chairman of the RETRAN Maintenance Group, sent a letter to the NRC requesting a review of the RETRAN-3D computer program. The request for review was accompanied by a list of 18 organizations that expressed support for the review [see Utility Support Expedites RETRAN-3D Review]. RETRAN-3D is the third generation of the RETRAN code series developed for the utility industry by EPRI and provides additional modeling capability compared with the NRCapproved RETRAN-02 computer program. Some of the new models include: C C C C multidimensional neutron kinetics, nonequilibrium thermodynamics, noncondensable gas flow, and implicit numerical solution methods.

In addition to these new models, a number of the RETRAN-02 models have been extended to broaden and enhance the analysis capabilities of the code. These new models and extensions will enable users to better simulate plant and system transient response with enhanced accuracy and efficiency. (continued on page 5)

1

Utility Support Expedites RETRAN-3D Review
Mark P. Paulsen, CSA The RETRAN Maintenance Group Chairman, Gregg Swindlehurst of Duke Energy, coordinated a request to the U.S. Nuclear Regulatory Commission to review the RETRAN-3D computer program [see article NRC Review of RETRAN-3D Underway]. Due to Mr. Swindlehurst’s efforts and the support of the maintenance group members, the review is in progress with completion scheduled for October 1999. One of the key factors influencing the NRC’s recent decision to review the RETRAN-3D computer program, was the number of organizations that expressed support of the review. These organizations include: Baltimore Gas & Electric Detroit Edison Duke Energy Illinois Power Northern States Power Pennsylvania Power & Light TU Electric Virginia Power Wisconsin Public Service Commonwealth Edison Duke Engineering & Services GPU Nuclear New York Power Authority PECO Nuclear South Texas Project Nuclear Operating Co. Union Electric Washington Public Power Supply System Wolf Creek Nuclear Operating Corp.

Some of these organizations either had made or were planning to make submittals to the NRC using RETRAN-3D to take advantage of the new or extended models. Duke Energy was one of these organizations and they have since received permission to use RETRAN-3D, in the RETRAN-02 mode, to perform analyses currently being done with RETRAN-02. Other organizations are awaiting review and approval before committing to transition to the new code. The successful effort to initiate the review of the RETRAN-3D computer program demonstrates an important benefit to all RETRAN users afforded by the RETRAN Maintenance Group.

Calendar
If there are any items you would like to have added to our calendar, please contact Garry Gose or Pam Richardson.

2

Duke Energy Upgrades to RETRAN-02 MOD005.2
Christy L. Ray, Duke Energy Duke Energy recently upgraded from RETRAN-02 MOD005.1 to RETRAN-02 MOD005.2 and has begun using RETRAN-02 MOD005.2 in licensing applications. The version of RETRAN-02 MOD005.2 being used by Duke Energy contains two additional modifications that were developed by CSA but have not yet undergone formal verification under the RETRAN-02 QA procedures. Since these modifications fall outside of the maintenance program discussed above, any licensing basis analyses performed with this version of the code must clearly identify these modifications in any submittals to the NRC. The first modification allows access to the condensation heat transfer correlations with the use of the forced convection heat transfer map. The second modification corrects an error in the calculation of the liquid region work term in the pressurizer model (Trouble Report No. 443). In order to validate the RETRAN-02 MOD005.2 computer code under the Duke Energy software quality assurance program, several transients that had been previously evaluated with RETRAN-02 MOD005.1 were analyzed with RETRAN-02 MOD005.2. The results from the two code versions were then compared to demonstrate the predictive capabilities of the RETRAN-02 MOD005.2 code. This approach was chosen to help determine if any of the error corrections made to the RETRAN-02 MOD005.1 code to produce RETRAN-02 MOD005.2 had a significant impact on any of the licensing basis analyses. The transients analyzed were: C C C C C C McGuire/Catawba Increase in Feedwater Flow McGuire/Catawba Uncontrolled Bank Withdrawal McGuire/Catawba Locked Rotor Oconee Turbine Trip Oconee Steam Line Break with OneDimensional Kinetics Oconee Steam Generator Tube Rupture

C C C

Oconee Large Steam Line Break with Condensing Heat Transfer Oconee Tornado Analysis McGuire/Catawba Steam Line Break Mass and Energy Release

For most transients, there was very good agreement between the RETRAN-02 MOD005.1 results and the RETRAN-02 MOD005.2 results. For a few transients, minor differences were seen between the results obtained with the two code versions. However, decreasing the maximum time-step size resolved most of the differences. The time-step sensitivity was most noticeable in cases with significant pressurizer relief (e.g., pressurizer PORV cycling). Thus, regardless of the code version being used, the maximum time-step size should be selected carefully to ensure a converged solution. The Oconee large steam line break analysis uses the forced convection only heat transfer map, and thus it was used to test the modification added to access the condensing heat transfer correlations. The overall results showed good agreement between RETRAN-02 MOD005.1 and RETRAN-02 MOD005.2. The only noticeable difference seen, due to the use of the condensing heat transfer option, was lower pressure in the intact steam generator as expected. Thus, the condensing heat transfer option appears to be working as intended RETRAN-02 MOD005.1 contains a code error (Trouble Report 426) involving interpolation between heat transfer mode 11 (free convection nucleate boiling) and mode 14 (free convection to vapor). This error was apparent in the McGuire/Catawba steam line break mass and
(continued on page 4)

3

Duke Energy Upgrades to RETRAN-02 MOD005.2 (Cont'd)
energy release analysis in which the steam generator tube bundle uncovers. The error that was seen with RETRAN-02 MOD005.1 is corrected in RETRAN-02 MOD005.2 and eliminates the spikes in primary-to-secondary heat transfer in the affected loop. The RETRAN-02 MOD005.2 case showed less overall primary-tosecondary heat transfer in the affected loop which impacted reactor power, pressurizer pressure, pressurizer level, and affected loop hot leg temperature. However, these differences had a negligible overall impact on the break flow rate and enthalpy. Transients with prolonged tube bundle uncovery may need to be examined to ensure that the error correction involving the interpolation between heat transfer modes 11 and 14 does not impact analysis results obtained using RETRAN-02 MOD005.1. The Oconee tornado analysis simulates the loss of main and emergency feedwater in conjunction with a loss of offsite power. For this particular transient, there were noticeable differences between the RETRAN-02 MOD005.1 and RETRAN-02 MOD005.2 results. The error correction that resulted in these differences has not been identified. RCS pressure and pressurizer level were lower for the RETRAN-02 MOD005.2 case due to greater relief through the pressurizer safety valves. Thus, it appears that the error corrections made to RETRAN-02 MOD005.1 to produce RETRAN-02 MOD005.2 may have an impact on transients with significant pressurizer relief, especially for those transients that experience substantial primary system voiding. It is also noted that this was the only transient examined that showed any impact from the modification to correct the error in the pressurizer liquid region work term.

Summary of RETRAN-3D Code Trouble Reports
A total of 154 trouble reports had been filed as of December 31, 1998. Of these, 138 reports have been resolved, while 16 remain unresolved. A summary of the unresolved trouble reports is shown below. Additional information for RETRAN-3D trouble reports is available at http://www.csai.com/retran/r3dtrpt/index.html.
NO. 22 30 40 48 52 54 60 70 81 116 122 142 144 145 150 152 TROUBLE REPORT TYPE OF PROBLEM Problem using Wilson bubble rise model & error when using low power initialization 2-loop Oconee w/5-eq. fails in steady state Results do not agree with data Steady state fails after 6 iterations MOC does not return to the initial temp. MOC solution; no null transient for two-phase Anomalous countercurrent flooding Fails in subroutine DERIVS Steady-state failure at iteration #6 Fails in steady-state initialization Problems with EOS convergence Timestep selection causes 3-D kin to fail TAUGL model doesn't apply for horiz. flow SS fails to converge for low press. and flow SS solution void fraction oscillation Junct pressure lags vol pressure 1 time step CORRECTION NO. IDENT *** *** *** *** 006 *** *** *** *** *** *** *** *** *** *** *** *** ******* MOD001 ******* ******* ******* MOD001g ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* COMMENTS

(partial fix)

(partial fix)

(water packing)

4

Summary of RETRAN-02 Trouble Reports
The following is a summary of RETRAN-02 Trouble Report/Code Maintenance Activity. There are 12 outstanding trouble reports, 2 from MOD004 and 10 from MOD005. A list of trouble reports and the status can be obtained directly from the EPSC (800-763-3772). Additional information is available from the RETRAN-02 Trouble Report Page at http://www.csai.com/retran/r02trpt/index.html.
NO. 354 376 394 408 439 440 442 443 444 445 446 447 TROUBLE REPORT TYPE OF PROBLEM Large Step Change in PHIR Control Reactivity, No Motion Anomalous Heat Trans. Behavior OTSG Heat Transfer Problems Decay Heat Input Kinetic Energy/Time Dep Area Poor Diagnostics Liquid Region Work Term Positive Slip Velocity Boron Transport Inconsistency Theory Manual for Bubble Rise Smoothing Algorithm in SVOID CORRECTION NO. IDENT *** *** *** *** *** *** *** *** *** *** *** *** ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* COMMENTS

TH Manual Modification

NRC Review of RETRAN-3D Underway (Cont'd)
After a preliminary review of the four-volume RETRAN-3D documentation, the NRC staff requested a meeting to discuss the submittal and planned review process. A meeting was held on September 15, 1998. At the meeting Mr. Swindlehurst discussed the review request and the role of the RETRAN Maintenance Group. Lance Agee of EPRI and Mark Paulsen of CSA presented an overview of RETRAN-3D including some RETRAN-02 background, the design review, discussion of the new models and model extensions, code assessment, and maintenance procedures. The NRC staff presented a preliminary review schedule for determining if the RETRAN-3D documentation was adequate for the review to continue. In early December, Mr. Swindlehurst was notified of NRC’s decision to proceed with the review. The review will be preformed in-house by staff members in the Reactor Systems Branch headed by Ralph Caruso, with Ralph Landry as the technical lead. While the ACRS Subcommittee on Thermal/Hydraulic Phenomena has previously been involved in reviewing LOCA-related submittals, they will also be participating in the RETRAN-3D review, a first for transient analysis codes. A meeting with the ACRS Subcommittee on Thermal/Hydraulic Phenomena was held December 17, 1998, where an overview was given of the RETRAN-3D computer code similar to that presented to the NRC staff in September. The NRC staff presented their review plan and review schedule that targets October 1999 for the issuance of the RETRAN-3D SER. The transcript for this meeting can be obtained at http://www.nrc.gov/ACRS/rrs1/Trans_Let/index_top/ACRS_sub_tran/Thermal_Hydr_Phen/th981217. The NRC is focusing on doing a well-managed review and will be working with the ACRS in an attempt to limit their requests for additional information to a single round of questions. The presentations were well received by the ACRS. Completion of the review will represent a significant achievement in the development of RETRAN-3D, allowing code users to use the advanced analytical capabilities to address emerging safety and licensing issues. 5

RETRAN-3D Cross-Section Model to be Changed
Garry C. Gose, CSA The next release of RETRAN-3D (RETRAN-3D MOD003) will include many new model features and the area of multidimensional kinetics modeling will be included. RETRAN-3D will be coupled more closely with the upstream physics codes, as the figure below indicates, relying upon them for cross-section data (the BXF file) as well as geometric data (the CDI file). The previous cross-section models will be replaced. Recent versions of CORETRAN (1.45 or greater) use a ‘TABLES’ form of the cross-section model used to represent the group constants for the ANM diffusion solution. The new format has been described and documented in other publications so it will not be repeated here (Eisenhart, et al., Proceedings of Eighth International RETRAN Meeting, EPRI-TR-106038, March 1996 and CORETRAN-01 Theory and Numerical Analysis, Volume 1, October 1997). The new formalism is no longer a polynomial representation but is instead a data base of cross-section dependencies that will allow the generation of cross sections based upon a variety of historical and instantaneous parameters using interpolation and partial cross-section summation. The format is passed on to RETRAN-3D and a modified format that contains only instantaneous data dependencies (as opposed to both historical and instantaneous) are written for the RETRAN-3D kinetics model. The older, ‘TAPE67' (based on polynomials), format that exists in RETRAN-3D MOD002.0 will no longer be supported. Support for the older SIMULATE-E model will also be removed in an effort to simplify the code maintenance efforts.

User Input

Engineering Data

Lattice Physics

Engineering Code Pre-Processor

CDI File

CORETRAN

Channel Model Input

RETRAN-3D

Binary Cross-Section File

Current RETRAN-3D Channel Model Data Flow The move to the new cross-section model will allow a more direct coupling to the CORETRAN code and will allow the user to spend less time constructing cross sections and more time analyzing kinetics cases. 6

CPM-3/CORETRAN Workshop
John L. Westacott, CSA A CPM-3/CORETRAN Workshop was held on October 27-30, 1998, at Rockville, MD. The goal of the workshop was to familiarize utility reactor physics engineers with the EPRI lattice physics code, CPM-3, and the core simulation code, CORETRAN-01. CORETRAN-01 includes steady-state depletion, operational transient, and design basis transients capabilities. Four EPRI contractors presented the workshop which was attended by 11 US and international utility organizations. During the morning lectures, code capabilities and applications were summarized, theory presented, testing and validation results presented, and input requirements outlined. In the afternoon, hands-on sessions were used to demonstrate the code's capabilities with various sample problems. Ten high-performance PCs and one IBM RISC workstation were provided. The hands-on sessions were especially popular. Attendees were able to observe and modify the CPM-3 benchmark lattice physics PWR and BWR cases on Tuesday. CORETRAN core depletion demonstration problems were used for the Wednesday session. On Thursday, CORETRAN transient models were provided. A sample problem was also provided which demonstrated the crosssection generation by CORETRAN, with the transient calculation performed using RETRAN-3D. Cross-section and core geometry data were passed directly from CORETRAN to the RETRAN-3D code. Preprocessors were demonstrated and available for use with the sample problems. The PISCIS interface for CORETRAN allows input manipulation and generation, and graphical representation of results. The RETRAN preprocessor was used to demonstrate automated core mapping of neutronics channels to thermalhydraulic channels for RETRAN-3D system transient calculations. The workshop was successful in introducing and demonstrating the capabilities of CPM-3 and CORETRAN to a number of organizations. The codes were well received and the hands-on sessions were particularly effective (the sessions often ran past the scheduled time). Both CPM-3 and CORETRAN-01 are in the late beta test stage and are to be formally released in 1999.

Please supply us with technical tips for our TechTips section and you will receive a RETRAN mouse pad. pad.
7

Hurray! New RETRAN Graduates!
Two RETRAN training sessions were held at CSA's office in Idaho Falls during September. A basic RETRAN training session was held during the week of September 14, 1998, followed by an advanced RETRAN training session the next week. The basic session lectures covered the theoretical basis of the RETRAN code including the balance equations, constitutive, and component models. Other topics included the selection of input options, common modeling practices, the interpretation of results, and common pitfalls and their resolution. Part of each afternoon was devoted to working sessions where the attendees prepared input for sample problems designed to illustrate material covered in the previous lectures. The lectures also presented comparisons of the RETRAN-02 and RETRAN-3D codes and the improved analysis capability of RETRAN-3D. Attendees were also given the opportunity to use the RETRAN input preprocessor and the PEGASYS plotting software during the problem sessions. Congratulations to the following Basic Training participants:
Robert Lee, Commonwealth Edison Timothy Byam, Illinois Power David Balfour, Northeast Utilities Sunyoung Kwon, Northeast Utilities Keith Higar, Northern States Power Melissa Higar, Northern States Power Dixon Yee, Pacific Gas & Electric Jeffrey Bonner, Pennsylvania Power & Light Owen Stevens, Public Service Electric & Gas Mark Strohecker, Public Service Electric & Gas Shih-Fang Huang, STP Nuclear Operating Co. John Fisher, Washington Public Power Supply System

Lectures for the advanced RETRAN training session were designed for experienced users and provided information about advanced modeling methods that were beyond the scope of the basic course. The new modeling capabilities of the RETRAN-3D code were presented, including the revised balance equation set, improved solution methods, and new models. Modeling recommendations were discussed for both BWR and PWR applications. These included implications arising from the RETRAN-02 SER/TER and RETRAN-3D activities that have addressed specified SER/TER limitations. The final day of the session was devoted to the theory, use, and qualification of the multidimensional kinetics model in RETRAN-3D. Each afternoon, participants were able to run their own plant models using RETRAN-3D, and many used the RETRAN input preprocessor to convert existing RETRAN-02 input decks to RETRAN-3D. During the afternoon work sessions, individualized consultation was provided by CSA staff members. The following participants are congratulated for their successful completion of the Advanced RETRAN session.
Robert Lee, Commonwealth Edison Enrique Vela, Consejo de Seguridad Nuclear Jennifer Furl, Public Service Electric & Gas Kent Halac, Public Service Electric & Gas Javier Iglesias, UITESA/IBERINCO John Fisher, Washington Public Power Supply System Linda Woosley, Washington Public Power Supply System Steven Chen, Wolf Creek Nuclear Operating Corp.

8

INER/TPC RETRAN-3D Training Seminar
James H. McFadden, CSA The Institute of Nuclear Energy Research (INER) and Taiwan Power Company (Taipower) have joined the group of RETRAN-3D users. A training seminar for analysts from these two organizations was conducted in November 1998 at the office of INER in Lungtan, Taiwan, ROC. The seminar included general sessions as well as several special sessions. On the first day, a special introductory session was conducted which gave an overview of RETRAN-3D and the NRC review of the program. This session was chaired by Dr. Yuan-Ching Chou and was attended by about 60 engineers and analysts from INER, Taipower, and the Taiwan regulatory agency as well as university faculty members. The final day included a special session on quality assurance and configuration control. About 20 people attended the general sessions. The topics discussed included C C changes in RETRAN-02 models, the nonequilibrium field equation option and related constitutive models,

C C C

the noncondensable gas option and associated constitutive models, the three-dimensional neutron kinetics option, and general RETRAN modeling practices for BWR and PWR applications.

Dr. Jan-Ru Tang (INER) and Mr. Julian Chiang (Taipower) organized the seminar, which was conducted by Mark Paulsen and Garry Gose of CSA. INER and Taipower are long time users of RETRAN-02 and participants in the RETRAN project, and the RETRAN bibliography includes many reports from these organizations. In the nearterm, they plan to focus on BWR stability and ABWR analyses with RETRAN-3D.

