当前位置:首页 >> 信息与通信 >>

Condition Monitoring and Diagnosis of POWER TRANSFORMERS


2008 International Conference on Condition Monitoring and Diagnosis, Beijing, China, April 21-24, 2008

Condition Monitoring and Diagnosis of Power Transformers
E. Gockenbach, H. Borsi
Leibniz Universit?t Hannover, Schering-Institut Callinstr. 25A, 30167 Hannover, Germany *E-mail : gockenbach@si.uni-hannover.de

Abstract--Power transformers are one of the most important components of electric networks. These devices are very expensive and therefore diagnosis and monitoring systems will be valuable for preventing damage to them. A facility for viewing the status of transformers remotely by experts who will make an appropriate decision in case of a problem is needed to prevent premature damage to the transformers. The transformers are geographically spread and with the aid of Internet, it is possible to collect appropriate information from these transformers to a central node for diagnostic purposes under the supervision of high voltage engineering experts. The aim of this paper is to show the main problems related to power transformers and to review mitigation methods for the monitoring and diagnosis of power transformers. Index Terms--High voltage technique, Monitoring, Power transformer insulation Insulation life,

I. INTRODUCTION

P

ower transformers are important and expensive equipment in electric energy networks. The majority of these devices have been in service for many years under different environmental, electrical, and mechanical conditions. In addition to the costs associated with equipment repair or replacement, the capital loss of an accidental power transformer outage is often counted in million dollars for output loss only [1]. Because of this economic motivation, online monitoring and diagnosis (M&D) systems are of benefit to predict fault conditions and maintenance of the high voltage transformers.

The selection of the monitoring functions is determined mainly by two goals, the faults must be promptly recognized, so that the operator can avoid critical conditions and the maintenance work is carried out only, if the condition of the plant requires it. The processing of the measured values represents a further aspect. Monitoring is a component of the service concept of the manufacturer. As direct expectation to a diagnostic system, the following terms are important: x extension of the remaining useful life of the transformer x improvement of loading possibility of the transformer x higher availability and service reliability x condition-based maintenance and repair x prevention of loss and destruction Diagnosis contains interpretation of data to determine the current condition of a transformer. The diagnostic task has an important influence on the overall maintenance cost as well as on reliability and availability. The use of advanced technologies has the potential to greatly reduce the time and increase the accuracy of transformer diagnostics. There are many different techniques for diagnostic purposes, such as: Expert Systems, Case-Based Reasoning (CBR), Model-Based Reasoning (MBR), Artificial Neural Network (ANN), Fuzzy Logic, Knowledge-Based Systems and Genetic Algorithm. All the technologies have their advantages and disadvantages. In most cases they cannot work alone to solve the diagnostic problems and have to complement each other to form an integrated solution. Figure 1 shows the block diagram of M&D system for power transformers.

Data Acquisition Feature Extraction
PD, FRA, Bushing

Data Mining
Clustering Discrimination Information Fusion

Diagnosis
ANN Fuzzy CBR MBR

Pre-processing

Voltages, Currents Dissolved Gas in Oil

Online & Offline Sensors

Fig. 1. Block diagram of monitoring and diagnosis system for power transformers

978-1-4244-1622-6/08/$25.00 ?2007 IEEE

II. ACTUAL SITUATION A transformer consists of several independently working components. These components are windings and cores as electric and magnetic active parts in which voltage is induced and magnetic flux is guided. Additionally several bushings, insulating oil and tank, tap-changers and coolers are required. A survey of the age structure of transformers from a large German utility clearly shows in Figure 2 that the transformer majority are approaching or exceeding the age of 25-35 years [2]. This situation is more or less similar in many industrialized countries [3].
80 70 60

thermal chemical stresses Statistics [1] show that the most frequent causes followed by long outage damages are in tap changer, active component and in bushings, according table I.
TABLE I FAILURE CAUSES OF POWER TRANSFORMERS WITH DOWN-TIMES MORE THAN 1 DAY.