9

TechTips

Converting RETRAN-02 Control Systems to RETRAN-3D

Sometimes when a RETRAN-02 control system model is converted and solved by RETRAN-3D, the new model may not converge. This problem can sometimes be overcome by increasing the number of iterations on the 701000 data. This, however, can increase running time and some control blocks still may not converge. These nonconvergence problems are typically related to blocks that use feedback signals. Another solution is to replace direct feedback signals with the output of a signal that is passed through a COUT block. Why does the RETRAN-3D control solver fail to converge occasionally? Isn’t it a newer and improved solver? Well yes it is and that’s part of the answer. In the RETRAN-02 code, the control blocks are evaluated using an explicit solver in the order that they are numbered from input. In RETRAN-3D, the solution method has been revised and the output of all blocks are evaluated at the same time level. This results in a system of coupled equations that are solved simultaneously. A Gauss-Seidel iterative solution is used, rather than a direct matrix solver. The Gauss-Seidel method converges well as long as the system of equations is diagonally dominant. However, when an off-diagonal element in the matrix is dominant the method may converge slowly. Feedback signals can introduce off-diagonal elements that are large compared with diagonal elements leading to nonconvergence of the Gauss-Seidel solution scheme. Passing the feedback signal through a COUT block places the value at the old time and effectively moves the element from the coefficient matrix to the right-hand side. This eliminates the nonconvergence problems caused by feedback signals. The example shown below illustrates the situation.

Original RETRAN-02 Control System

-60 FNG

Modified RETRAN-3D Control System

-60 FNG

-61 LAG

-61 LAG

502 COUT -65

-62 SUM

-62 SUM

501 COUT -64

-63 SUM

-63 SUM

-64 INT

-64 INT

-65 INT

-65 INT

-66 FNG

-66 FNG

10

About This Newsletter
RETRAN Maintenance Program
The RETRAN Maintenance Program is part of a program undertaken by EPRI to provide for the support of the software developed in the Nuclear Power Division. The main features of the Subscription Service include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN.

Newsletter Contributions
The RETRAN Newsletter is published for members of the Subscription Service program. We want to use the newsletter as a means of communication, not only from EPRI to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. The next newsletter is scheduled for March 1999 and we would like to include a brief summary of your RETRAN activities. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to the E-mail addresses below by March 5, 1999. Contributors of a

The RETRAN Maintenance Program now has 31 organizations participating in the program, including 23 member utilities, 5 organizations from outside of the U.S., and 3 nonmember utilities from the U.S. A Steering Committee, composed of representatives from the participating organizations, advises EPRI on various activities including possible enhancements for the code and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from Lance Agee EPRI P. O. Box 10412 Palo Alto, CA 94303 lagee@epri.com or (650) 855-2106

feature article will receive a RETRAN polo shirt. We are looking forward to hearing from all
RETRAN licensees. James McFadden Mark Paulsen Garry Gose Pam Richardson mcfadden@csai.com (208) 529-1700 paulsen@csai.com gcg@csai.com pam@csai.com

The RETRAN Web Page is located at http://www.csai.com/retran/index.html.

EPSC Contacts
EPSC Hours: EPSC Hotline: EPSC Fax: 7 a.m. to 8 p.m. EST (800) 763-3772 (619) 453-4495

For Nuclear Quality Assurance related questions, call Clark Wallace at (619) 622-6611.

11

INSIDE
RETRAN 2010 Work Summary ................................. Tech Tip ....................................... RETRAN Development Activities for 2011 ................... 2011 RVUG Membership Fees .. RETRANUI Version 2.5.0 ......... VIPRE-01 Development Activities for 2011 ................... New VIPRE-01 Code Release .... RETRAN Visualization Tool ..... Steering Committee Members... About this Newsletter ................. Calendar of Events ..................... 1 4 4 5 5 5 6 6 7 7 7

Published by CSA Inc.

RETRAN-3D Work Summary for 2010
A summary of RETRAN-3D related activities for 2010 is given below 3-D-Kinetics RESTART While restart has never been functional for any version of RETRAN-3D running 3-D kinetics, it is a necessary feature since many 3-D kinetics cases can take a long time to run. To improve the usefulness of the 3-D kinetics option, restart capability has been added. This allows users to continue (restart) 3-D kinetics cases, which terminate intentionally or unintentionally before the required solution period is complete. Restart problems continue a terminated problem solution from the original problem data stored in the restart file, which can save valuable time and resources. The standard PWR 3-D Kinetics Sample Problem was used to test the functionality of RESTRT saving a restart file with the sample problem run. A RESTRT was then run by restarting partway through the original solution using the time-step, trip, and minor edits from the original RETRAN-3D run. No changes were made to boundary conditions or control actions. The results shown below illustrate that the RESTRT results (restarted at 0.1 seconds), are essentially identical

13.00

11.00

NORM. POWER

9.00

7.00

5.00

3.00

PWR: SYSTEM
1.00 0.0000

RST: SYSTEM
0.0500 0.1000 0.1500 0.2000 0.2500 0.3000 (SEC) 0.3500 0.4000 0.4500 0.5000

SYSTEM ELAPSED TIME

RESTRT and RETRAN Results Comparison for the PWR Sample Problem

The RETRAN & VIPRE Newsletter – February 2011

1

RETRAN-3D Work Summary for 2010 (Cont'd)
to the original results between 0.1 and 0.5 seconds. Similar agreement was obtained for the other 3-D kinetics sample problems. Improved RETRAN-3D INPUT Processing Error Messages As input is processed within RETRAN-3D, various routines examine the input to insure that it is formatted correctly, that all required data is supplied, and that the input values are consistent with physical and known limits. While RETRAN-3D writes an error message(s) when an error is encountered, the messages are often terse and in some instances misleading. There is no guidance as to how to correct the error. Several years ago, the error messages associated with the transient solution were revised. The error messages were revised to aid code users (many of the original error messages contained information that was only useful to a programmer). An appendix in the User’s Manual – Volume 3, contains a detailed description of the error, possible causes and recommended action to eliminate the error. The recent effort has revised the input processing error messages to use a similar error handling scheme or process where the error messages are written to the output file and errlog files are processed and edited as they occurred. The input processing error messages have been grouped by category and each is assigned a unique error message. The errors are categorized according to component type or specific model option. The messages have been revised to be more informative to the code user. An attempt has been made to eliminate and/or revise confusing messages. Each message has a more detailed description in Appendix C of the User’s Manual which includes a discussion of specific action that can be taken to eliminate the error. Short Form Problem Description Data Replacement Cards The input for the Problem Control and Description Data Card 01000Y was revised at MOD004.1 to allow a short form that removed all dimensions and many option flags from the card series. A maximum of 12 data items are supplied on the new form compared with the 42 that were originally allowed. Use of the short form input simplifies use of the code and avoids many commonly encountered errors associated with adding or removing a component or minor edit and not changing the corresponding input on the 01000Y data card. While the option has been available for some time, many users’ models still use the classic form of input where up to 42 variables are defined. To simplify migration to the short form, a new output edit was added to the output file immediately following the edit for the problem control and description data. The edit includes the replacement cards required to convert the deck to use the short form of the problem control and description data. The replacement cards can simply be copied into the clipboard and pasted at the end of the RETRAN-3D input deck. RETRAN-3D Trouble Report Resolution Thirty-five trouble reports have been filed since the release of MOD004.5f95. Three trouble reports were still unresolved when MOD004.5f95 was released. All 38 trouble reports have been resolved and will be included in MOD004.6f85. 64-Bit RETRAN-3D Windows 7 Application The Windows version of RETRAN-3D MOD004.5f95 was released as a 32-bit application, which should run on both 32- and 64-bit processors. However, it included license control software that used a 16-bit dll, which does not function on 64-bit machines. When the 32-bit RETRAN-3D application is run on a 64-bit machine, the following error message is issued “Cannot execute machnm1.exe!”. For most organizations the error does not affect execution of RETRAN-3D other than the fact that a response is required to clear the message, after which the code runs normally. If an automated procedure is being used to run multiple cases, it will require a response for each case. This error would be fatal for organizations whose license locks the RETRAN-3D installation to a particular computer, but none have reported the problem. The Protection Plus (PP) license control software used with RETRAN-3D MOD004.5f95 required upgrading for use with 64-bit installations on Windows 7. This required revisions to C++functions used to implement the licensing features, including the new PP model for uniquely identifying hardware platforms. The revised code supports use of a license file to uniquely identify and validate organizations licensed to use RETRAN-3D. Various types of license restriction models need to be supported. They include (1) licenses where no restrictions are made for installation hardware or expiration, (2) licenses where the executable is locked to a particular computer, (3) licenses that are give an expiration date after which the installation will not run, and (4) and combination of 2 and 3. Both 32-bit and 64-bit implementations of the C++ interface libraries were tested and are now available for use with the next release of RETRAN-3D. During the implementation of the revisions for the 32- and 64-bit locked RETRAN-3D applications, the Intel Fortran compiler was successfully migrated from Version 10.1 to Version 11.1. This compiler version will be used for new releases of the Windows and Linux code versions.

2

The RETRAN & VIPRE Newsletter – February 2011

RETRAN-3D Work Summary for 2010 (Cont'd)
RETRAN-3D MOD004.6 Code Version A new RETRAN-3D code version will be released soon. It will include the error corrections and code modifications described above. The release will support the HPUX, AIX, and Solaris, Red Hat Linux and Windows XP or 7 platforms. Both 32-bit and 64-bit applications will be provided for the Windows installation. They will also employ software locking similar to previous releases. Two transmittal CD-ROM formats will be available, one for Windows only installations that does not include source code and a second for UNIX/Linux platforms with executable files and source code. It also includes the PC application installation. All four volumes of the documentation will be updated for the new version. See the discussion below for the update of Volume 4. Assessment Manual - Volume 4 Analyses During RETRAN-3D development and maintenance efforts, verification and validation is used to make certain the software satisfactorily performs all intended functions. As RETRAN-3D was developed, the analyses included in the Assessment Manual – Volume 4 were run in support of the validation effort. Since that effort was an ongoing work, several different code versions were used to produce the results. Initially versions MOD001f and MOD002 were used. Some analyses were re-run with later versions, primarily MOD003 but MOD003.1 for a few. As part of the MOD004.5f95 validation effort, 13 separate effects analyses and 35 system effects analyses form the Assessment Manual were used to validate the Fortran 95 conversion effort. This effort required revising the input for the new code version and it also discovered some errors that had to be corrected before the analyses could be completed. They were valuable in validating the code. As a result of this validation effort, it became apparent that the analyses presented in the Assessment Manual needed to be updated using the current code version. This effort would insure that the validation problems could be run on the current code version and that the results presented are true representations of the solution from the current code. All of the Assessment Manual cases for which CSA has input decks have been re-run. Some of the decks in the Assessment Manual are not available because the analyses were performed by organizations that participated in the original validation of RETRAN-3D and the decks were not provided to CSA. This was an extensive effort that involved running 177 validation cases representing a wide combination of individual models, components, and numerical techniques. All of these cases have been re-run using MOD004.5f95 with various error corrections for some of the analyses. As soon as MOD004.6f95 is released, they will be re-run on this code version. These results will then be plotted against the experimental data and other code results as necessary. The update plots will be included in the Assessment Manual.

The result of this effort will be an Assessment Manual where most of the results are representative of the current code. The results that cannot be rerun due to the lack of an input deck will be retained in the manual and the code version on which they were run will be identified. Efforts will be made in the future to locate the decks and update the results. As part of this effort, procedures have been developed to simplify re-running the validation cases on a new code version.

The RETRAN & VIPRE Newsletter – February 2011

3

get_R3D_Plot_Vars Utility Program
RETRAN-3D MOD004.5f95 contains a new feature to optionally create an output file named R3D_PLOT that can be used to provide solution results to plotting or postprocessing applications. It is described in Section IV.2.8 of the Programmer’s Manual – Volume 2. A utility program, get_R3D_Plot_Vars, has been developed to allow solution results to be extracted from the R3D_PLOT file. The information to be extracted is requested using minor edit variable request pairs (variable flag and region number) and the format of the new output file is user defined. An option allows get_R3D_Plot_Vars to create an auxiliary file (TAPE60). Appendix E of the Programmer’s Manual – Volume 2, describes the use of the get_R3D_Plot_Vars utility program. The source code for the utility is provided with the transmittal package.

Possible RETRAN Development Activities for 2011
A list of development activities have been sent to the RUG members, with a request for additional suggestions if there is something they would like to see done that is not on the current list. Responses will be added to the full list, which will be prioritized by the RUG Steering Committee. Items of interest from prior years included: ?

?

?

New User Features ? Continue Preparation of a RETRAN-3D User Guidelines Document ? Add Enhanced Steady-State Error Messages and Corrective Action ? Add New Control Elements ? Add New Minor Edit Variables and/or User-Supplied Minor Edit Labels Solution Enhancements ? Implement a Volume versus Height Model for Separated Volumes ? Develop a Pressurizer Thermal Stratification Model ? Implement a Water Packing Mitigation ? Implement a Taitel-Dukler Flow Regime Map for Dynamic Slip ? Design and Implement a VIPRE-Based 3-D Core Model ? Develop and Implement a Six-Equation Model New Validation Analyses ? Five-Equation Studies for Cold Water Injection into Two-Phase Fluid ? RETRAN-3D SBLOCA Benchmark to Test Facility Data

4

The RETRAN & VIPRE Newsletter – February 2011

2011 RVUG Membership Fees
RVUG membership fees for 2011 were set by the Steering Committee. Their decision was to keep the fees at the 2010 level rather than increasing them by 3% as had been done the past few years. Membership notices were sent to past members in December of 2010.

RETRAN User Interface, Version 2.5.0 Coming April 2011
CSA periodically updates the RETRANUI to resolve trouble reports and provide user requested enhancements. Version 2.5.0 is a new release of the RETRANUI that contains new features and error corrections. New features and enhancements for Version 2.5.0 are summarized below. New Features and Enhancements ? ? ? ? ? ? ? Added a new Stop button to the toolbar. When pressed, it results in the problem being terminated as if an end problem trip were encountered and all edits and auxiliary files are closed normally. Compatible with Windows 7 and Excel 2007. When using change decks, the output files are now saved in the change deck directory instead of the input file directory. When setting up an executable file, you now have the ability to type in or select the executable location and the local temporary files directory. Added REEDIT and GENRST to tape60 file types. Added the capability to connect to a remote UNIX/Linux machine using either a REXEC or SSH connection. Extended FTP functionality to include the use of SSL connections.

The RETRANUI Version 2.5.0 User's Guide has been updated to include descriptions and instructions on how to use the new features. Version 2.5.0 of the RETRANUI is available to organizations that have purchased an annual maintenance subscription. Subscriptions include software updates and telephone or e-mail support for one year. Organizations purchasing the RETRANUI receive a subscription to the maintenance group for one year from the purchase date, after which an annual membership may be purchased. The annual fees provide support and funds for additional RETRANUI development.

Possible VIPRE Development Activities for 2011
A list of development tasks has been sent to the VUG members, with a request for additional suggestions if there is something they would like to see done that is not on the current list. All responses have been received. Responses will be added to the full list, which will be prioritized by the VUG Steering Committee. Items of interest from prior years included: ? ? ? ? ? ? VIPRE-01 Benchmark Based on NUPEC PWR Subchannel Bundle Test (PSBT) MDNBR Search Improvement Improve Input Error Messages Review SI Options Conversion Multiple Case Options - External File for Power Data Revised VIPRE-01 Programmer's Manual

If you do not have access to the detailed task descriptions, please contact Garry Gose, gcg@csai.com.

The RETRAN & VIPRE Newsletter – February 2011

5

New VIPRE-01 Code Transmittal MOD02.4F95 Released February 2011
A new VIPRE-01 code transmittal has been developed and sent to RUG members during February 2011. VIPRE-01 MOD02.4F95 is the most recent VIPRE-01 MOD02 code version to be issued to the VIPRE User Group. VIPRE-01 MOD02.4F95 is the result of a VUG funded multi-year project that modernizes the VIPRE-01 source code and data storage using up-to-date Fortran language constructs. VIPRE-01 MOD02.4F95 also contains corrections for trouble reports and VUG requested enhancements. ? New CHF correlations can be developed using a dynamic link library (DLL) feature as a way to add third party, proprietary correlations to VIPRE-01 without recompiling the VIPRE-01 source. VIPRE-01 run-time errors are categorized and summarized in a post-execution error summary file. Information is given about the type of error and location in the model. User guidelines are written that describes the error, the probable cause, and potential user action. An optional drift flux option based on the RETRAN-3D implementation of the Chexal-Lellouche drift flux model has been added. Code modifications that correct ten VIPRE-01 trouble reports have been added. The trouble reports were filed since the release of VIPRE-01 MOD02.3 (September 2007).

?

?

?

Here are some of the significant new features of VIPRE-01 MOD02.4F95: ? Dynamic or execution time data array allocation. This should significantly reduce (or eliminate the need for recompiling the code as new options or more detailed input models are developed. Another benefit of the dynamic memory allocation project is a significant reduction in the restart file size.

The formal code transmittal of VIPRE-01 MOD02.4F95 is a CDROM consisting of source, documentation, test cases, and installation. VIPRE-01 MOD02.4F95 is supported Windows XP/VISTA/7, IBM, SUN, and HP UNIX platforms and PC compatible machines operating under Linux. For more information regarding VIPRE-01 MOD02.4F95 contact: Garry C. Gose (208) 529-1700, Ext. 22, or gcg@csai.com

?

RETRAN Visualization Tool
A preliminary version of the RETRAN Visualization Tool (RVT) was unveiled at the May 2010 RUG meeting as a “work in progress.” Based on the feedback, the RVT scope was expanded to include a plant view that displays various regions as different colors and shades along with dials, gauges, recorders, digital channels, and other indicators for displaying data. Hence its release has been delayed, but is now nearing completion. An online demonstration will be available soon. An updated description of the RVT can be found at http://www.csai.com/rvt/index.html.

6

The RETRAN & VIPRE Newsletter – February 2011

About This Newsletter
RETRAN/VIPRE Maintenance Programs
The RETRAN/VIPRE Maintenance Programs provide support of software developed and maintained by CSA. The main features of the maintenance programs include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN and VIPRE.

Newsletter Contributions
The RETRAN/VIPRE Newsletter is published for members of the Maintenance Program. We want to use the newsletter as a means of communication, not only from CSA to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. We would like to include a brief summary of your RETRAN and VIPRE activities on the next newsletter. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to one of the email addresses below by December 9, 2011. We are looking forward to hearing from all RETRAN and VIPRE licensees. Mark Paulsen Garry Gose Pam Richardson paulsen@csai.com gcg@csai.com pam@csai.com

The RETRAN and VIPRE Maintenance Program both have 15 organizations participating, including 11 U.S. utilities and four organizations from outside of the U.S. A Steering Committee, composed of representatives from the participating organizations, advises CSA on various activities including possible enhancements for the codes and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from: Mark P. Paulsen or Garry C. Gose CSA Inc. P. O. Box 51596 Idaho Falls, ID 83405 paulsen@csai.com or gcg@csai.com (208) 529-1700

The RETRAN web page is located at http://www.csai.com/retran/summary.html. The VIPRE web page is located at http://www.csai.com/vipre/summary.html Previous issues of the RETRAN/VIPRE Newsletter are available from the RETRAN or VIPRE web pages.

Steering Committee Members
John Lautzenheiser, Dominion (Chairman), john_lautzenheiser@dom.com Kurt Flaig, Dominion, Kurt_Flaig@dom.com Jeff Abbott, Duke Energy, jlabbott@duke-energy.com Jorge Arpa, FPL, jorge_apra@fpl.com Rafael de la Fuente Frutos, Iberinco, rff@iberinco.com Tsutomu Ikeno, NFI, t-ikeno@nfi.co.jp Daren Chang, STPEGS, dchang@stpegs.com

Calendar of Events
User Group Meeting May 24 & 25, 2011 Salt Lake City, Utah http://www.csai.com/retran/rvug/ugm.html Basic RETRAN Training Session June 13-17, 2011 Idaho Falls, Idaho http://www.csai.com/retran/summary.html

The RETRAN & VIPRE Newsletter – February 2011

7

INSIDE
Dominion Hosts RVUG Meeting in Richmond ........................ 3 VIPRE-01 Modifications....... 6 Tech Tip.................................. 7 Steering Committee Members ............................. 9 About this Newsletter ............ 9 Calendar of Events ................ 9

Published by Computer Simulation & Analysis, Inc.

RETRAN Modification Removes Enthalpy Transport Model Limitation
The RETRAN enthalpy transport model is applied in heated volumes to account for the enthalpy distribution that occurs when a uniform heat flux is applied over the length of the volume. Use of this model yields more realistic volume temperature distributions for use in single-phase heat transfer calculations or a more accurate quality, which gives a more accurate mass distribution. The enthalpy transport option is frequently used in the reactor core region and steam generators. While the model is very useful, it can have limitations in application. A Limitation in the Current Model One well known limitation is that the enthalpy transport model can predict anomalous superheating when it is used in a tube bundle region that dries out. The problem occurs when the resulting wall heat flux yields superheated vapor at the exit junction. It is possible for the model to predict exit junction steam superheating hundreds of degrees higher than the temperature of the corresponding primary volume. This is nonphysical. 1200 The adjacent figure shows a typical result from the situation described above. In the figure, temperatures for the secondary-side of a U-tube steam generator are shown during a tube uncovery transient. Junctions 471 through 474 represent the exit flow paths for the secondaryside volumes in the steam generator. The problem is observed at about 210 seconds when the lower region (471) becomes highly voided and the junction temperatures begin to increase. The temperature in Junction 471 increases very rapidly, ultimately superheating by hundreds of degrees. Note that all of the secondary temperatures exceed the hot leg (Volume 420) temperature, while they should be lower. The temperature distribution is also inverted in that
1100

the Junction 471 temperature should be the lowest rather than the highest. The aphysical secondary-side temperatures cause 'reverse' heat transfer to occur, where the secondary-side volumes actually lose heat to the primary-side. There are two issues that lead to the secondary temperature exceeding that of the corresponding primary volume. The first is the assumption that the heat addition source term in the enthalpy transport equation is treated as a known quantity. In reality, it is dependent on the junction enthalpy that is being computed, a limitation that is amplified when the conditions in the tube region change from two-phase to single-phase vapor. The other issue is that the heat transfer within a volume assumes a twophase heat transfer regime, even though single-phase vapor conditions exist at the exit.

1000

Junction 471 Junction 472 Junction 473 Junction 474 Volume 420

Temperature (F)

900

800

700

600

500

400 0 50 100 150 200 Time (sec) 250 300 350 400

SGTB Uncovery with Enthalpy Transport (Original)

The RETRAN & VIPRE Newsletter – March 2007

1

RETRAN Modification Removes Enthalpy Transport Model Limitation (Cont'd)
There are at least two distinct types of heat transfer occurring in such a volume, one that is driven by the saturation temperature (two-phase) and the other (singlephase) that is driven by the average temperature of the superheated vapor, which directly depends on the heat addition to the single-phase subregion. Simplifying assumptions in the standard enthalpy transport model neglect this coupling. The New Model: A Subdivided Node The anomalous superheating behavior is eliminated and accuracy improved by a new enthalpy transport approach that dynamically subdivides a secondary node undergoing a transition from two-phase to superheated conditions. The model computes the length of the two-phase and superheated subregions within the secondary volume and applies appropriate heat transfer fluid boundary conditions to the subregions. The model is activated when enthalpy transport is active at a junction and when the donor volume is two-phase but the exit junction is superheated steam (as determined from the standard enthalpy transport model). Heat conductor(s) associated with the two-phase volume are required to be two sided for the current implementation. The left boundary volume(s) will be single-phase liquid (steam generator tube volumes). The right boundary will be the secondary volume. The adjacent figure illustrates a cross section of a typical steam generator model bundle region that is one node high. It is comprised of a volume representing the tubes on the upflow side, a volume modeling the tubes on the downflow side, a single secondary-side volume, and the two-sided heat conductors that allow for heat transfer between the tube volumes and the secondary-side volume. In the new model, when the secondary volume exit junction dries out (superheats), the axial cross-section nodes are divided into two regions, a two-phase region (Region 2), and a superheated steam region (Region 1). Additional heat conductors (auxiliary conductors) are added automatically to Region 1. The base conductor will be applied over the two-phase length A2 and the auxiliary heat conductors over the superheated length (1-A2). Separate mass and energy equations are solved over the subregions defined by A2. The model assumes: (1) (2) (3) (4) (5) (6) volumes are oriented vertically, heat conductors are two-sided, the two-phase volume is the right boundary for the heat conductors, the tube volumes are on the left surface and they will be single-phase liquid, axial heat conduction can be neglected, the subregion (two-phase) length can be applied to the tube volumes and associated heat conductors, all liquid in the secondary-side volume is below the dryout point, and all associated exit junctions use enthalpy transport superheated length (1-A2).

(7) (8)

When the two-region model is selected via user input, it is inactive until the normal enthalpy transport model predicts superheated steam for the junction exiting from a twophase volume. The volume must also satisfy the geometry and modeling assumption stated above.

Steam

(1 ? A 2 ) Region 1

Two-Phase

A2

Region 2

U-Tube Upflow

Secondary Volume

U-Tube Downflow

Typical Steam Generator Model Segment

2

The RETRAN & VIPRE Newsletter – March 2007

RETRAN Modification Removes Enthalpy Transport Model Limitation (Cont'd)
Result: An Improvement The figure to the right shows the result when the new model is applied to the same tube uncovery transient described above. The results show the expected behavior that the junctions become superheated but not excessively and the exit temperature (Junction 474) is lower than the hot leg temperature by a few degrees. Summary A two-region enthalpy transport model has been developed for RETRAN-3D. The model overcomes a known limitation of the enthalpy transport model when used in highly voided volumes. The model: ? ? ? ?
700 650

Temperature (F)

600

550 Junction 471 Junction 472 Junction 473 Junction 474 Volume 420

500

450 0 50 100 150 200 Time (sec) 250 300 350 400

Automatically subnodalizes at the node SGTB Uncovery with Enthalpy Transport (Corrected) dryout point Enthalpy transport is applied to each region Applies appropriate heat transfer correlations to each subregion Uses energy equation to determine temperature for single-phase subregions

Users of RETRAN-3D will have access to the new model when the next code version is released.

Dominion Hosts RETRAN/VIPRE User Group Meeting in Richmond
Last November, RETRAN and VIPRE users traveled east to attend the RETRAN/VIPRE User Group (RUG) Meeting in Glen Allen, Virginia. Attendees were welcomed by Kerry Basehore, Director of Nuclear Analysis and Fuel at Dominion. The 31 attendees represented five US utilities, three international organizations, one U.S. commercial vendor, EPRI, and CSA. Following introductory remarks by Gregg Swindlehurst, the RVUG Steering Committee Chairman, CSA made presentations summarizing the status of the RUG and VUG projects. CSA discussed the financial status of RETRAN project, including a summary of the revenues and expenses for 2006. Summaries of the unresolved and recently resolved RETRAN-02 and RETRAN-3D trouble reports were given. The status of the development work being performed during 2006 included: alternate (optional) turbulent wall friction models, two-region enthalpy transport model for utube steam generator dryout, code architecture modernization, enthalpy transport model studies, and improved time-step selection algorithms. Summarizing the VIPRE project, CSA showed year-to-date expenses and the funding available to perform maintenance and development tasks during the remainder of the year. VIPRE 2006 development included an enhanced CHF calculation summary edit, reduced output file size option, and implementation of the RETRAN-3D water properties. Following the organizational presentations, member organizations made formal and summary presentations of RETRAN and VIPRE activities. ? Duke discussed Feedwater Train Modeling in RETRAN. The presentation described a method for initializing feedwater piping networks that include feedwater heaters. The overall energy balance

The RETRAN & VIPRE Newsletter – March 2007

3

Dominion Hosts RETRAN/VIPRE User Group Meeting in Richmond (Cont'd)
neglects heat addition by heaters. This method can be applied to both u-tube and once-through steam generators. ? Dominion summarized sensitivity studies with the VIPRE rod conduction model. A comparison of the transient MDNBR was given for locked rotor and complete loss of flow transients using dummy rods and rods using the heat conduction model. Duke described a VIPRE-01/MARK B-HTP CHF Correlation Evaluation. Steve Nesbit (Duke), Mike Howard (Westinghouse), and The Framatome BHTP Critical Heat Flux Craig Peterson (CSA) (CHF) correlation was added to VIPRE-01 MOD02 with the ability to switch from the BHTP CHF correlation to the Framatome ? Westinghouse discussed the development of the nonmixing vane BWU-N CHF correlation and with an RETRAN model development for Waterford Unit 3. option of constant or variable axial Turbulent Mixing The digital reactor protection system will be modeled Coefficient (TMC). The VIPRE-01 BHTP results using the CPC-Fortran program, which currently compared favorably with the original LYNX results. utilizes a manual iteration between RETRAN and CPC-Fortran. In the future it will be interfaced to IBERDROLA showed results of a CRDA in a high RETRAN-02 using the PVM code. burnup fuel core for Cofrentes NPP using RETRAN-3D. The current safety limits were ? KEPRI summarized their RETRAN activities. The developed using low burnup fuel data. With an KNAP non-LOCA methodology development effort industry trend toward longer fuel cycles, safety limits was described. Other work summarized included use require additional evaluation. Analyses were of RETRAN to support a simulator updated required performed for high burnup fuel under various inlet for power uprate of the KORI units, and Westinghouse subcooling conditions. three-loop plant equipment qualification for SLB and FLB scenarios. Westinghouse discussed RETRAN modeling of inactive loop flow stagnation during natural circulation ? Duke gave a summary of the applications for which cooldown. With loss of the reactor coolant pumps, VIPRE is used at their facilities. VIPRE is used for natural circulation is the preferred method of removing seven units for both steady-state and transient DNBR stored energy and decay heat from the reactor core. analysis. Twenty-one different CHF correlations have An excessive cooldown rate can lead to flow been implemented to support the Duke units and over stagnation under certain circumstances. 75 cycles have been analyzed. Duke summarized their RETRAN and VIPRE activities which included reanalysis to support changing fuel vendors, examining rod ejection limits and Crystal River Unit 3 steam generator replacement. Duke's Catawba SGTR analysis with MSIV single failure was also discussed. A SGTR overfill analyses with RETRAN-02 determined the limiting single failure to be the secondary PORV on the ruptured steam generator. The analysis determined that steam generator overfill was not predicted, but additional evaluation of steam loads with the failed MSIV is necessary. ? Summarizing their RETRAN work, Westinghouse presented past and current RETRAN-02W applications. Plans include the development of methods for CE plants with digital protection systems. TEPCO gave an update of their activities and plans for RETRAN analysis. The work included use of RETRAN-3D to check vendor analyses, experimental data and other code comparisons, and training. TEPCO also uses RETRAN-3D and SIMULATE3/K as a coupled code system for analyzing events requiring three-dimensional kinetics. Benchmarking comparisons of the Peach Bottom tests were presented.

?

?

?

?

?

?

4

The RETRAN & VIPRE Newsletter – March 2007

Dominion Hosts RETRAN/VIPRE User Group Meeting in Richmond (Cont'd)
model for typical u-tube and oncethrough steam generator models. Another study also examined use of the model for scenarios where u-tube steam generators dry out. CSA also discussed the TwoRegion Enthalpy Transport Model (see lead article "RETRAN Modification Removes Enthalpy Transport Model Limitation" in this newsletter). Limitations with the enthalpy transport model can lead to a violation of the second law of thermodynamics when it is applied to the secondary side of steam generators that dryout. A model is being added to RETRAN-3D that accounts for the changes in heat transfer that occur at the dryout point. ? CSA gave an update on the RETRAN-3D code modernization task and their progress to develop a plan for converting RETRAN-3D to Fortran 95. The purpose of the conversion is to replace FTB memory management with Fortran 95 allocate features and eliminate use of equivalence masks. Additionally, restructuring the code using Fortran 95 control structures will simplify code maintenance. The conversion plan was summarized. ?

Informal Discussion ? Duke discussed a modified Barnett CHF correlation application for steam line breaks. Duke is experiencing fuel failures with Mk-B10 and Mk-B11 fuel associated with the mixing vane design. They are evaluating HTP fuel which requires use of the BHTP CHF correlation. Since the lower pressure limit for the correlation is 1400 psia, vendor recommended use of the modified Barnett correlation for lower pressures. CSA reported on enthalpy transport use studies. The enthalpy transport model is commonly used in heated regions such as cores and steam generators. Model limitations often lead to solution errors in steam generator models when flow reversals and countercurrent flow patterns occur. Several studies were performed to establish guidance for use of the

?

The spring 2007 RVUG meeting will be held in San Diego, California, on April 23 & 24.

Sang Jun Ha's KEPRI Presentation

The RETRAN & VIPRE Newsletter – March 2007

5

VIPRE-01 Modifications
Several modifications to the current VIPRE-01 version (VIPRE-01 MOD02.2.1) are available as a result of 2006 VIPRE-01 work scope development items or corrections for trouble reports. Modification descriptions are provided below. VUG members that are interested in obtaining the modifications can contact Garry Gose at gcg@csai.com. Modification 249 Error Correction for Trouble Report 241 An executable, installed using a newer compiler, resulted in run-time errors when executing sample problems. The errors are due to an array index that is not within the declared array dimensions. In these cases, the array index is 0. In addition, the compiler issues an error due to a missing comma in routine recirc.f. These problems have not been encountered on any previous Unix or PC compiler versions. Modification 249 corrects “bounds checking” errors. A missing comma was also fixed in subroutine RECIRC. Modification 250 Error Correction for Trouble Report 242 VUG requested work on the CHFDMP (Unit 17) CHF summary file (see Modification 252) and it was noted that using the SI output option yields different results which are not due simply to unit conversions (in the summary file). Investigation showed the parallel logic used to produce results in SI units (Subroutine RESLT2) is missing a variable definition for smdnbr which is used to determine the MDNBR location. Modification 250 corrects logic used to produce results in SI units (Subroutine RESLT2). Modification 251 VIPRE-01 Software Revision Request 009 This modification is a VUG requested effort to add RETRAN-3D water properties functions (subroutines taken from RETRAN-3D MOD004.2) as a user-selected option to use these newer functions in place of the current implementation of the EPRI water properties. The properties from the new routines are obtained with new VIPRE-01 Option NFPROP = 3 (VIPRE-01 PROP.1, Word 4). Testing from the VIPRE-01 sample problem test suite indicated no significant differences between NFPROP = 2 or 3. No differences should be expected unless pressures greater than Pcrit or lower than 15 psia are encountered. Modification 252 VIPRE-01 Software Revision Request 010 VIPRE-01 generates a CHF summary file ("CHFDMP") on Unit 17. This VUG requested modification cleans up the format of the CHFDMP file. The information is more organized and the operating conditions are presented in units which match the input specifications. Also, a convergence indicator has been added to inform the user if the CHF data were produced by a nonconverged solution. Modification 253 VIPRE-01 Software Revision Request 011 Modification 253 is a VUG requested modification to optionally suppress the input reflection, edit summary, and bundle summary edits from the VIPRE-01 output file. This option is obtained by setting the new Word 5 (ISUPP on VIPRE.1) card to 1. The default of 0 produces the normal, unsuppressed output.

6

The RETRAN & VIPRE Newsletter – March 2007

Transforming Pump Data from Characteristic Curves to Homologous Curves
This tech tip illustrates how to get RETRAN centrifugal pump head and torque data from characteristic curves. The procedure focuses on the first quadrant but it can be applied to all of the regions in the four quadrants. Figure 1 is a reproduction of a typical four quadrant pump characteristic curve. The first quadrant of the pump characteristics curve is called the 'normal' pump quadrant because the pump head and torque curves are supplied as functions of flow and speed. In this quadrant, both head and torque can range from positive to negative values. The homologous curves are produced from the characteristics curves by a transformation of variables in which the resultant curves are dimensionless parameters. The parameters are normalized head and torque as functions of dimensionless parameters a, v, h, and b where these parameters are defined in the RETRAN-3D Theory Manual (Page VI-6) as a = Speed/Speed (Rated) v = Flow/Flow (Rated) h = Head/Head (Rated) b = Torque/Torque (Rated) The desired homologous curves (Figure 2 for example) are produced by plotting the head and torque ratios as functions of pump speed and volume ratios.
2 h/a = FH (v/a) 2 h/v = GH (a/v) b/a2 = FT (v/a) b/v2 = GT (a/v)

Figure 1. Four-Quadrant Pump Characteristic Curves

1

As an illustration, we will define the homologous pump head data for the first quadrant. The torque curve will follow identically. First, if one examines the homologous pump curve numbering system given by Table 1, the normal pump quadrant (+v, +a) requires two head curves for a complete description. These curves are numbered 1 and 2 in the RETRAN input. Head curves (3 through 8) are used for representing the other pump quadrants. Curves 9 through 16 are for the specification of torque. The first curve specifies the normalized head as a function of v/a for values of v/a less than or equal to 1. Therefore, for Curve 1, the dependent variable is 2 normalized head (h/a ) as a function of the independent variable (v/a). This is FH defined above.