x x

Tap changer coil + core bushings vessel accessories

40 % 35 % 14 % 6% 5%

50 40 30 20 10 0

A. Electrical Causes Inside the transformers, insulation systems (made of insulating oil and cellulose), can be locally overstressed, whereby the aging process of the insulating system will be accelerated, leading to acid generation and pollution in the oil, an increase in the water content, gas and mud formation. As a consequence, the breakdown voltage will decrease. The increase in water can lead to an acceleration of the depolymerisation of the cellulose. Paper aging process, depends primarily on temperature and water content 5. The existence of oxygen will also remarkably increase the rate by as much as 2.5 to 10 times normal. Fig. 4 depicts the relation between relative depolymerisation velocity and the water content and temperature.
service age in years

number of transformers

1000 relative depolymerisation velocity 80 °C 100 °C 100 120 °C

Fig. 2. Age structure of transformers from a large German utility [2]

Regarding the information of Fig. 2 the main point in monitoring and diagnosis is the increase of the failure rate with increasing years of service. Fig. 4 shows the typically curve whereby a punctual revision causes a shift of the failure rate of approximately 5 years [4].
60 50 40 failure rate in % 30 20 10 0

10

1

0

1

2 water content in paper

3

[%]

4

Fig. 4. Relative depolymerisation velocity at different water content and temperature

1

3

5

7

9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 service years

Fig. 3. Example of failure rate as a function of transformer age [4]

III. FAILURE CAUSES The premature and unexpected failures in transformers can be caused through the following stresses: x electrical x electromagnetic x dielectric

B. Electromagnetic Causes High forces acting on the coil, which can lead to a deformation, can be caused e.g. by a short-circuit in the network. These deformations can lead to the insulation paper breaking in particular with aged papers. The consequence is first the generation of partial discharge and finally the total breakdown. C. Dielectric Causes The quality of the paper insulation in a transformer depends on the polymerization degree (number of the glucose rings and chains) of the paper. During aging process, the chains are up-smelled consisting of glucose rings and develop

water, gases (CO, CO2), groups of carboxyl’s (organic acids). The ageing speed depends thereby on different parameters such as temperature, water content, oxygen content, number and kind of temperature cycles and material properties. D. Thermal Causes As a thermal cause, the losses result from the demagnetization of the core (eddy current losses) and the Ohm's resistance of the coil inevitably. Heating up and cooling processes of the insulation with high temperature gradient, represent additional load of the insulation system which influences the life span of the transformer insulation. E. Chemical Causes Organic acids are developed as aging by-products of the solid and liquid insulation, which attack in particular the paper insulation and accelerates the aging. In addition metals such as copper, iron, aluminum and zinc which exist in each transformer are catalytic, and have an additional accelerating effect on the aging processes of the insulation. IV. TOOLS FOR MONITORING AND DIAGNOSIS Various tools and methods for monitoring and diagnosis of high voltage power transformers are actually available [2]. Basically, they can be divided into: x traditional diagnostic methods that have been used for many years x new methods that range from methods that are starting to be used and x methods that are still at the research stage Some examples of traditional methods are: Dissolved Gas Analysis (DGA) [5], Insulating oil quality, Power factor testing [6], winding resistance, and thermograph. New methods include online PD testing [7], recovery voltage measurement [8], tap changer monitoring, internal temperature measurement, on-line power factor measurement, dielectric spectroscopy, and winding deformation detection. Some of these mentioned methods are implemented using software systems, which gives more definite indications of transformer problems than conventional analysis. The use of software can improve the reliability and provide facilities to analyze the test data. It can also be used to extract information and knowledge that is not available and not visible from the data directly using advanced information technology methods such as Data Mining. The advancement in artificial intelligent (AI) modeling techniques has enabled power engineers and researchers to develop useful artificial intelligent software for diagnosing transformer faults. Artificial neural network or ANN approaches are used for DGA method. Fuzzy logic concept is another AI technique in power system associated with the uncertainty of changing power operational condition, numerous power system configurations, imprecise information input by human operators, disturbances and faults. Expert systems have been proposed to manage knowledge processing. An expert system is a computer program that performs a complex decision making task within a particular