2

Figure 2. Homologous Pump Head Curves

The RETRAN & VIPRE Newsletter – March 2007

7

Transforming Pump Data from Characteristic Curves to Homologous Curves (Cont'd)
Table 1. Homologous Curve Numbering System Independent Variable v/a a/v v/a a/v v/a a/v v/a a/v v/a a/v v/a a/v v/a a/v v/a a/v Curve Number 1 2 9 10 3 4 11 12 5 6 13 14 7 8 15 16 Curve Quadrant Normal Pump (+v, +a) Type Head Torque Energy Dissipation (-v, +a) Head Torque Normal Turbine (-v, -a) Head Torque Reverse Pump (+v, -a) Head Torque v/a <1 >1 <1 >1 >-1 <-1 >-1 <-1 <1 >1 <1 >1 >-1 <-1 >-1 <-1

There are situations where v/a is greater than 1, but this will not be specified using Curve 1. As Table 1 shows, for data where v/a is greater than 1, Curve 2 is used. Curve 2 requires normalized head dependent (h/v2) as a function of the independent variable (a/v). This is GH defined above. There is a corresponding set of curves required for torque, using RETRAN Curves 9 and 10. Curves 11 through 16 would be used for the other quadrants for torque data. As a specific example, if one examines the homologous head curve given in Figure 2, the first quadrant normalized head data is represented by two curve segments. For those values of v/a less than or equal to 1, 2 the curve segment in the first quadrant marked (h/a , v/a) is used. Thus, values of v/a from 1.0 to 0.0 correspond to values of h/a2 from 1.0 to about 1.3 or 1.4 terminating on the y axis. This is Curve 1 in the RETRAN input file. The second curve of the first quadrant normalized head curve is indicated by (h/v2, a/v). This curve is used when values of v/a are greater than 1. Values of a/v from 1.0 to 0.0, correspond to values of h/v2 from 1.0 to about -.4 or -.5 terminating on the y axis. The curve crosses the zero head line at a value of a/v of about .5. This is Curve 2 in the RETRAN input file. Typical values of the normalized head input are shown below (not to be used as real model input).

* CURVE 1 *CARD Num v/a Normalized Head (h/a**2) 101011 5 + 0.00 1.7 + 0.25 1.5 + 0.50 1.3 + 0.75 1.2 + 1.00 1.0 * *CURVE 2 *CARD Num a/v Normalized Head (h/v**2) 101021 5 + 0.00 -2.0 + 0.25 -1.1 + 0.50 -0.4 + 0.75 0.30 + 1.00 1.0 There is a corresponding set of curves (9 and 10) for the normalized torque curve segments, but the process is exactly the same. In practice, it is a matter of selecting values for the independent variables, a/v or v/a from 0.0 to 1.0 and making a determination of the corresponding normalized head and torque values from the characteristic curves and the rated speed, head, flow, and torque for the given pump. One may describe the curves with as many points as necessary to accurately describe the pump characteristics, but the curves are generally smooth polynomials.

8

The RETRAN & VIPRE Newsletter – March 2007

About This Newsletter
RETRAN Maintenance Program
The RETRAN/VIPRE Maintenance Program is a program that provides for the support of software developed and maintained by CSA. The main features of the Subscription Service include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN and VIPRE.

Newsletter Contributions
The RETRAN/VIPRE Newsletter is published for members of the Subscription Service program. We want to use the newsletter as a means of communication, not only from CSA to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. The next newsletter is scheduled for December 2007 and we would like to include a brief summary of your RETRAN and VIPRE activities. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to one of the email addresses below by December 4, 2007. We are looking forward to hearing from all RETRAN and VIPRE licensees. Mark Paulsen Garry Gose Pam Richardson paulsen@csai.com gcg@csai.com pam@csai.com

The RETRAN Maintenance Program now has 19 organizations participating in the program, including 13 U.S. utilities and six organizations from outside of the U.S. Eight U.S. utilities and three organizations outside the U.S. are currently participating in the VIPRE maintenance program. A Steering Committee, composed of representatives from the participating organizations, advises CSA on various activities including possible enhancements for the code and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from: Mark P. Paulsen Computer Simulation & Analysis, Inc. P. O. Box 51596 Idaho Falls, ID 83405 paulsen@csai.com or (208) 529-1700

The RETRAN web page is located at http://www.csai.com/retran/summary.html. The VIPRE web page is located at http://www.csai.com/vipre/summary.html Previous issues of the RETRAN/VIPRE Newsletter are available from the RETRAN or VIPRE web pages.

Steering Committee Members
Gregg Swindlehurst, Duke Energy (Chairman), gbswindl@duke-energy.com Andres Gomez Navarro, Iberdrola, agn@iberinco.com Sama Bilbao y Leon, Dominion, Sama_Bilbao@dom.com David Huegel, Westinghouse, huegelds@westinghouse.com Adi Irani, Entergy Nuclear Northeast, airani@entergy.com Daren Chang, STPNOC, dchang@stpegs.com

Calendar of Events
User Group Meeting April 23 & 24, 2007 San Diego, California http://www.csai.com/retran/rvug/ugm.html RETRAN Training Session June 18-22, 2007 Idaho Falls, Idaho http://www.csai.com/retran/summary.html ANS Annual Meeting June 24-28, 2007 Boston, Massachusetts

The RETRAN & VIPRE Newsletter – March 2007

9

The RETRAN Newsletter
Summary of Activities This issue of the RETRAN Newsletter contains information on the new RETRAN Mailing List, addition of CORETRAN to the CSA Web Page, and interesting articles from code users. Your contributions are greatly appreciated. We, EPRI and CSA, encourage everyone to participate in this newsletter. Previous issues of the RETRAN Newsletter are available from the RETRAN Web Pages at http://www.csai.com/retran.
In This Issue . . . RETRAN-3D Experience at Duke Power Company . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 RETRAN-02 Activities in INVAP . . . . . . . . 3 CORETRAN Added to the CSA Web Page 3 Domain Decomposition Methods for 3-D Flow Simulation in RETRAN-3D . . . . . . . 4 Responses to EPRI Survey on RETRAN-02 and RETRAN-03 Software Experience Still Needed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 RETRAN-02 Trouble Reports . . . . . . . . . . 6 RETRAN-3D Trouble Reports . . . . . . . . . . 7 Announcing the RETRAN Mailing List . . . 7 Calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Newsletter Contributions . . . . . . . . . . . . . . . 8

Published by Computer Simulation & Analysis, Inc. for the Electric Power Research Institute

March, 1997

RETRAN-3D Experience at Duke Power Company
G. Seeburger, Duke Power For many years now, Duke Power has done its own safety analysis with RETRAN-02 for four Westinghouse nuclear units and is currently preparing to submit a topical report to allow the same work on three B&W units. In an effort to gain experience with RETRAN-3D, and with an eye towards a possible upgrade to the licensed methods, the locked rotor event was recently analyzed with RETRAN-3D with the 3D kinetics option. The analysis was a best-estimate exercise of the technology, not a limiting FSAR-like worst case. The RETRAN-3D system model was based on a four-loop Westinghouse plant. A two-loop model was used: one loop/steam generator was modeled individually for the pump with the locked rotor while the other three loops/steam generators were lumped into the other loop. The 3D kinetics model was taken from an ARROTTA core model that had been used previously for a rod ejection analysis. No adjustments were made to the cross sections of the core model, so the power distribution, reactivity coefficients, and shutdown margin are not at their limiting values. RETRAN-3D was first initialized with a three-level point kinetics core model. Then an 18-axial level, 20 flow channel 3D core model was added. Eighteen-axial levels were chosen because the number must match the ARROTTA core model. 1

RETRAN-3D Experience at Duke Power Company (Cont'd)
One flow channel represents the core bypass and the remaining 19 flow channels represent the 193 fuel assemblies. Each of the 19 channels represents from one to six lumped assemblies. The choice of which assemblies to lump together was based on the assembly average power and the proximity to or presence of a control rod location. This choice of fuel assembly to lumping is specific to the power distribution and the transient to be simulated. Note that assemblies that are lumped together are not necessarily adjacent to each other. Since the locked rotor transient is symmetric as far as the core is concerned over the time period of interest (a few seconds), having lumped assemblies not adjacent to each other is acceptable. The 3D core was made hydraulically equivalent to the three-level point kinetics model that it replaced. Even so, some small adjustments were required to the core loss coefficients to get a converged flow solution.
8 9 13 11 12 11 11 9 7 5 4 6 9 8 6 17 3 8 7 7 6 17 3 2 10 5 17 3 8 1 4 3 2 4 6 8 18 14 16

10 12 11 7

19 16

12 10 13 14 1 4

19 15

19 14 16

19 15 16

15 18 14 16 16 RETRAN-3D Channel Layout

The transient was initiated by instantly setting the pump speed to zero on the single loop. The reactor tripped on the RPS low flow signal. The three remaining pumps on the lumped triple loop were tripped on the ensuing turbine trip to simulate the loss of offsite power. Once the transient was run, the core inlet flow versus time was taken from RETRAN-3D and input to an equivalent ARROTTA case. The minor core inlet temperature variation and the pressure variation were ignored. The ARROTTA assembly average power distribution closely matched the RETRAN-3D power distribution at the start of the transient, with the maximum error less than 1% of assembly power. By the end of the transient the agreement was not as good, with the maximum error about 7% of assembly power. However, the more important parameter of core power versus time matches very well between the two models, with the largest error being 1.5% of full power throughout the transient. The excellent agreement of core power versus time shows that the 19 core flow channels adequately model the thermal-hydraulic feedback, even though the power distribution comparison does degrade over the duration of the transient. The transient results are of little interest in absolute terms because it was done as best estimate and there is nothing to compare to. The next step in the process is to re-run the transient as a limiting, FSAR-type analysis. This will involve initializing RETRAN-3D with low flow, high vessel average temperature, high pressure (looking at peak pressure as opposed to DNBR). There will also be changes to the core model cross sections to obtain a limiting power distribution, limiting reactivity coefficients, a limiting delayed neutron fraction, and a limiting shutdown margin. These changes will facilitate a comparison to the current licensing basis analysis to determine the relative merit of the 3D kinetics option (for the locked rotor transient, at least). Along the way, some interesting experience should also be gained.

2

RETRAN-02 Activities in INVAP
C. Mazufri, INVAP INVAP, under contract of CNEA (National Atomic Energy of Argentina), is developing an advance design of a nuclear reactor for electrical generation named CAREM. The main features of the reactor are: small size (25 MWe), core cooling by natural circulation, steam generator integrated into the pressure vessel, and selfpressurized. According to its acceptable capability for licensing applications, the RETRAN-02 code was chosen to analyze the reactor behavior under transient and abnormal situations. Typical operation transients like power ramp, energy unbalances, load following; and abnormal events such as blackout, LOCA, reactivity insertions; have been analyzed and included in the Safety Analysis Report. As part of the CAREM project, assessments of RETRAN-02 dynamic response predictions against experimental data in natural circulation conditions are now being carried out. Experimental results are obtained from a natural circulation loop developed ad hoc for CAREM thermal-hydraulic conditions by INVAP.

CORETRAN Added to CSA Web Page
J. Westacott, CSA CSA is the contractor to EPRI for CORETRAN maintenance support. CORETRAN (developed by S. Levy and Battelle for EPRI) is designed to perform LWR core steady-state and transient analyses. CORETRAN is a product of combining the ARROTTA three-dimensional neutron kinetics and VIPRE-02 thermal-hydraulic codes into a unified package. The CSA Web page now includes information of interest to CORETRAN users and is available on http://www.csai.com/coretran/index.html. The web page includes a summary of the CORETRAN-01 code, a list of trouble reports, and a form to submit CORETRAN trouble reports electronically.

3

Domain Decomposition Methods for 3-D Flow Simulation in RETRAN-3D
T. Downar and J.-Y. Wu, Purdue University The simulation of Light Water Reactor transients has typically been limited to one-dimensional fluid dynamics modeling. For licensing based accidents in which spatial variations of the power, flow, or temperature in the core are important, conservative operating margins are imposed resulting in some degradation of plant efficiency. High fidelity, three-dimensional flow modeling can help reduce uncertainties in code predictions and can improve the basic understanding of transient phenomena. One of the primary obstacles to 3-D flow simulation in the RETRAN-3D code has been the large computational burden incurred in solving the linearized fluid dynamics equations. Recent advances in semi-iterative solution methods and parallel computing has motivated an investigation into new methods for solving the RETRAN-3D fluid dynamics equations. The objective of this exploratory research project was to establish the computational feasibility of 3-D flow modeling in RETRAN-3D. Because cross flow between fuel assemblies has been shown to be important for the PWR steam line break transient [Dias, 1992], the SLB analysis was chosen as the application for demonstrating the methods developed here. The first phase of the research focused on the parallelization of the RETRAN-3D transient algorithm. Spatial domain decomposition methods were applied in which groups of contiguous volumes and junctions were assigned to separate processors for a sequence of computations. The parallelized program was tested for a typical 1-D flow problem on the distributed memory INTEL Paragon. Only moderate speedups were achieved primarily because of the heavy overhead in message passing and the difficulty of balancing the computational load among the processors. It became apparent that the tightly coupled thermal-hydraulic algorithms in RETRAN-3D could more efficiently be parallelized on a shared memory multiprocessor in which message passing overhead is eliminated. The second phase of the work focused on the RETRAN-3D linear solver which consumed over 99% of the CPU time for reactor models with 3-D flow channels such as depicted in Figure 1. The existing direct solver in RETRAN was optimized for 1-D flow and employs a Sparse Block Elimination scheme which takes advantage Figure 1. A PWR SLB Analysis Model 4

Domain Decomposition Methods for 3-D Flow Simulation in RETRAN-3D (Cont'd)
of the tridiagonal matrix structures encountered in 1-D flow models. For 3-D flow this method was very inefficient and the computational time increased as the cube of the matrix size. An innovative iterative linear solver was developed based on the Generalized Minimum Residual (GMRES) method and accelerated with a domain decomposition preconditioner in which the core and the ex-core blocks of the linear system were solved separately. The 3-D core flow problem was solved by a second level GMRES and accelerated using a Schwarz preconditioner consisting of overlapping subdomains with multiple channels. The performance of the new iterative solution algorithm was studied using a PWR SLB model with explicit cross flow modeling between fuel channels as shown in Figure 1. The core in the model shown consists of 25 channels and 12 axial planes. A second model was also examined in which each assembly in a half-core was modeled separately as a total of 104 channels. The solution with the direct solver was reproduced by the iterative solver for both the 25- and 104-channel models. The computational time for the 25-channel model was reduced by a factor of 44, whereas comparisons to the direct solver time for the 104-channel model were not meaningful because the memory requirements of the direct solver exceeded the workstation core memory capacity. The direct solver required an excess of 1 GB whereas the iterative solver required less than 100 MB. Since the CPU time of the direct solver increased as the matrix order cubed, a reasonable estimate of the reduction in the execution time for 104-channel problem would be more than a factor of two orders of magnitude. Parallel performance of the iterative solver was examined only on the distributed memory INTEL Paragon. The speedup compared to the serial performance was again modest because of the fine graining of the algorithm and the heavy message passing overhead. The best parallel efficiency achieved with the 104-channel model was about 50% on three processors. A considerably better parallel efficiency for the iterative solution method can be expected on shared memory workstations which are more suitable for fine grain parallelism. Future work will examine the adaptation of the methods developed here onto a shared memory multiprocessor with the expectation of further performance improvement. However, the results of the work appear to establish the computational feasibility of three-dimensional flow modeling and provide encouragement for practical three-dimensional coupled thermal-hydraulics and neutronics in RETRAN-3D. This research is sponsored by the Strategic R&D Group at EPRI.

Responses to EPRI Survey on RETRAN-02 and RETRAN-3D Software Experience Still Needed
G. Swindlehurst, Duke Power By letter dated January 8, 1997, from Lance Agee to the members of the RETRAN User Group, your response to a survey regarding your use of RETRAN-02 and RETRAN-3D was solicited. EPRI and the RETRAN Steering Committee are still very much interested in receiving your completed survey forms. The results will be used to improve how EPRI software and, in particular, the RETRAN codes are developed, distributed, and maintained. Please send your responses in at your earliest convenience. 5

RETRAN-02 Trouble Reports
The following is a summary of RETRAN-02 Trouble Report/Code Maintenance Activity. Unresolved Trouble Reports ? 1 From MOD001 ? 5 From MOD002 ? 4 From MOD003 ? 3 From MOD004 ? 8 From MOD005 A list of trouble reports and the status can be obtained directly from the EPSC. Additional information is available from the RETRAN-02 Trouble Report Page at http://www.csai.com/retran/r02trpt/index.html.

Summary of RETRAN-02 Code Trouble Reports
NO. 1 61 121 139 140 177 209 272 317 334 342 354 366 376 394 408 413 436 437 438 439 440 *441 TROUBLE REPORT TYPE OF PROBLEM Error 209 in TEMZ Delta T for Conductor with TDV OTSG Low Power Initialization Failed Using Large Time Step Spurious Trips on High Level Overflow in WAT9 Pump Coast Down Rates Junction Properties at Break Junction Property Error Time-Dep. Volume Input Control Block Output near Zero Large Step Change in PHIR Mixture/Liquid Level Difference Control Reactivity, No Motion Anomalous Heat Trans. Behavior OTSG Heat Transfer Problems Incorrect Vsn No. in IBM Output Prandtl Number is Discontinuous Heat Transfer Logic/CHF Restart Failure/Pipe Transport Decay Heat Input Kinetic Energy/Time Dep Area Anomalous Power Increase CORRECTION NO. IDENT *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** 405 ----407 *** *** ----******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* MOD005P3 ----------MOD005P3 ******* ******* ----------COMMENTS MOD001 Error Need Input Deck Need Input Deck Need Input Deck Need Input Deck Need Correct Deck Need Input Deck Cannot Reproduce Error Need Input Deck

Cannot Reproduce Error Not a Code Error

Not a Code Error

6

Summary of RETRAN-3D Code Trouble Reports
RETRAN-3D MOD002.0 has been completed and transmitted to the Electric Power Software Center. A total of 122 trouble reports had been filed as of February 28, 1997. Of these, 103 reports have been resolved, while 19 remain unresolved. A summary of the unresolved trouble reports is shown below. Additional information for RETRAN-3D trouble reports is available at http://www.csai.com/retran/r3dtrpt/index.html.
NO. 7 30 33 39 40 41 43 45 47 48 51 52 54 60 70 116 119 121 122 TROUBLE REPORT TYPE OF PROBLEM Steam separator model fails 2-loop Oconee w/5-eq. fails in steady state 000040 data not read during restart Time-step error; pressure is 5997 psia Results do not agree with data Anomalous downcomer level Steady-state convergence error Restart incorrect transient values Standard Problem One difference Steady state fails after 6 iterations Pressure search failure for two-phase MOC MOC does not return to the initial temp. MOC solution; no null transient for two-phase Anomalous countercurrent flooding Fails in subroutine DERIVS Fails in steady-state initialization SS fails to converge for some cases (algebraic) Calculation failure on second time step Problems with EOS convergence CORRECTION NO. IDENT *** *** *** *** *** *** *** *** *** *** 006 *** *** *** *** *** 052 *** *** *** *** ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* MOD001g ******* ******* ******* ******* ******* MOD001g ******* ******* ******* ******* COMMENTS

(partial fix)

(partial fix)

Announcing the RETRAN Mailing List
G. Gose and K. Kvarfordt, CSA An internet mailing list has been established for the purpose of exchanging ideas on RETRAN related issues. The mailing list is intended to provide a forum for an on-going group discussion of RETRAN methods, models, ideas, or other related topics. A mailing list is similar to a news group because when a member of the group posts a message to the list address (defined below), it is distributed to all of the members or subscribers to the list. The list is different from a news group because access to the list is controlled and the discussion will be moderated. While the moderator will reject nonappropriate or offensive postings, a free idea exchange is encouraged. Subject matter related to thermal hydraulics or core physics methods are sought, but any topic that may enable engineers to better model the plant are valid. Please keep the discussion to technical issues.

7

Announcing the RETRAN Mailing List (Cont'd)
To subscribe, send email to majordomo@domo.srv.net leave the subject line blank and enter subscribe retran in the body of the message. A digest version of the list is also available by entering subscribe retran-digest in the body of the message. For more information, contact Garry Gose at gag@csai.com or Kent Kvarfordt at kek@csai.com.

Calendar
Basic RETRAN Training Idaho Falls, ID June 16-20, 1997 Advanced RETRAN Training Idaho Falls, ID June 24-27, 1997 RETRAN User Group Meeting September - October, 1997 Location To Be Announced International RETRAN Meeting 1998 Location, Date To Be Announced

About This Newsletter
RETRAN Maintenance Program
The RETRAN Maintenance Program is part of a program undertaken by EPRI to provide for the support of the software developed in the Nuclear Power Division. The main features of the Subscription Service include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN.

Newsletter Contributions
The RETRAN Newsletter is published for members of the Subscription Service program. We want to use the newsletter as a means of communication, not only from EPRI to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. The next newsletter is scheduled for June 1997 and we would like to include a brief summary of your RETRAN activities. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to the E-mail addresses below by June 1, 1997. Contributors will receive a RETRAN mouse pad. We are looking forward to hearing from all RETRAN licensees. James McFadden Mark Paulsen Garry Gose Pam Richardson jm-csa@csai.com (208) 529-1700 mp-csa@csai.com gag@csai.com pam@csai.com

The RETRAN Maintenance Program now has 31 organizations participating in the program, including 23 member utilities, 5 organizations from outside of the U.S., and 3 nonmember utilities from the U.S. A Steering Committee, composed of representatives from the participating organizations, advises EPRI on various activities including possible enhancements for the code and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from Lance Agee Electric Power Research Institute P. O. Box 10412 Palo Alto, CA 94303 lagee@epri.com or (415) 855-2106

The RETRAN Web Page is located at http://www.csai.com/retran/index.html.

8

The RETRAN Newsletter
Summary of Activities This issue of the RETRAN Newsletter contains information on the 1998 Steering Committee, Ninth International RETRAN Meeting, and interesting articles from code users . Your contributions are greatly appreciated. We, EPRI and CSA, encourage everyone to participate in this newsletter. Previous issues of the RETRAN Newsletter are available from the RETRAN Web Pages at http://www.csai.com/retran.
In This Issue... Use of EPRI Codes at the Paul Scherrer Institute . . . . . . . . . . . . . . . . . . Ninth International RETRAN Meeting . . 1 3

Published by Computer Simulation & Analysis, Inc. for the Electric Power Research Institute

April, 1998

Use of EPRI Codes at the Paul Scherrer Institute in Switzerland
M. Zimmermann, PSI At the Paul Scherrer Institute, the active involvement with the EPRI codes began with RETRAN-02, and the use of this code culminated in its application within a design study of a 20 MWth district heating reactor. Based on this positive experience, the participation in the RETRAN-03 development project was a logical extension of PSI's commitment to RETRAN. Meanwhile, the analytical activities relating to the safety analysis of the Swiss NPPs have been organized within the STARS project. The goal of this project is to provide an independent capability for the deterministic safety analysis of the Swiss NPP (two BWRs and three PWRs of four different designs) to the Swiss Nuclear authority (HSK). The scope includes the analysis of plant behavior for conditions both within and beyond design bases transients, but excludes the analysis of severe accident sequences. The STARS project includes the following technical fields:
4

TechTips . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 A New Interphase Mass Transfer Model for RETRAN-3D . . . . . . . . . . . . .

1998 Steering Committee . . . . . . . . . . . . . . 5 Summary of RETRAN-3D Code Trouble Reports . . . . . . . . . . . . . . . . . . . . 5 Summary of RETRAN-02 Trouble Reports . . . . . . . . . . . . . . . . . . . . 6 RETRAN-02 MOD005.2 Ready for Licensing Applications . . . . . . . . . . . . . . Calendar . . . . . . . . . . . . . . . . . . . . . . . . . . About this Newsletter . . . . . . . . . . . . . . . . 7 8 8

C C C C C

steady-state and transient core analysis, plant transient analysis, loss of coolant accidents (SB and LB), fuel behavior, and containment analysis (limited activity).

Plant-specific models are being developed for all the codes used for this broad field of analysis and are validated against available test data from the NPPs. It is evident from the above description, that a collection of well validated codes is needed. Figure 1 shows the current system of codes used in the STARS project. 1

Use of EPRI Codes at the Paul Scherrer Institute in Switzerland (Cont'd)
Pin Power BU

SIMULATE

RELAP5

CORETRAN
XSec Tables
SLICK

RETRAN-3D

DNBR CPR

TABLES-3

CORETRAN

Boundary Conditions

TRANSURANUS

Fuel-Rod Parameters

XSec Pin Power

POLGEN

RAMONA-3.10

GOTHIC

P, T Containment

CASMO-4

0D/1D Nucl. Data

TRAC-BF1

Fine Group XSec

System Parameters

Figure 1: Codes Used in the STARS Project Nuclear cross sections are generated with the help of CASMO-4 and SIMULATE, both from Studsvik, Inc. CORETRAN is used for the analysis of reactivity initiated accidents while RETRAN-3D is used for transient analysis of both BWRs and PWRs. LOCA analysis is conducted with RELAP5 for PWR and TRAC-BF1 for BWR. Subchannel analysis is performed with the help of CORETRAN (VIPRE-02) while the TRANSURANUS code (from the TRANSURANUS Institute) is used for fuel analysis. BWR stability is currently analyzed with the help of RAMONA-3.10 from Scandpower Inc. It is evident from the above list that EPRI codes play a central role in the STARS project. There are several reasons for this: C EPRI took the lead in the long-term development of a fully integrated code package (which eventually will cover the whole spectrum needed in the STARS project); EPRI's codes are developed within a framework of strong relations to the utility industry and therefore address "real life" problems; EPRI's codes apply both for BWR and PWR; and EPRI's codes attract a sizable user community that exchanges its experiences and shares the cost of maintenance, a key element for the survival of any computer code.

C

C C

The operational implementation of 3D kinetics into RETRAN-3D is a high priority item for 1998. Here at PSI this will follow the route: two-group nuclear cross sections generated with CASMO-4 will be transferred to CORETRAN and integrated into the plant core model, from there the transient cross sections together with the core model will be
(Continued on Page 7)

2

Ninth International RETRAN Meeting
J. McFadden, CSA The Ninth International RETRAN Meeting will be held in Monterey, California on June 7-10, 1998. The meeting is sponsored by the Electric Power Research Institute and Computer Simulation & Analysis, Inc. The objectives of the meeting are: 1) to allow an exchange of RETRAN-related information between EPRI and the code licensees and 2) to document the analyses effort of the RETRAN code users. The Plenary Session on Monday, June 8, will include an introductory address summarizing various activities related to RETRAN and a panel discussion on "Analysis Needs for Supporting Nuclear Plants in the 21st Century". The panel is comprised of representatives from EPRI, the utility industry, and other organizations involved in nuclear power activities. The Technical Program will include presentations of RETRAN development efforts and applications of RETRAN for boiling water reactors and pressurized water reactors. The technical

sessions on Wednesday, June 10, will be devoted to presentations in EPRI's new software analysis tools, CORETRAN, and RETRAN-3D. Proceedings containing the complete text of the keynote address and all invited and contributed papers will be printed. One copy of the proceedings will be sent to each meeting participant. Additional copies may be ordered at the meeting or from EPRI. For additional information or registration forms, please contact Pam Richardson at (208) 529-1700 or pam@csai.com.

TechTips

Sometimes bubble rise volumes can experience numerical difficulties when the mixture level passes through the elevation of the outflow junction. The numerical problems show up as volume pressure and energy spikes. You can generally eliminate the problem by changing the junction from a horizontal "point" to a vertical junction. Just orient the junction vertically (JVERT =1 on the 08XXXY record) and then overlap the connected volumes as shown in the figure. Make sure to move the junction elevation to the midpoint of the overlapping region. You don't need very much of an overlap. Usually a few tenths of a foot will do. The result of this little change will smooth the junction enthalpy as Volume Volume Pressure Pressure the mixture level passes through the junction elevation, eliminating those problem pressure BEFORE spikes! AFTER

3

A New Interphase Mass Transfer Model for RETRAN-3D
P. Cebull, CSA, and R. Macian, PSI The five-equation model in RETRAN-3D was implemented in order to allow the simulation of nonequilibrium thermal-hydraulic phenomena. It currently includes a pressure change, or dP/dt-based model which is used to calculate the rate of vapor generation during pressure transients, such as turbine trips, small-break LOCAs, etc. However, experience gained from analyses of these types has shown that the model can produce unrealistic mass transfer rates and liquid superheat values which are too large, often resulting in numerical instabilities and subsequent code failure. Coefficients used to mitigate the problem are problem dependent and have not been correlated against a reasonable independent parameter. CSA is currently involved in a collaborative effort with the Paul Scherrer Institute to implement a new interphase mass transfer model. It is based on a concept commonly used in the six-equation simulation codes, in which the interfacial mass transfer is treated as a heat transfer process. Each phase transfers heat to an interface which exists at the local saturation temperature. The heat transfer coefficient is calculated based on the flow regime and local thermodynamic conditions. Implementation of the new model includes replacement of the flow regime map currently used in RETRAN-3D. The map being used now was empirically derived for heated, small-diameter tubes under high pressure, and thus its use under conditions much different from these (such as low pressure transients) is somewhat questionable. The new flow map is based on the work of Taitel and Dukler. It has been shown to be applicable to a wide range of pressures and flow configurations and was developed from an analysis of the physical phenomena affecting flow regime transitions. The new model is entering the testing phase of development and should be included in the next formally released version of RETRAN-3D. Its implementation should allow the range of application of RETRAN-3D to be expanded, encompassing new types of transients which were previously beyond its capabilities. 4

1998 Steering Committee
L. Agee, EPRI The results of the 1998 election for the RETRAN Steering Committee are as follows. Chairman: Gregg Swindlehurst Duke Energy Corp. Phone: (704) 382-5176 Fax: (704) 382-7852 E-mail: gbswindl@duke-energy.com

The other three steering committee members are: James Boatwright TU Electric Phone: (214) 812-8232 Fax: (214) 812-8687 E-mail: tboatwright@tuelectric.com Yuki Fujita Duke Engineering & Services Phone: (978) 568-2130 Fax: (978) 568-3700 E-mail: fujita@yankee.com Shie-Jeng Peng Illinois Power Co. Phone: (217) 935-8881 x4023 Fax: (217) 935-8294 E-mail: shie-jeng_peng@illinova.com

Please keep one or more of these people aware of your evolving requirements and what you think is important with respect to RETRAN-02 and RETRAN-3D. The steering committee's role is to assure that the utilities' RETRAN needs are fulfilled in as timely and cost effective a manner as possible. The committee advises EPRI as to priorities and assists in determining the future years' work scope for the code maintenance project.

Summary of RETRAN-3D Code Trouble Reports
A total of 147 trouble reports had been filed as of February 28, 1998. Of these, 132 reports have been resolved, while 15 remain unresolved. A summary of the unresolved trouble reports is shown below. Additional information for RETRAN-3D trouble reports is available at http://www.csai.com/retran/r3dtrpt/index.html.

NO. TROUBLE REPORT TYPE OF PROBLEM 7 Steam separator model fails 22 Problem using Wilson bubble rise model & error when using low power initialization 30 2-loop Oconee w/5-eq. fails in steady state 40 Results do not agree with data 48 Steady state fails after 6 iterations 52 54 60 81 116 122 127 142 144 145 MOC does not return to the initial temp. MOC solution; no null transient for two-phase Anomalous countercurrent flooding Steady-state failure at iteration #6 Fails in steady-state initialization Problems with EOS convergence Lack of convergence error Timestep selection causes 3-D kin to fail TAUGL model doesn't apply for horiz. flow SS fails to converge for low press. and flow

CORRECTION NO. IDENT *** *** *** *** *** 006 *** *** *** *** *** *** *** *** *** *** ******* ******* MOD001 ******* ******* ******* MOD001g ******* ******* ******* ******* ******* ******* ******* ******* ******* *******

COMMENTS

(partial fix)

(partial fix)

(water packing) (mass transfer)

5

Summary of RETRAN-02 Trouble Reports
The following is a summary of RETRAN-02 Trouble Report/Code Maintenance Activity. Unresolved Trouble Reports ? ? ? ? ? 1 From MOD001 5 From MOD002 4 From MOD003 3 From MOD004 9 From MOD005

A list of trouble reports and the status can be obtained directly from the EPSC. Additional information is available from the RETRAN-02 Trouble Report Page at http://www.csai.com/retran/r02trpt/index.html.

TROUBLE REPORT NO. TYPE OF PROBLEM 1 61 121 139 140 177 209 272 317 334 342 354 366 376 394 408 413 439 440 442 443 444 Error 209 in TEMZ Delta T for Conductor with TDV OTSG Low Power Initialization Failed Using Large Time Step Spurious Trips on High Level Overflow in WAT9 Pump Coast Down Rates Junction Properties at Break Junction Property Error Time-Dep. Volume Input Control Block Output near Zero Large Step Change in PHIR Mixture/Liquid Level Difference Control Reactivity, No Motion Anomalous Heat Trans. Behavior OTSG Heat Transfer Problems Incorrect Vsn No. in IBM Output Decay Heat Input Kinetic Energy/Time Dep Area Poor Diagnostics Liquid Region Work Term Positive Slip Velocity

CORRECTION NO. IDENT *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* *******

COMMENTS MOD001 Error Need Input Deck Need Input Deck Need Input Deck Need Input Deck Need Correct Deck Need Input Deck Cannot Reproduce Error Need Input Deck

Cannot Reproduce Error

6

RETRAN-02 MOD005.2 Ready for Licensing Applications
G. Swindlehurst, Duke Energy Based on a recent discussion with the NRC, RETRAN-02 MOD005.2 can now be referenced in licensing applications. The licensing history leading up to this recent development is as follows. RETRAN-02 MOD005.0 received an SER from the NRC in a letter dated November 1, 1991. On February 10, 1994, the RETRAN Maintenance Group notified the NRC of the release of MOD005.1. The NRC extended the MOD005.0 SER to cover MOD005.1 by letter dated April 12, 1994. On March 27, 1997, the RETRAN Maintenance Group notified the NRC of the release of MOD005.2 and requested a similar extension of the SER. A supplemental submittal on MOD005.2 was provided on September 2, 1997. During several phone calls with the NRC, the NRC indicated that they could not extend the SER for MOD005.2 as had been done previously for MOD005.1, and that the code would need to be submitted for a formal review. This was not a desirable process for the Maintenance Group, and so a meeting with the NRC was pursued. On November 20, 1997, Gregg Swindlehurst of Duke Energy (Maintenance Group Chairman) met with Tim Collins (NRC Reactor Systems Branch Chief) to

Accepted

explore options for moving forward with the NRC review. Based primarily on MOD005.2 not including any new models, and also since MOD005.2 has been developed under an Appendix B QA program, the NRC has taken the position that formal NRC review is not necessary for MOD005.2. The NRC did request a letter from the Maintenance Group describing the MOD005.2 code release. The requested letter was sent from Duke Energy to the NRC on November 24, 1997. A copy of the letter will be distributed to the users in a future mailing. RETRAN users can now utilize MOD005.2 in licensing applications, provided that the limitations of the original SER are observed. Future code versions that do not involve any new models will be handled in the same manner. The NRC's oversight of the use of MOD005.2 and other earlier versions will continue to be by audit or inspection.