narrow problem domain that is normally done by a human expert. It is based on the idea of taking the knowledge from a specialist and expressing it in an appropriate representation to exploit the knowledge in the same way as the human expert does and above all with the same result. The use of expert systems for transformer diagnosis offers the potential of reducing the overhead required by substations for the maintenance of transformers. V. MONITORING PROCEDURES In contrast to laboratory measurements, on-site measurement is disturbed by extreme and mostly influenceable conditions. In particular, within the range of the high voltage areas, within which the transformers work, precise measurements are more difficult because of the interference of electrical and magnetic fields. On-line procedures make it possible to detect failures at any time and also during the normal operation of the transformer. Furthermore on-line measurements offer the advantage that measuring data can be pursued during a longer period with almost same operating conditions. Thus, slow changes can be detected and a warning message or command to immediate disconnection can take place, if given limit values are crossed [9]. A. Chemical Procedures With the help of chemical procedures, some failures, in the transformer insulation can be determined. One of the standard investigations is the so called “gas in oil analysis (DGA)”, with which a sample of the insulating liquid is taken from the transformer and examined. This sample, after a vacuum extraction undergoes an analysis by gas chromatography to reveal the dissolved gases. The interpretation of different quotients of low-molecular hydrocarbon connections serves thereby to the determination of the failure. As an evaluation criterion, in particular, the triangular method after Duval as well as the MSS procedure after Mueller, Schliesing and Soldner are used. The disadvantage of these procedures is the fact that they allow only one integral evaluation to the insulation. In addition, sampling and the following treatment of the sample can affect the measurement. Further procedures are the furan analysis (HPLC) and the analysis of free gases collected in Buchholz relay. While with the furan analysis, vital information on the quality of the solid insulation is obtained, the analysis of the collected free gases in Buchholz relay provides information about the range of an available failure. The Buchholz relay indicates only gas amounts developed since the last exhaust, however not the history of the gas emergence. So long lasting, but nevertheless low energy failures e.g. PD lead to a continuous gas production, while failures with high energy content like local overheating generate high gas rates within a short period. In order to make a better evaluation of the available failure, it is meaningfully to determine the gas rate. This statement makes the electronic Buchholz relay possible, which is to serve as an extension for the Buchholz relay, without limiting its function. This system determines thereby the history of the

gas emergence, which, during simultaneous recording of the operating conditions, permits additional conclusions on an available failure. B. Electrical Procedures While the chemical give a cumulative statement about the period passed since the last analysis, the electrical procedures allow a statement about the current condition of the transformer. The main electrical procedures are the partial discharge (PD) measurement) and off-line the transfer function measurement. Partial discharges are in most cases determined chemically during gas in oil analysis, since they show up typically and strongly increased hydrogen content. A continuous measurement of PD during the operation is not used at present yet, since the narrow-band measurement of the partial discharge signals on site is expenditure and in addition, no localization of failure is possible. A diagnostic system was therefore developed which permits from the wide-band measurement of partial discharge signals a determination of the PD source and a determination of the apparent charge converted at the defective equipment. The diagnostic procedure is based on the evaluation of the signal deformation of PD pulses within the transformer by mathematical algorithms and permits by the determination of PD location as well as the charge quantity a qualified analysis of the failure. For this procedure, it is necessary to note the high frequency partial discharge signals both at the bushing and at the neutral point. In these data both the current pulse caused by the PD and the deformation, experienced by this impulse during its transmission trough the turns of the transformer are contained. If this deformation is known, then it can be reckoned back from the ends of the windings of measured signals on different original places. These computations are accomplished both from the high voltage and the neutral point side. The place where the signals calculated for the different sections of the windings are identical is the true location of the PD source. A further advantage of this procedure is that the calculated signal corresponds to the actual partial discharge signal. It is thus possible to estimate the charge contents of the PD at its origin. Defects can also be recognized by the measurement of the transfer function of the individual transformer coils such as turn short-circuits or deformations. The current measurement is compared with a "finger print", which must be determined first as reference. Deviations between the reference transfer function and the current transfer function indicates then a failure. Whereby, with the bandwidth of the transfer function measurement, the sensitivity of the method can be increased. Such transfer function measurements can be accomplished on and offline, whereby with offline measurements, the impulse response can be determined by different procedures and with online measurement, a switching impulse can be used. C. Dielectric Procedures The dielectric methods are used as offline procedure and permit a more exact view of the insulating system, if an