Use of EPRI Codes at the Paul Scherrer Institute in Switzerland (Cont'd)
transferred to RETRAN-3D. The successful implementation of this task will represent a major step towards unifying the code system with the result that: The routine use of the RAMONA code can be terminated upon the successful migration of the stability analysis to RETRAN-3D, and the use of the 'cumbersome' one-dimensional kinetics can be reduced to a bare minimum. In addition, all of the core dynamics analysis (previously performed with RAMONA for BWR's and a German 3D code for PWRs) can be performed with CORETRAN without any further transformation of the nuclear data.. The goals of STARS cannot be achieved without constant involvement in research activities to assess inherent limitations of the current methodology and to open up new modes of safety analysis. Two examples in relation to the EPRI codes are: PSI is taking the lead in collaborating with CSA to enhance the models pertaining to mass transfer in the RETRAN-3D five-equation flow model (see companion article by Peter Cebull, CSA) in an attempt to extend the range of application of RETRAN-3D. One promising way to achieve acceptably low turn-around times for large simulations including 3D kinetics is to spread the computational effort over several processors. The STARS project intends to take profit of this new calculational technology through a collaboration with Purdue University in the area of parallel RETRAN/CORETRAN that will start this summer. In conclusion, the EPRI codes and RETRAN-3D in particular have become much appreciated tools for the analysis of reactor transients at PSI and our involvement will continue. 7

About This Newsletter
RETRAN Maintenance Program
The RETRAN Maintenance Program is part of a program undertaken by EPRI to provide for the support of the software developed in the Nuclear Power Division. The main features of the Subscription Service include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN.

Newsletter Contributions
The RETRAN Newsletter is published for members of the Subscription Service program. We want to use the newsletter as a means of communication, not only from EPRI to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. The next newsletter is scheduled for June 1998 and we would like to include a brief summary of your RETRAN activities. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to the E-mail addresses below by June 5, 1998. Contributors will receive a RETRAN mouse pad. We are looking forward to hearing from all RETRAN licensees. James McFadden Mark Paulsen Garry Gose Pam Richardson mcfadden@csai.com (208) 529-1700 mpaulsen@csai.com gcg@csai.com pam@csai.com

The RETRAN Maintenance Program now has 31 organizations participating in the program, including 23 member utilities, 5 organizations from outside of the U.S., and 3 nonmember utilities from the U.S. A Steering Committee, composed of representatives from the participating organizations, advises EPRI on various activities including possible enhancements for the code and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from Lance Agee Electric Power Research Institute P. O. Box 10412 Palo Alto, CA 94303 lagee@epri.com or (650) 855-2106

The RETRAN Web Page is located at http://www.csai.com/retran/index.html.

EPSC Contacts
EPSC Hours: EPSC Hotline: EPSC Fax: Email: 7 a.m. to 8 p.m. EST (800) 763-3772 (619) 453-4495

Calendar
Ninth International RETRAN Meeting June 7-10, 1998 Marriott Hotel Monterey, CA Basic RETRAN Training September 14-18, 1998 CSA Idaho Falls, ID Advanced RETRAN Training September 22-25, 1998 CSA Idaho Falls, ID

For Nuclear Quality Assurance related questions, call Clark Wallace at (619) 622-6611.

Please supply us with technical tips for our new section

TechTips

and you will receive a

RETRAN mouse pad. pad.

8

INSIDE
Features Articles NRC Review News ........ From the Editor............... SPERT III E Tests ............ PWR MSLB Benchmark.

1 1 2 5

Published by Computer Simulation & Analysis, Inc. for EPRI

In Every Issue R2 Trouble Reports ........ 7 R3D Trouble Reports ..... 8 Tech Tip ........................... 9 Calendar .......................... 10

NRC Review News
G. B. Swindlehurst, Duke Energy
The NRC Review of RETRAN-3D, which began in August 1998, continues. The most recent meeting was May 8, 2000, in Washington, D.C. and the objective was to obtain a detailed list of the remaining NRC review comments. The intent is to respond with the required information and to obtain as favorable of an SER (Safety Evaluation Report) as possible. In the May meeting, the status of the responses to the previous two RAIs (Request for Additional Information) were discussed. While no formal conclusions were given at the meeting, the NRC staff will be providing specific information on the acceptability of the RETRAN-3D validation work and the responses to the second RAI within the next few weeks. EPRI/CSA will be responding to any resultant action items. The NRC is currently drafting the SER, with completion planned this summer and more meetings with the NRC staff, the ACRS Subcommittee, and the full ACRS are expected. It has become apparent that the NRC staff is conducting a very thorough review, more typical of the scope and depth of review that has been associated with LOCA codes. This approach can be expected on future reviews, and in particular on plant-specific applications of RETRAN-3D. The NRC has requested that EPRI, through the RETRAN Maintenance Group, guide RETRAN users on the transition from RETRAN-02 to RETRAN-3D and communicate the NRC's expectations. This will be communicated to the RETRAN community upon issuance of the SER later this year. Previous to the May meeting, discussions were held with the NRC staff on December 16, 1999, and a submittal dated March 6, 2000, focused on providing responses to the remaining known NRC issues. The RETRAN review was also on the agenda for the March 15, 2000, ACRS Subcommittee meeting on ThermalHydraulic Phenomena.

From the Editor
This issue of the Newsletter features two technical articles on the use of RETRAN-3D and applications to multidimensional kinetics models. Both tasks were undertaken to demonstrate the ability of the code to perform analysis that were not possible with RETRAN-02 and to add to the code validation database. We would like to encourage all members of the RETRAN community to contribute articles for the newsletter that show how the code is used in your group. We have had good support in the past from DUKE Energy, PSI, TEPCO, IBERINCO, UITESA, INVAP, KEPRI/KEPCO, and GPUN to name a few, and we invite all of you to submit any articles that might be interesting to the user community. If you have any ideas or suggestions about how we might improve the newsletter, please give us a call or send an e-mail. If you have been considering an article, remember that there is a reward. We are offering the world famous RETRAN polo shirt. It has been shown to lower your golf score, make you look ten years younger, and makes a great conversation starter at class reunions and wedding receptions.

The RETRAN Newsletter – May 2000

1

RETRAN-3D Multidimensional Kinetics Calculations for SPERT III E Tests 81 and 86
G. C. Gose, CSA
Two RETRAN-3D calculations were performed to simulate SPERT (McCardell, 1969) experimental reactor tests. The project was undertaken to expand the validation base for the RETRAN-3D multidimensional kinetics model and to demonstrate the ability of the code to reproduce experimental data. The calculated results and experimental data were compared for transient power and energy release. Even though it can be difficult to interpret the calculated results within the context of the experimental uncertainties, measured responses from significant and rapid reactivity changes in a nuclear core are rare. It is important to expand the validation base of the codes using these experiments when possible. The SPERT reactor was a small oxide-fueled, PWR that was characterized as generally having characteristics of commercial PWRs at that time (1969). During the 1960s, reactivity accident tests were performed in the SPERT III E-Core Reactor under the SPERT experimental program. The program was designed to obtain the kinetics response data of reactivity accidents and evaluate computer codes that were used to predict reactor kinetics behavior. The SPERT experimental program is unique because it is one of the few facilities where prompt critical tests have been performed. The two tests that were analyzed, Tests 81 and 86, were initiated by rapid reactivity insertions from hotstandby and operating-power initial conditions, respectively. These two tests were selected because they represented conditions from the SPERT high-initial-power test series. The high-initial-power tests were considered the most severe of the SPERT tests because the steady-state fuel temperatures were nearer the melting point, the reactor core would contain more energy, and the power burst energy release would be considerably large. SPERT III Reactor Description The SPERT III E-Core Reactor had the characteristics of an unborated commercial PWR, except for its small size. The E-core fuel is comprised of 4.8% enriched UO2 fuel rods contained within stainless steel assembly cans. The fuel is in the form of 0.42-inch (0.010668-m) diameter pellets contained in stainless steel tubes. The core characteristics are summarized in Table 1. Table 1. SPERT III E-Core Characteristics Core Configuration Assembly Types Fuel Rod Length Active Fuel Length Assembly Pitch Fuel Rod Outer Diameter Clad Thickness Control Rods Transient Rod ~ 26-inch (0.6602-m) Diameter 48 25-Rod Assemblies 16 16-Rod Assemblies 40.8 in. (1.03632 m) 38.3 in. (0.97282 m) 0.585 in. (0.014859 m) 0.466 in. (0.011836 m) 0.020 in. (0.000508 m) 8 Total – 2 per Quadrant Poison Section: 1.1684 m Fuel Follower: 1.1593 m 1 Central Cruciform Shape Rod Poison Section: 0.9652 m Cylinder

43

42

41

R N

14

24

34

34

33

R 32 W

31 N 21 R 12

23

R

33

43

24

23 W

32

42

14

12

11

R

11

21

R E

31

41

41

31

21 S

11

R

11 E

12

14

42

32

R

12 S

R

21

23 R

24

43

33

23

31

32

33

34

34

R 24

14

R

41

42

R

43

Figure 1. SPERT III E Core Configuration

The SPERT core coordinate system is given in Figure 1. The majority of the fuel rods are contained in 48, 3.0-inch (7.62-cm) square canisters each containing 25 fuel rods in a 5 by 5 rectangular array. To provide control, there are 12 smaller 2.5-inch (0.0635-m) square fuel assembly cans each containing 16 fuel rods in a 4 by 4 rectangular array. Four of these assemblies surround the central transient rod and the remaining eight form the fuel followers of the eight control rods. The cruciform shaped transient rod that is located at the core center is used for initiating the reactivity accidents.

Poison Section: Type 18-8 Stainless Steel with 1.35 Wt% B-10 Upper Section: 18-8 Stainless Steel; Poison Section: 1.35 Wt% B-10

2

The RETRAN Newsletter – May 2000

The four control rod pairs, indicated on Figure 1, by shading, are placed in the core in rotational symmetric locations. The transient rod is bottom inserted and the control rods are top inserted. The RETRAN-3D thermal-hydraulic model consists of the active core region and thermal-hydraulic boundaries. The core consists of 48 active channels and a single bypass path. Within each channel, there are 30 axial nodes consisting of thermalhydraulic control volumes and core conductors. There are 31 flow junctions per channel. Figure 2 shows the thermal-hydraulic model. Reactor Modeling Cross-section files that represent the core conditions for Tests 81 and 86 were supplied from a previous calculation. These files were in the format used by the NESTLE code (Turinsky, 1996). The NESTLE crosssection model is based upon polynomials for both controlled and uncontrolled states, and the RETRAN-3D cross-section routines were modified in order to use this format. The RETRAN-3D core model for SPERT is very simple, consisting of three assembly types (Types 1-3) for the active fuel and one (Type 4) for the reflector. Each fuel assembly has 30 axial nodes for the active portion, and a single node for both the top and bottom reflectors. The active core is surrounded by one row of assemblies representing the radial reflector. The layout of assembly types is illustrated in Figure 3. For thermal-hydraulic calculations, the fuel assemblies are modeled by 48 flow channels, and the reflector region is model by a single bypass path. The RETRAN-3D channel map is shown in Figure 4. Transient Modeling The modeling of the control rods and the transient rod is crucial to the simulation of the reactivity accident tests. The excursions were initiated by dropping the transient rod poison section from the core.

8 (Upper Plenum)
131 1310 1300 1290 1280
Conductors

48310 48300 48290 48280 48300 48290 48300 48280 48290 48280 130 129 128 130 129 128

1300 1300 1290 1280 1290 1280
Volumes Junctions Active Core 48 Channels

Core Bypass 1 Channel

1030 1020 1010
Channel 1

1040 1030 1030 1020 1020 1010 1010

48030 48020 48010

48030 48020 48010

48030 48030 48020 48010

103 102 101

104 103 102

Channel 48

101

9 ( Lower Plenum)

Figure 2. SPERT RETRAN-3D Model To simulate the procedure, the first task is finding the initial transient rod position required for the power excursion. The transient rod tip was first positioned to the top of the bottom reflector, and then a criticality search was performed manually by moving the control rod positions until a critical condition was obtained. For Tests 81 and 86 the required reactivity insertion was $1.17. The transient rod was then inserted into the bottom of the core with the thermal-hydraulic feedback frozen at the initial conditions until the desired rod worth was obtained. The transient was modeled by moving the transient control rod to the final position determined from the above procedure, and the core was maintained at critical by adjusting the position of the control rods. The transient rod was then dropped out of the core with the designed acceleration of 2 2 (50.8 m/sec ) to 2000 in./sec simulate the reactivity insertion rate. Test 81 SPERT Test 81 is one of the highinitial-power test cases. The steadystate core power is about 1 MW, which represents the hot-standby condition. The amount of reactivity insertion is $1.17. The comparison of the RETRAN-3D results with experimental data are plotted in

4 4 4 4 4 4 4 4 4 4

4 4 3 3 3 3 3 3 4 4

4 3 3 3 2 3 3 3 3 4

4 3 3 2 3 3 2 3 3 4

4 3 3 3 1 1 3 2 3 4

4 3 2 3 1 1 3 3 3 4

4 3 3 2 3 3 2 3 3 4

4 3 3 3 3 2 3 3 3 4

4 4 3 3 3 3 3 3 4 4

4 4 4 4 4 4 4 4 4 4

Figure 3. RETRAN-3D Assembly Layout

40 41 41 42 43 43 43 24 25 26 27 28 29 44

40 23 9 10 11 12 30 44

40 22 8 2 3 13 31 45

39 21 7 1 4 14 32 46

38 20 6 5 16 15 33 46

38 19 18 17 36 35 34 46 37 37 37 48 47 47

Figure 4. RETRAN-3D Channel Layout

The RETRAN Newsletter – May 2000

3

Figures 5 and 6 which include transient power and total reactivity. Given the experimental uncertainties, the agreement between the RETRAN-3D results and the experimental data are excellent. RETRAN-3D captured both the time and the magnitude of the peak power, also the pre- and post-peak power agreed very well with the experimental data. The added energy and the net reactivity (calculated based on the measured power and point kinetics equations) were correctly predicted. Test 86 SPERT Test 86 is a full power test case. The initial core power is 19 MW. The amount of reactivity insertion is $1.17. The comparison of the RETRAN-3D results with experimental data were plotted in Figures 7 and 8 which include transient power and total reactivity. In general, the results are similar to those obtained in Test 81 but the magnitude and the time of the calculated power peak (710 MW at

0.114 sec) are different from the experimental data. It is noted that the experimental power peak that was shown in Table VII of the SPERT report (McCardell, 1969) was 610 MW at 0.110 sec. The power peak plotted in Fig. D-70 of the report, however, was 598 MW at 0.122 sec. The reason for this discrepancy is unknown. Conclusions The RETRAN-3D code has been applied to the simulation of SPERT Tests 81 and 86, which are reactivity accident tests in a small experimental PWR. The transients were initiated by rapid reactivity insertions and experimental data included the subsequent time-dependent power and energy release. The results from RETRAN-3D indicate that the fundamental trends in both tests are correctly reproduced. The transient power and added energy were accurately predicted in Test 81, suggesting that the reactivity insertion rate be correctly modeled. The uncertainties in the fuel conduction model were not pronounced in the hot-standby case.

A significant factor in the interpretation of the results presented here is the degree of experimental uncertainty. For example, the documented standard deviation in reactor power for Tests 81 and 86 was +10% (as much as 60 MW for Test 86), the time of the peak power, +5 msec, and the uncertainty in the inferred reactivity (not a measured parameter) was about +4%, translating to about $0.05 uncertainty in the reactivity insertion for the $1.17 cases. To summarize the RETRAN-3D analysis work, the results show that the peak values and timing of the two cases are reasonably captured with the best comparison for the hotstandby initial power Test 81. More detailed sensitivity studies involving the fuel pin conductance may identify the significant parameters that affect the higher initial power case performance. It can be concluded that RETRAN-3D can produce good comparisons with experimental data from transient kinetics systems.

Figure 5. Test 81 - Reactor Power

Figure 6. Test 81 - Net Reactivity

Figure 7. Test 86 - Reactor Power

Figure 8. Test 86 - Net Reactivity

4

The RETRAN Newsletter – May 2000

Pressurized Water Reactor Main Steam Line Break Benchmark
J. G. Shatford, CSA
A PWR main steam line break (MSLB) benchmark was jointly sponsored by the Organization for Economic Cooperation and Development and the U.S. Nuclear Regulatory Commission. The benchmark was established to demonstrate coupled threedimensional neutronic/thermalhydraulic behavior and compare predictions from different codes. The MSLB event is characterized by a significant radial power shift caused by asymmetric cooling and the assumption of a stuck-out control rod. The power shift can only be predicted with three-dimensional kinetics. One concern with the MSLB is a return to power after the scram due to reactivity addition from reactor coolant system temperature decrease. Historically, this transient has been analyzed with point kinetics using conservative assumptions that compensate for the inability to simulate the power shift. The benchmark consists of three phases: Exercise I – a system simulation using point kinetics, Exercise II – a core only simulation using three-dimensional kinetics,
88 8
208 208 9 19 108 108
131 181 208 131 108

S T E AM G EN ER A T O R B
J209

L O O P A H O T L EG

L O O P B H O T L EG

135

P ORV

S T EA M
J109

13 0
136 H E A TE R S

GE NE RA T O R A 10 9

P RE S S U RI Z E R
230

2 09
J 2 64 26 3 21 0

11 0

16 3

J1 64

E FW

EF W

2 62

21 1

130

1 11

16 2

26 4

2 61

21 2

1 12

16 1

16 4

26 0

21 3

11 3

16 0

132
J 2 65 25 9 21 4 11 4 15 9

J16 5

J 2 51

J 2 50

25 8

21 5

228 227

128 127

11 5

15 8

J1 50

J1 51

25 7

2 16

116

1 57

25 6

2 17

224

225

12 5

124

117

1 56

242
25 5

14 2
11 8

25 1
25 4

21 8

15 5

21 9

R EA C T O R C O O L AN T P U MP S

H PI

HPI

R E A CT O R C O O LA N T P U MP S

15 1
11 9 15 4

P RE S S U RE V ESS EL

22 0

2 52 J 25 2

L O O P A CO LD L E G S

2 53

1 20

153

LOOP B C OLD LE GS

22 1

1 21

152 J1 52

2 22

1 22

226 223

126 12 3

Figure 1. TMI RETRAN-3D Two-Loop Model Nodalization Diagram and Exercise III – system simulation using coupled three-dimensional neutronics/thermal-hydraulics. The benchmark was intended to provide enough detail to ensure that all simulations use consistent input. The cross sections for Exercises II and III were provided. The specification defines most key parameters and modeling methods. The MSLB benchmark analysis was performed jointly by Computer Simulation & Analysis and GPU Nuclear. These analyses were performed with RETRAN-3D. The RETRAN code was modified to allow the cross-section data to be used as supplied. The plant selected for the MSLB benchmark was the Three Mile Island Unit 1 (TMI-1). This is a twoloop plant with a once through steam generator on each loop. A nodalization diagram is shown in Figures 1 and 2. The core is split into faulted and unfaulted halves (Figure 2). The transient was initiated by a guillotine rupture of a steam generator steam line. The MSLB results in the blowdown of a single steam generator, causing extreme cooling of the faulted loop but minimal cooling from the unfaulted steam generator. Flow from the two loops mix to some degree. However, the temperature in the half core on the faulted loop side decreases much more than the unfaulted side due to incomplete