extended statement about their condition is required. Three procedures are used and well described in the literature: x Recovery Voltage Measurement (RVM) x Polarization and Depolarization Current (PDC) x Frequency Dielectric Response (FDS) D. Further Procedures The overload capacity of a transformer is limited among other things by reaching the permissible maximum temperatures of coil and oil. For the determination of the hot spot temperature different procedures can be used using a thermal model of the transformer and some temperature measurements on accessible points of the transformer. VI. CONCLUSIONS x Monitoring and diagnosis systems for power transformers are available. x The failure sources are well known and also the parameter which strongly influence the ageing of the insulation system. x The procedures for monitoring the various effects are mainly based on chemical, electrical and dielectric measurements and can be recorded. x The crucial point, which needs more statistical data, is the evaluation of the failure risk and the estimation of the residual life time. But this will be improved in future by using more and more monitoring and diagnosis systems on power transformers and by increasing the statistical data basis. VII. REFERENCES
[1] [2] CIGRE-WG 12-05: ?An international survey on failures in large power transformers in service“, Electra No.88 (1983), pp. 21-48. M. Stach,: ?Betriebswirtschaftliche Gesichtspunkte im AssetManagement im Zeitalter der Fusion“, Micafil Symposium, 2021.M?rz 2002, Stuttgart. M. Wang, A. J. Vandermaar, “Review of condition assessment of power transformers in service”, IEEE Electrical Insulation Magazine, vol. 18, No. 6, pp. 12-25, Nov/Dec 2002. Bengtsson, C., Persson, J-O., Svenson, M.: ?Replacement and Refurbishment Strategies for Transformer Populations,“, CIGRE 2001 Colloquium, Dublin SC 12.20. “Guide for the Sampling of Gases and Oil from Oil-Filled Electrical Equipment and for the Analysis of Free and Dissolved Gases,” IEC Publication 60567, 1992. T. Noonan, “Power transformer on-site condition assessment testing,” presented at the International Council on Large Electric Systems (CIGR?), Paris, France, 2000. Asghar Akbari, et al.,Transfer Function-Based Partial Discharge Localization in Power Transformers: A Feasibility Study, EEE Electrical Insulation Magazine, 18, 2002. R.S. Brooks , G.S. Urbani, “Using the recovery voltage method to evaluate aging in oil-paper insulation,” in Proc. IEEE Int. Conf. Conduction and Breakdown in Solid Dielectrics., Vasteras, Sweden, pp. 93-97, 1998. Han, Y.; Song, Y.H., “Condition monitoring techniques for electrical equipment-a literature survey”, IEEE Transactions on Power Delivery, Volume: 18, Issue: 1, Pages:4 - 13, Jan 2003. I. Fofana, H. Borsi, E. Gockenbach and M. Farzaneh “Aging of transformer insulating materials under selective conditions”, European Transactions on Electric Power, 2006; 16:1–21

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]


相关文章:
Condition Monitoring and Diagnosis of POWER TRANSFORMERS.pdf
International Conference on Condition Monitoring and Diagnosis, Beijing, China, April 21-24, 2008 Condition Monitoring and Diagnosis of Power Transformers E. ...
A-62. 频率响应法在线监测变压器绕组变形.pdf
Monitoring and Diagnosis, Beijing, China, A.W....[14] S. Birlasekaran, FRA Fred Condition F...winding displacements in power transformers ...
外文翻译---基于单片机的油浸式变压器温度监测系统的设计.doc
condition monitoring and fault diagnosing of power ...immersed power transformers is easy to change ...monitoring and fault diagnosis of power transformer...
...Based Condition Monitoring System for Power transformers (....pdf
The developed system has became useful platform for an on-going project on the condition monitoring and fault diagnosis for large power transformers. ...
大型变压器在线监测技术及其发展.pdf
condition monitoring and fault diagnosis for large - sized transformers based ...fiber - op tic temperature sensor system for Power transformer monitoring EEE...
变压器状态监测_图文.pdf
978-1-4244-1622-6/08/$25.00 ?2007 IEEE I Power transformers are among... condition-monitoring and diagnosis of the equipment are essential to ...
Improved BP Neural Network for Transformer Fault Diagnosis_....pdf
Improved BP Neural Network for Transformer Fault Diagnosis_能源/化工_工程科技...Nuclear power plant fa... 42人阅读 6页 2.00 Fault Diagnosis Based ...
A New Diagnostic Test for Power Cables Based on_图文.pdf
fault diagnosis methods are increasingly developed. ...real time condition monitoring and trend prediction...of power transformers is very common [11-17]. ...
power transformer som.pdf
Diagnosis of Power Transformers Using Modified Self Organizing Map J. P. ...Development of condition monitoring and diagnosis techniques has been very ...
电力变压器的协同训练故障诊断算法.pdf
TrainingforFaultDiagnosisofPowerTransformersUUJunl,一,LILi3,LIUDao.qun2,4(1?DepartmentofEngineeringScience,ChongqingRadio&TelevisionUniversity,Chongqing400052,China...
英文_图文.pdf
2008 International Conference on Condition Monitoring and Diagnosis, Beijing, ...The insulation system of power transformers consists of oil and cellulose (...
Koltunowicz be submitted_final_图文.pdf
Challenges in continuous monitoring of transformer insulation 变压器绝缘持续监测...Tenbohlen et al.: “Enhanced Diagnosis of Power Transformers using On- and...
Fault Diagnosis of Power Transformer Based on Heuristic ....pdf
Fault Diagnosis of Power Transformer Based on Heuristic Reduction Algorithm_...D. Srivastava, “Review of condition assessment of power transformers in ...
模糊专家系统在电力变压器在线故障诊断中的应用初探.pdf
(河海大学电气工程学院电气工程系 A PRIMⅡIVE STUDYONONLINEFAULTDIAGNOSISOFPOWERONFUZZYEXPERT TRANSFORMERSBASEDSu SYSTEM Wenhui Ju Ping (HohaiUniversity,...
基于BP神经网络的电力变压器故障诊断的研究_86_90.pdf
and emit heats in large power transformers. ...Under the condition of normal running and early ...(3) The Transformer Fault Diagnosis Model Based ...
Fault Diagnosis On Power Transformers Using Non-electric ....pdf
Fault Diagnosis On Power Transformers Using Non-electric Method_电力/水利_...Power systems fault di... 6页 免费 Condition Monitoring a... 4页 2...
Oil-immersed Power Transformers condition analysis.pdf
Oil-immersed Power Transformers condition analysis_理学_高等教育_教育专区。transformer Oil-immersed Power Transformers Condition Diagnosis with Limited Dissolved ...
2009-02-25-ON-LINE_PARTIAL_DISCHARGE_MONITORING_AND_DIAGNOSIS....pdf
_MONITORING_AND_DIAGNOSIS_AT_POWER_CABLES_能源/...The condition-based maintenance of power cables ...power transformers connected to the power cable ...
基于BP网络算法优化模糊Petri网的电力变压器故障诊断.pdf
关键词:变压器;故障诊断;油中溶解气体分析;BP 网络;模糊 Petri 网 Fault diagnosis of power transformers based on back propagation algorithm evolving fuzzy Petri...
高压断路器故障诊断.pdf
condition monitoring and diagnosis in order to ...voltage or current transformers and limit switches....Computing the diagnosis requires high computer power...