234 228 27 233 225

5

15

134 128

4

14

30 133

7
26 3 6 13 29

17

125

25

2

12

28

1

11

Figure 2. TMI RETRAN-3D Split Vessel Nodalization

The RETRAN Newsletter – May 2000

Faulted Loop

Unfaulted Loop

5

loop flow mixing. The benchmark defines how the loop flow mixing and asymmetric core cooling is to be modeled. The split plenums and core shown in Figure 2 allow simulation of incomplete loop flow mixing. The moderator temperature coefficient is negative; consequently, the power will increase and tilt to the core side with the colder fluid. After the scram, there is the possibility of the net reactivity becoming positive if reactivity addition due to cooldown (from fuel and moderator) becomes larger than the scram worth. The RETRAN-3D system power and component reactivities from Exercise I are shown in Figures 3 and 4. The power drops rapidly after the reactor trip and then at 20 seconds the power begins to increase and reaches a maximum around 60 seconds. As can be seen from the component reactivities in Figure 4, shortly after 50 seconds the reactivity contributions from the fuel and moderator exceed the control reactivity and the total reactivity momentarily becomes positive.
3.5E+09 3.0E+09 Total Power, W 2.5E+09 2.0E+09 1.5E+09 1.0E+09 5.0E+08 0.0E+00 0 20 40 60 80 100 Time, seconds

Exercise II is a plenum-to-plenum model. The thermal-hydraulic conditions are completely defined by the benchmark, so Exercise II shows the effect of different threedimensional kinetics modeling. The benchmark provided cross sections for 24 axial nodes for each of the 177 fuel assemblies that are a function of fuel temperature and moderator density. Consequently, there are 4248 neutronic nodes within the reactor core. The benchmark defined thermalhydraulic nodalization that was less detailed than the neutronic nodalization. The core was divided into 18 parallel “flow channels” between the lower and upper plenums. The split plenums were retained from Exercise I, so there are nine flow channels between the two plenum volumes on each side. The 18 flow channels are distributed in a symmetric radial manner. Since there are less flow channels than fuel assemblies, several fuel assemblies are placed within a single flow channel. The flow channel/fuel assembly map is shown in Figure 5 where each square represents a fuel assembly and the numbers (1 to 18) represent the flow channel.

these conservative assumptions, point kinetics models typically predict a return-to-power while three-dimensional kinetics models do not. A second set of cross sections based on a less conservative tripped rod worth were supplied as a better test of the coupled three-dimensional kinetics/ thermal-hydraulic codes. The transient power response and total reactivity for Exercise II for both cross-section sets are shown in Figures 6 and 7.
3.5E+0.9 3.0E+09 Total Power, W 2.5E+09 2.0E+09 1.5E+09 1.0E+09 5.0E+08 0.0E+00 0 20 40 60 80 100 Time, seconds Return-to-Power No Return-to-Power

Figure 6. Exercise II Total Core Power
1.0E-02 Reactivity, dK/K 0.0E+00 -1.0E-02 -2.0E-02 -3.0E-02 -4.0E-02 -5.0E-02 0 20 40 60 80 100 Time, seconds Return-to-Power No Return-to-Power

Figure 7. Exercise II Total Reactivity For RETRAN-3D, Exercise II results varied significantly from the full system model (Exercise III) since Exercise II uses fixed flow rates, specified temperature and pressure boundaries and the scram time is predetermined. Exercise II was used mainly to verify code modifications allowing use of the specified crosssection data. Exercise III uses the Exercise I system model with the three-dimensional kinetics core model from Exercise II. Consequently, this provides a direct comparison of the core behavior using the two reactor kinetics models.

Figure 3. Exercise I Total Core Power
3.0E-02 2.0E-02 Reactivity, dK/K 1.0E-02 0.0E+00 -1.0E-02 -2.0E-02 -3.0E-02 -4.0E-02 0 20 40 60 80 100 Time, seconds Return-to-Power Total Doppler Control Moderator

Figure 5. Exercise III 3-D Kinetics Core Fuel Flow Bundle/Flow Channel Map The benchmark provided two separate sets of cross-section data. The first set, based on current licensing practices, uses a very conservative tripped rod worth. For

Figure 4. Exercise I Reactivity Components

6

The RETRAN Newsletter – May 2000

There is a power increase after scram in the three-dimensional kinetics case but less than point kinetics (Figure 8) and there was not a return to power as can be seen in the total reactivity response (Figure 9). The response of the rest of the system was quite similar in the two cases. The different power response in the two cases is due to the ability to simulate the extreme radial flux shift to the faulted half of the core in addition to the more detail in the
3.5E+09 3.0E+09

local feedback in the region of power increase. Additional margin relative to a return-to-power concern can be achieved by using three-dimensional kinetics and more detailed thermalhydraulic nodalization in the regions where the power excursion is expected to be the most significant. More accurate thermal-hydraulic reactivity feedback can be achieved
5.0E-03 0.0E+00 Reactivity, dK/K
Phase 3 Phase 1

when several assemblies are not placed within a single flow channel. Further from the stuck rod, the number of fuel assemblies per flow channel should become less important. To optimize the number and placement of flow channels, sensitivity studies should be performed to define the point where more detail does not change the power response.

Total Power, W

2.5E+09 2.0E+09 1.5E+09 1.0E+09 5.0E+08 0.0E+00 0 20 40 60

-5.0E-03 -1.0E-02 -1.5E-02 -2.0E-02 -2.5E-02 -3.0E-02 -3.5E-02 Phase 3 Phase 1

80

100

0

20

40

60

80

100

Time, seconds

Time, seconds

Figure 8. Exercise III/Exercise I Total Core Power Comparison

Figure 9. Exercise III/Exercise I Total Reactivity Comparison

Summary of RETRAN-02 Trouble Reports
The following is a summary of RETRAN-02 Trouble Report/Code Maintenance Activity as of April 30, 2000. There are 12 outstanding trouble reports. A list of trouble reports and the status can be obtained directly from the EPSC (1-800-763-3772). Additional information is available from the RETRAN-02 Trouble Report Page at http://www.csai.com/retran/r02trpt/index.html. NO. TROUBLE REPORT TYPE OF PROBLEM 354 376 394 408 431 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 Large Step Change in PHIR Control Reactivity, No Motion Anomalous Heat Trans. Behavior OTSG Heat Transfer Problems Failure in JN Properties Prandtl Number is Discontinuous Heat Transfer Logic/CHF Restart Failure/Pipe Transport Decay Heat Input Kinetic Energy/Time Dep Area Anomalous Power Increase Poor Diagnostics Liquid Region Work Term Positive Slip Velocity Boron Transport Inconsistency Theory Manual Problem in Bubble Rise Smoothing Algorithm in SVOID Decay Heat Input LOFW Transient Behavior Momentum Flux Error Non Right Angles CORRECTION NO. IDENT *** *** *** *** 406 405 --407 *** *** --*** *** *** *** *** *** ----*** ******* ******* ******* ******* MOD005P3 MOD005P3 ------MOD005P3 ******* ******* ------******* ******* ******* ******* ******* ******* ------------******* COMMENTS

Not a Code Error

Not a Code Error

TH Manual Mod. User's Manual Mod. Code Limitation

The RETRAN Newsletter – May 2000

7

Summary of RETRAN-3D Code Trouble Reports
A total of 209 trouble reports had been filed as of April 30, 2000. Of these, 177 reports have been resolved, while 32 remain unresolved. A summary of the unresolved trouble reports is shown below. Additional information for RETRAN-3D trouble reports is available at http://www.csai.com/retran/r3dtrpt/index.html. NO. 22 30 40 48 52 54 60 70 81 116 122 142 144 145 150 152 164 165 168 170 174 181 182 190 197 198 200 201 202 203 204 205 206 207 208 209 TROUBLE REPORT TYPE OF PROBLEM Problem using Wilson bubble rise model & error when using low power initialization 2-loop Oconee w/5-eq. fails in steady state Results do not agree with data Steady state fails after 6 iterations MOC does not return to the initial temp. MOC solution; no null transient for two-phase Anomalous countercurrent flooding Fails in subroutine DERIVS Steady-state failure at iteration #6 Fails in steady-state initialization Problems with EOS convergence Timestep selection causes 3-D kin to fail TAUGL model doesn't apply for horiz. flow SS fails to converge for low press. and flow SS solution void fraction oscillation Junct pressure lags vol pressure 1 time step 3-D kinetics causes floating point exceptions 3-D kinetics unable to specify profile fit for subcooled boiling model Incorrect null trans w/3d Kin., mod ht & 5eq PARCS numerics will not hold a null transient 5-EQ error in steam lines No rod cusping treatment in 3D kinetics Kinetics problem type is fixed at 3 Error when reversing from/to junc. w/ angle >1 geometry data set is supplied on the CDI Momentum flux error – if junction angles are not 0, 90, 180, 270 SS failure for NCG (WAT0 error maybe WAT17) SS failure when flow split option used Error when pcrit reached during tran – 5-Eq Pressurizer time step selectn when Przr solid Impl Przr – Int reg HT and spray mdl errors Channel model doesn’t allow dyn gap cond mdl PARCS inner iteration BICGSTAB fails Xsec Extrapolation on DM is not supported PARCS BC=2 (no return flux) is not allowed SLB sample problem using direct mod. heating CORRECTION NO. IDENT *** *** *** *** 006 *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** 020 *** *** *** *** *** *** *** *** *** *** 022 023 024 *** ******* MOD001 ******* ******* ******* MOD001g ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* ******* COMMENTS (partial fix)

(partial fix)

(water packing)

Model limitation

Model limitation 3D kinetics

ARROTTA Source ARROTTA Source ARROTTA Source

8

The RETRAN Newsletter – May 2000

Simulating a Second-Order Control System in RETRAN-02
G. C. Gose, CSA
Sometimes it is necessary to simulate the response from a second-order control system but the elements for this task are not directly available in RETRAN-02. A recent update to RETRAN-3D has added a second-order control block, STF, allowing the user to directly model the elements of a second-order system, given by the Laplace transform 1/(1 + a2s + a1s ) . The RETRAN-3D input allows the user to specify the block gain and the characteristic constants, a1 and a2. But, what about the RETRAN-02 user? Can the behavior of a second-order system be simulated? Well, the answer is yes, with a little algebra. The Laplace transform of a LAG block looks like 1/(1 + t1s) . By combining two LAG blocks together (that is, feeding the output of the first into the input of the second) an equivalent Laplace transform results 1/(1 + (t1 + t2) s + t1t2s2) . This looks very similar to the transform for the secondorder system. An equivalent behavior to the second-order block can be obtained if the following constraints are followed a1= t1*t2 and a2= t1+t2 .
2

Thus, subject to this constraint, users can combine LAGLAG blocks to achieve a second-order system response. As an example, let’s look at a comparison of the method for a second-order system with the following constants, 4y” + 4y’ + y = x(t) where x(t) is some arbitrary function, say a step or a ramp. In Laplace transform space, this is equivalent to 1/(1 + 4s + 4s2) . A little algebra shows that this is equivalent to 1/((1 + 2s)*(1 + 2s)) which is the same as the combination of two LAG blocks, each with time constants of 2.0. For illustration, we have selected the response of a second-order block (STF) and the LAG-LAG block to a simple unitary step change at four seconds. The accompanying figure shows that for the chosen function (a unit step function) the two control system responses appear to be nearly identical. Thus in many cases, one can model a second-order control system response in RETRAN-02 if a set of equivalent constants can be found that match the ‘transformed’ second-order system.
CONTROL SYSTEM PROBLEM (SECOND DERIV BLOCK VS LAG-LAG) 01/05/00 12:17:57

1.50

LAG STEP INPUT

LAG
OUTPUT

LAG-LAG STF 1.00

STF

0.50

0.00 0.00

2.00

4.00

6.00

8.00

10.00

12.00 (SEC)

14.00

16.00

18.00

20.00

SYSTEM ELAPSED TIME

The RETRAN Newsletter – May 2000

9

About This Newsletter
RETRAN Maintenance Program
The RETRAN Maintenance Program is part of a program undertaken by EPRI to provide for the support of the software developed in the Nuclear Power Division. The main features of the Subscription Service include: ? ? ? the code maintenance activities for reporting and resolving possible code errors, providing information to users through the User Group Meetings and this newsletter, and preparing new versions of RETRAN.

Newsletter Contributions
The RETRAN Newsletter is published for members of the Subscription Service program. We want to use the newsletter as a means of communication, not only from EPRI to the code users, but also between code users. If this concept is to be successful, contributions are needed from the code users. The next newsletter is scheduled for August 2000 and we would like to include a brief summary of your RETRAN activities. Please provide your contribution to CSA, P. O. Box 51596, Idaho Falls, ID 83405, or to the E-mail addresses below by August 4, 2000. Contributors of a feature article will receive a RETRAN polo shirt. We are looking forward to hearing from all RETRAN licensees. Mark Paulsen Garry Gose Pam Richardson paulsen@csai.com gcg@csai.com pam@csai.com (208) 529-1700

The RETRAN Maintenance Program now has 26 organizations participating in the program, including 22 U.S. utilities and 4 organizations from outside of the U.S. A Steering Committee, composed of representatives from the participating organizations, advises EPRI on various activities including possible enhancements for the code and the scheduling of future code releases. Information regarding the Maintenance Program can be obtained from Lance Agee EPRI P. O. Box 10412 Palo Alto, CA 94303 lagee@epri.com or (650) 855-2106

The RETRAN Web Page is located at http://www.csai.com/retran/index.html. Previous issues of the RETRAN Newsletter are available from the RETRAN Web Pages at http://www.csai.com/retran.

EPSC Contacts
EPSC Hours: 7 a.m. to 8 p.m. EST EPSC Hotline: (800) 763-3772 EPSC Fax: (619) 453-4495

Calendar of Events
June 12-16 RETRAN Training Idaho Falls, ID Oct. 17-19 User Group Meeting TU Electric

Please supply us with technical tips for our section and you will receive a RETRAN mouse pad.

Your contributions are greatly appreciated. We, EPRI and CSA, encourage everyone to participate in this newsletter. 10 The RETRAN Newsletter – May 2000

The RETRAN Newsletter
Summary of Activities
This issue of the RETRAN Newsletter contains information on obtaining code versions from the EPSC and several interesting articles from a number of code users. You contributions are greatly appreciated. We, EPRI and CSA, encourage everyone's participation in this newsletter. In addition to these topics, a description of the RETRAN Maintenance Program is included, as well as information about how to make contributions to this newsletter.
In This Issue . . . EPSC Bulletin Board . . . . . . . . . . . . . . . . . . . . 1 Auxiliary Feedwater Pump Turbine Steam Supply and Drain System . . . . . . . . . . . . . . . . 2 User Group Meetings . . . . . . . . . . . . . . . . . . . . 2 The Development of the Evaluation Methodology of the LTOP Accident in PWR 3 CSA Launches Web Page . . . . . . . . . . . . . . . . . 3 Oscillations Due to Poor Control Block Ordering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 LOFTRAN-AP . . . . . . . . . . . . . . . . . . . . . . . . . . 5 RETRAN-02 Trouble Reports . . . . . . . . . . . . 5 RETRAN-3D Trouble Reports . . . . . . . . . . . . 6 ComEd BWR Transient Analysis Licensing Topical Report . . . . . . . . . . . . . . . . . . . . . . . . . 8 Calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Void Fraction Modeling for Steady-State and Dynamic Environments . . . . . . . . . . . . . . . . . 9 New RETRAN Users . . . . . . . . . . . . . . . . . . . . 10 Newsletter Contributions . . . . . . . . . . . . . . . . . 10

Published by Computer Simulation & Analysis, Inc. for the Electric Power Research Institute

June, 1996

Procedure for Obtaining QA Version of Code
Call the EPSC at 1-800-763-3772 to confirm membership status and verify license. EPSC requires a written record of the request which can be faxed, e-mailed, or sent via regular mail. A help desk technician will assist the caller in identifying the most appropriate communications channel. Please note: Controlled copies can only be distributed to utilities with signed letter agreements to all known nuclear utilities.

Availability of RETRAN-02 MOD005.2
RS-6000/AIX Version: Sun/UNIX Version: HP/UNIX Version: PC-DOS Version: Available 6/13/96 Available 6/13/96 Available 6/17/96 Waiting for software keys from Rainbow Technologies

EPSC Contacts
EPSC Hours: EPSC Hotline: EPSC Fax: E-Mail (EPRINET): 6 a.m. to 5 p.m. PST 1-800-763-3772 (619) 453-4495 ordepsc@eprinet.epri.com

For Nuclear Quality Assurance related questions, call Clark Wallace at 619-622-6611. 1

Auxiliary Feedwater Pump Turbine Steam Supply and Drain System Redesign Using RETRAN-02
J. Boatwright, TU Electric Several trips of the turbine-driven auxiliary feedwater pumps (TDAFWP) at Comanche Peak Steam Electric Station have been experienced during system startups. One of the primary contributors to these trips was determined to be excessive accumulation of condensate in the steam supply, and possibly, exhaust steam lines to and from the TDAFWP turbine. The causes of the excessive condensate ranges from long runs of steam supply piping subjected to cold start conditions to malfunctioning of steam traps in the drain system. To improve the system reliability, a major design modification was undertaken for the steam supply and drains systems. Briefly, the redesign of the system involved (1) decreasing the total volume of steam supply piping by connecting the two steam supply lines immediately downstream of the steam supply valves, (2) providing additional drain pots in the steam supply piping with flow restricting orifices to allow removal of steam traps while still limiting steam flow to the drains, and (3) providing a flash tank with steam vent and loop seal to separate condensate in the drains from the steam to prevent steam blowdown to the floor drains system. The design was analyzed using two RETRAN models: one of the steam supply, exhaust, and drain piping (the system model) and another of the flash tank (the flash tank model). Steam supply and turbine exhaust conditions for the full operating range of the turbine were imposed on the system model using time-dependent volumes. Condensation rates were determined using conservative hand calculations and imposed on the system model using a fill flux into one of the drain pots. No attempt was made to model the dynamic separation of the condensate from the steam. It was simply assumed that all condensate was separated from the steam and entered a single drain pot. Results from the steam model runs were used: (1) to verify that the steam flows to the turbine met design requirements over the full range of operating pressures, (2) to ensure that the capacitance of drain pots in the steam supply lines coupled with their flow restricting orifices was sufficient to prevent adverse condensate accumulation in the steam supply lines, (3) to size the drain pot flow restricting orifices and steam vents to limit steam blowdown to the drains system and direct it to the atmosphere, and (4) to obtain the mass and energy blowdowns to the flash tank. These mass and energy blowdowns were imposed on the flash tank using fill fluxes, one for the blowdown from the steam supply side drains and another from the exhaust side drains. The flash tank and loop seal were modeled using separated volumes. Results from the flash tank model were used to size the flash tank, loop seal, and steam vent in order to prevent steam blowdown to the floor drains system.

User Group Meetings
The RETRAN and VIPRE User Group Meetings will be hosted by Commonwealth Edison this year. The meetings are scheduled to begin the afternoon of September 30 and end the morning of October 3, 1996, and will be held at the CECo office in Downers Grove, Illinois. A block of rooms has been reserved at the Marriott Suites in Downers Grove for this meeting. The hotel is one block from the CECo office. Users are requested to prepare brief presentations of current RETRAN and VIPRE activities at their organization. Further information on this meeting will be mailed to all members of the RETRAN and VIPRE Maintenance Groups. If you have any questions, contact Pam Richardson or Garry Gose at CSA or Lance Agee at EPRI. 2

The Development of the Evaluation Methodology of the LTOP Accident in PWR
C. Lee and Y. Kim, Korea Electric Power Research Institute, KEPCO The low-temperature overpressure (LTOP) accident may cause brittle failure of the reactor pressure vessel during startup and cooldown operations in pressurized water reactor. To protect the reactor pressure vessel from the LTOP accident, a pressure versus temperature curve given by plant technical specification conservatively bounds the maximum allowable pressure for the temperature range where the reactor pressure vessel can suffer brittle fracture. Nuclear power plants have been equipped with either redundant power operated relief valves or redundant safety relief valves in the residual heat removal systems to prevent pressure increase from exceeding the embrittlement fracture limit. The purpose of KEPRI's project is to develop the evaluation methodology of LTOP accident in PWR. KEPRI has defined the severe LTOP events into two classes, such as the mass addition and the heat addition events. The limiting mass injection events assumed is the inadvertent start of one safety injection pump when RCS is water-solid condition. The possible source of injected flow should be decided by plant specific technical specification. For the analysis of these events, KEPRI has developed a simple computer program, LTOP_MI, and verified it with the results of RETRAN-03 MOD001f of the same inputs. The limiting heat addition event is assumed to be initiated by the start of a reactor coolant pump with steam generator secondary-side temperature 50oF higher than the primary side when RCS is water-solid condition. This leads the rapid expansion of the RCS inventory and is causing overpressure event. For the analysis of these transients, we have used RETRAN-03 MOD001f. The results of this project will be presented in the near future.

CSA Launches Web Page
The CSA Web page that includes information of interest to RETRAN users will be available to users August 1, 1996. The page will have ? ? ? ? general information on RETRAN activities, lists of RETRAN-02 and RETRAN-3D trouble reports, a bibliography of RETRAN-related publications, and a form to submit RETRAN trouble reports electronically.

Other features that are planned for the future include a news group. We will send the URL to all members of the Maintenance Group as soon as the RETRAN pages are available.

3

Oscillations Due to Poor Control Block Ordering
N. Newman, Scientech, Inc. In its description of the control system model, the RETRAN-02 MOD005.1 User's Manual clearly states that "the order in which the output of each block is calculated affects the numerical results ..." and that "the burden is upon the user to order computations in the control system in cascading order ...". An illustration of the truth of this was encountered recently. A control system had been constructed to model the torque applied to the recirculation pump in a BWR. Upon running the model for what should have been a null transient, fairly significant and rapid oscillations in the pump speed were observed to begin after a few seconds. Fortunately, this particular control system had been computer-generated, and it was a trivial matter to have the control system generated in cascade order. With this re-ordering, the control system behaved perfectly. A comparison of the computed pump speeds is shown in Figure 1. It is emphasized that the only difference in the input decks for these two cases was the selection of the card sequence numbers for the 703XXX control cards. This illustrates that the warning in the RETRAN User's Manual needs to be respected. If care is not taken to order the control system blocks correctly, incorrect results may be obtained.

Figure 1. Illustration of the Effect of Control Block Ordering

4

LOFTRAN-AP Analysis
S. Oh, EPRI Westinghouse used LOFTRAN-AP, a modified version of LOFTRAN, to analyze certain SSAR Chapter 15 transient analyses for AP600. One of the analyses is a steam line break transient (SLB). EPRI has performed the SLB analysis using RETRAN-03 MOD001b. The objective of this study is to understand similarities and differences between results from the two codes and models and to show that LOFTRAN SLB analyses are conservative. Similar to current operating plants, the main SLB at the hot zero power condition is examined. The scope of this study is limited to the thermal-hydraulic behavior. The core power history, obtained from the LOFTRAN run, was input into RETRAN. The main focus is possible flow stagnation and two-phase conditions in the unaffected loop, and their impact on core makeup tank (CMT) behavior. The CMTs are connected to the unaffected loop in this study. Two sets of RETRAN analyses were made. The first set was the counterpart of the LOFTRAN-AP analysis. RETRAN was run for 700 seconds. The second set was to examine long-term behavior. RETRAN was run out to 3000 seconds. The first set of results indicate that the LOFTRAN analysis is conservative and has a more rapid cooldown than that computed by RETRAN. The main difference is that the unaffected loop became stagnant and two phase in the RETRAN result. In LOFTRAN, forward flow through the unaffected loop was maintained. The second set of results indicate that the core makeup tank recirculation continued throughout the transient. AP600 is a 600 MWe advanced light water reactor, currently in the design certification process. Unlike current operating plants, AP600 utilizes passive safety injection systems, such as core makeup tanks, IRWST, and a passive RHR system. The reliance on passive safety systems requires a different challenge to the codes. Correct prediction of low flows induced by density differences and natural/mixed convection heat transfer are important. The RETRAN prediction of PRHR heat transfer and core makeup tank recirculation have been examined and found to be reasonable for the SLB analysis. However, it is recommended that a more systematic evaluation be performed before further application of RETRAN to AP600 transients.

RETRAN-02 Trouble Reports
The following is a summary of RETRAN-02 Trouble Report/Code Maintenance Activity. Unresolved Trouble Reports ? 1 From MOD001 ? 5 From MOD002 ? 4 From MOD003 ? 3 From MOD004 ? 3 From MOD005 A list of trouble reports and the status can be obtained directly from the EPSC.

5

Summary of RETRAN-02 Code Trouble Reports
NO. 1 61 121 140 177 209 272 317 334 342 354 366 376 394 408 413 TROUBLE REPORT TYPE OF PROBLEM Error 209 in TEMZ Delta T for Conductor with TDV OTSG Low Power Initialization Spurious Trips on High Level Overflow in WAT9 Pump Coast down Rates Junction Properties at Break Junction Property Error Time-Dep. Volume Input Control Block Output near Zero Large Step Change in PHIR Mixture/Liquid Level Difference Control Reactivity, No Motion. Anomalous Heat Trans. Behavior OTSG Heat Transfer Problems Incorrect Vsn No. in IBM Output CORRECTION NO. IDENT *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** COMMENTS MOD001 Error Need Input Deck Need Input Deck Need Input Deck Need Correct Deck Need Input Deck

Cannot Reproduce Error Need Input Deck

Cannot Reproduce Error

Summary of RETRAN-3D Code Trouble Reports
The Design Review of RETRAN-3D MOD001f is nearly finished. A summary of the Trouble Report/Code Maintenance Activity is shown below. It includes all errors that were not resolved in MOD001f and errors that have been reported since its completion. MOD001g has not been finalized and includes all error corrections and modifications that have been made to date.
NO. 2 5 6 7 8 10 11 19 24 26 27 28 29 30 31 32 33 34 35 TROUBLE REPORT TYPE OF PROBLEM Slip differences between prerel & pre56 Pressure increase in last SL volume 1-D Kin power level not converging Steam separator model fails Two-phase junction choking error 1st iteration failure during steady state Negative enthalpy when flow reverses Steady-state area adj. for powered cond. Error when using low power initialization Pressurizer mix. level not consistent with the liquid level (RETRAN-03 and -02) Choked flow failure Flow oscillation as jun void goes to zero Low power steam generator init. fails Fails with minimum time-step size 2 loop Oconee w/5-Eq fails in st-state Failure in QDOT14 Fails in the two region nonequil. model 000040 data not read during restart Different laminar flow friction transition Initial NCG states not propagated correctly CORRECTION NO. IDENT ---------*** *** *** *** *** *** *** ---*** ---*** 005 *** *** 001 *** ---------------------------------******** ******** ******** ******** ******** ******** ******** -----------******** -----------******** MOD001g ******** ******** MOD001g ******** COMMENTS Change in defaults Not a code error Model limitation

Model limitation Input error

6

Summary of RETRAN-3D Code Trouble Reports (Cont'd)
NO. 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 TROUBLE REPORT TYPE OF PROBLEM CHF calculation for a single volume Choked flow numerical instability Core dump occurs for one case not others Time-step error - Pressure is 5997 psia Results do not agree with data Anomalous downcomer level Fails with a time step error in pressurizer Steady-state convergence error Direct moderator heating error Restart incorrect transient values Steady-state does not converge Standard Problem One difference Steady-state fails after 6 iterations Edits for Volume Data Actually Used Void fraction comparisons are poor Pressure search failure for 2-phase MOC MOC does not return to the initial temp MOC does not work with noncondensables MOC soln - no null tran for 2 phase Condensation heat tran -- core dump Stdy-state fails - dyn slip used -- 135 vol Enthalpy error at steady-state iter #6 Condensation mass transfer model error Error in fill and tdv BCs for pure error Anomalous countercurrent flooding REEDIT job causes and FTB error Code error discovered on PC installation Multi-D kinetics input and ss error proc Format Error for Card 146000 Error turning enth tran off when 5-eq used Error in TRIPDT when dt < TMIN No check for TMIN or TRTG < TMAX Error in five-eqn wall heat/mass transfer Five-equation option fails to initialize Fails in subroutine DERIVS Przr model doesn't match theory Allow old TDV input when no NCG Output format error in sub. INGAS Correct EOS failure in GENOPT Eliminate call to CCFPRP in stdy-st Remove extraneous deriv. in XANDH Reactivate slip after single phase Correct phase flag in STATNE Error in Wilson Bubble Rise Model Error with restart - Pressure reported as non. Steady state failure at Iteration #6 Water properties routines don't give unique answers Separated volumes fail to converge A reedit error - FTB Error #11 CORRECTION NO. IDENT 007 013 *** *** *** *** *** *** *** *** *** *** 006 009 002 *** *** ---*** 008 035 *** 002 011 *** 014 003 004 016 018 028 004 019 024 *** *** 025 017 020 021 022 023 026 027 030 032 *** *** *** *** MOD001g MOD001g ******** ******** ******** ******** ******** ******** ******** ******** ******** ******** MOD001g MOD001g MOD001g ******** ******** -----------******** MOD001g MOD001g ******** MOD001g MOD001g ******** MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g ******** ******** MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g MOD001g ******** ********* ********* ********* COMMENTS

(partial fix)

Model limitation

(partial fix)

7

ComEd BWR Transient Analysis Licensing Topical Report
B. Tsai, Commonwealth Edison Company In June 1995, Commonwealth Edison Company (ComEd) submitted a licensing topical report that presents ComEd's BWR transient analysis methods. The report demonstrates the validity of the methods and the qualification of ComEd to perform transient analysis for reload licensing and operational support applications. This is accomplished by presenting the results of the benchmarking studies of ComEd BWR plant startup tests, Peach Bottom turbine trip tests, and the Peach Bottom NRC licensing basis transient. Related material including the core thermal limit, and reload application methodologies to qualify the reload licensing application will be provided in a separate report. ComEd's transient analysis methods are primarily based on the computer codes developed by the Electric Power Research Institute (EPRI): RETRAN-02 MOD005, ESCORE, FIBWR2 (EPRI/Scientech), and PETRA (Scandpower). The ComEd benchmarking analysis of the plant startup tests was chosen to validate the ComEd transient analysis methods for a variety of plant transients. Comparison studies of the turbine trip tests performed at Peach Bottom Unit 2 Cycle 2 demonstrate the acceptability of the ComEd transient analysis methods for more challenging pressurization events similar to the licensing basis events. An NRC licensing basis transient case of Peach Bottom turbine trip without bypass was analyzed to demonstrate ComEd method's capability of predicting system response under conditions which challenge operating limits. The comprehensive nature of the benchmarking scope and the good agreement of all the benchmarking results have fully demonstrated the capability of the ComEd transient analysis methods and the qualification of the ComEd staff to use the methods presented to perform transient analysis for reload licensing and operational support applications. We would like to acknowledge the consu

相关文章:
RETRAN程序在300MW核电厂事故分析中的建模及应用.doc
RETRAN 程序在 300MW 核电厂事故分析中的建模及应用 【摘要】本文以 300MW 核电厂为研究对象,利用 RETRAN 程序对其一、 二回路的主要设备和系统进行建模。选择...
RETRAN程序在300MW核电厂事故分析中的建模及应用_论文.pdf
RETRAN程序在300MW核电厂事故分析中的建模及应用 - ACadem IC 学术 RETRAN 程序在 300MW核 电厂 事故 分析中 的建 模及...
RETRAN-3D和RELAP5程序壁面换热模型 比较研究.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...了RETRAN-3D和RELAP5程序的特点,分析了两个程序...沸水堆核电站系统热工水力瞬态分 析系统程序[1]...
RETRAN程序在300MW核电厂事故分析中的建模及应用.doc
RETRAN 程序在 300MW 核电厂事故分析中的建模及应用 【摘要】本文以 300MW 核电厂为研究对象,利用 RETRAN 程序对其一、二回路的主要设备和系统进行建模。 选择...
RETRAN-3D程序PWR DNB模型评价研究.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...系统分析程序(如: RELAP5、 RETRAN、 THEMIS 等)...秦山核电厂各种控制棒弹棒事故的审核计算[J]. 原子...
典型压水堆核电厂一回路热力系统小破口失水事故计算分析.doc
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...对核电厂一 回路的热力系统进行建模分析,模拟小破口...在国内,自 1986 年 从美国引进 RETRAN-02 程序...
核动力船陆奥反应堆自然循环RETRAN分析.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...核动力船陆奥反应堆自然循环RETRAN分析_机械/仪表_...如果 自动 控制 系统 在自然循环模式中运行 良好 ,...
AP1000主回路系统热工水力瞬态计算程序RETAC的开发.pdf
新闻 网页 贴吧 知道 音乐 图片 视频 地图 百科...应堆系统安全分析程序RELAP、RETRAN、TRA C及AP1000...非能动安全先进核电1 林诚 格, [厂A 原子能出版...
核反应堆安全分析-4.ppt
登录注册新闻网页贴吧知道音乐图片视频地图百科文库 搜...Retran 等;子通道 分析程序: Cobra 等;燃料行为分析...《轻水堆核电厂安全分析报告标准格式和内 容》(第...
第三章 核电厂事故分析基本知识_图文.doc
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...总体上分析核电厂在失水 事故及各种瞬变过程中系统...是事故分析中最主要的程序,如 RETRAN,RELAP5,TRAC...
压水堆核电厂一回路仿真可视化研究(论文).pdf
新闻 网页 贴吧 知道 音乐 图片 视频 地图 百科...[3],美国EI公司研制的 RETRAN[43;后者主要包括...编写程序对压水 堆核电厂一回路系统进行仿真分析,且...
核电厂控制与保护系统动态仿真_图文.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...系统工程系,200030) 摘要:分析了 CHASHIMA 核电站...两程序分别为 FROTRAN 语言和 C 语言编制, 通过...
压水堆核电站热工水力系统程序的研发现状与趋势_图文.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...分析了目前世界上典型的压水堆核电站热工水力系统...应用的热工 水力系统程序还有KORSAR、RETRAN程序 等...
部分内容.doc
这种建模方法是现代反 应堆安全分析程序中广泛使用的...较为成熟的核电站全范围模拟仿真系统, 以及适用于...在SUN上作站上开发了基于RETRAN-02计算软件的仿真...
压水堆核电站运行堆芯物理过程的PC仿真_图文.pdf
新闻 网页 贴吧 知道 音乐 图片 视频 地图 百科...分析 关键 词:PWR核 电站 ;物理 数 学模 型 :...研究 ,开发了RELAP5、RETRAN、TRAC等大 型程序。我...
RETRAN-3D和RELAP5MOD4.0程序两相临 界流模型评价_图文.pdf
RETRAN-3D和RELAP5MOD4.0程序两相临 界流模型评价 - 为研究系统分析程序中临界流模型的准确性和喷放过程动态趋势,采用RETRAN-3D和RELAP5/MOD4.0 程序对高温...
一体化核动力装置的热工水力特性模型及计算.pdf
根据模型编制了程序 , 经过实堆计算及 RETRAN02 ...可在陆地上建造核电厂 , 截至 1995 年底 , 全世界...一体化核动力系统热工水力分析程序研制 [ D ] . ...
系统分析程序预测MDNBR的保守性研究.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...RETRAN-02/3D、THEMIS 等、子通道分析 程序 FLICA...以秦山核电厂事故分析为例,研 究了通过内部耦合一次...
海洋条件反应堆热工水力系统分析程序开发及验证_论文.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...海洋条件反应堆热工水力系统分析程序开发及验证_电子...分析 ;浮动核电站 RETRAN -02/G R A V 等程序...
地理信息系统在核电站事故场外后果预测评价系统中的应用.pdf
网页 新闻 贴吧 知道 音乐 图片 视频 地图 文库 |...地理信息系统核电站事故场外后果预测评价系统中的...模式程序 ( 以下统 ) 一般都用 FO R TRAN 或 ...
更多相关标签: