A New and Comprehensive Software for Condition Assessment and LifeTime Management of Power Transformers

Transcription

1 No. E-13-ITP-1815 A New and Comprehensive Software for Condition Assessment and LifeTime Management of Power Transformers M. Goodarzi, P. Khazaee, M. Rahmani Power Transmission & Distribution Research Center Niroo Research Institute (NRI) Tehran, Iran mgoodarzi@nri.ac.ir, pkhazaee@nri.ac.ir I. Sabetghadam, P. Karimifard Power Research Office Tavanir Tehran, Iran issasabet@yahoo.com, p.karimifard@tavanir.org.ir Abstract Power transformer is one of the most important equipments of the transmission networks. It is essential to have good information of this equipment's condition in order to improve the operation and increase the reliability of the network. Since power network has too many transformers and it is too complicated and time-consuming to analyze all of these equipments, it seems that it is necessary to have a software to carry out this duty; this software should be able to not only save and organize the information, but also to assess the condition of the transformers. TransDiag-1 utilizes fourth generation of programming languages. Creating the classified databases, analyzing the condition assessment tests' results based on standards and estimating the remaining lifetime of the transformers are some features of TransDiag-1 software. In this software, transformers with critical condition can be detected by formation of the risk assessment curve. In this paper, computational algorithms and operational procedures of the software have been introduced and studied. Finally, information of a sample transformer have been used as a study-case to test this software. Keywords software developing; power transformer; condition assessment; prioritization; lifetime management I. INTRODUCTION Software development, that is, the development of a software product, may include research, new development, prototyping, modification, reuse, re-engineering, maintenance, or any other activities that result in software products [1]. Software can be developed for a variety of purposes, the three most common being to meet specific needs of a specific client/business, to meet a perceived need of some set of potential users, or for personal use. Embedded software development, that is, the development of embedded software such as used for controlling consumer products, requires the development process to be integrated with the development of the controlled physical product [2]. Many methods such as waterfall, spiral, agile and etc are available to develop a software; each one of these methods has many weaknesses and strengths. TransDiag-1 software which is designed for condition assessment and lifetime management, has been developed through a combination of these methods. In this software, it is easy to develop and/or modify the probable errors. TransDiag-1 software has many features such as 1) creating a database, 2) analyzing the condition assessment tests' results based on credible international standards, 3) estimating the remaining lifetime of power transformers, 4) prioritizing power transformer in order to detect the transformers with critical condition, and so on. The results of the transformers' condition assessment tests will be classified and saved in this software; it is also possible to import and/or export these data from the software. One of the most important features of any software is to be user friendly which has been thoroughly considered in TransDiag- 1. A help tab is designed for any section of the software and gives user some information about the modality of the test and computational algorithms. In this paper, computational algorithms which are used in different sections of the software have been studied. Data of a sample transformer has been used as a study-case to see how the software operates. II. SOFTWARE STRUCTURE AND COMUTATIONAL ALGORITHMS TransDiag-1 has been prepared for condition assessment and lifetime management of power transformers. Fig. 1 shows a general view of this software. Fig. 1. general view of TransDiag-1 software

2 As it is seen in the configuration, this software consists of two major parts which are "input data" and "results"; each part includes some minor parts. Results consist of five sections: Importance Index, Health Index, Prioritization, Condition Assessment and Remaining Lifetime Estimation. All of these features are explained completely in this section. Also, computational algorithms of the software have been described thoroughly in the following. A. Input Data Like any other software, this software needs some information as input data which should be inputted by user. The input data in TransDiag-1 software consists of two parts: 1) General information: data such as serial number, utility and etc should be selected by the user just to specify the transformer type; but it is not necessary to input these data and it does not lead to any error if the user does not input these data. 2) Input Data: these data consist of information about condition assessment tests, loading history and maintenance history. Software processes these data and then, the user can select between different parts of the output section to be shown. Output section consists of 5 catagories. B. Importance Index Importance index is defined and calculated based on the economical features of the transformer. In order to calculate the importance index, transformer present worth value, annual cost and equivalent uniform annual cost (EUAC) should be known. To calculate the annual cost, bathtub curve has been utilized which is used in many papers as a good approach to model the failure rate [3]. Present value of the transformer has been calculated through depreciation method [4]. After calculating the annual depreciation, the present worth value of the transformer can be obtained through (1) as follows [5]: n ( ) ( ) PW n C d j = (1) j = 1 Where PW is the present worth value of the transformer, C is the capital cost of transformer and d is the annual depreciation. Future worth-value of the transformer which can be calculated by subtracting the transformer expenditures from the present worth value is formulated in (2): ( 1 ) m m = + m m m j j = FV C i PW 1( Co j + Crj + Cp j ) ( 1+ i ) Where Co j and Cr j are operational annual cost and repair annual cost respectively and i represents interest rate. After calculating the future worth value of transformer, EUAC is obtained from (3); then by normalizing the EUAC, the importance index is calculated. (2) EUAC i = FV (3) ( i ) m m m It can be seen that the process of calculating the importance index is too complicated and needs so much time; so, it seems that it is necessary to utilize a computational software. In the TransDiag-1 software, user can see the output of these steps in a fraction of a second by inputting the data and clicking on the Importance Index button. In this software, importance index has been used to prioritize a set of transformers to detect the transformers with critical condition. C. Health Index According to importance of the power transformers' health and complicated calculating process of the health index, it is necessary to benefit from a computational software; so, one of the output parts of the software has been allocated to determination of the health index. In this software by using information from condition assessment tests, the health subindex is calculated for any of the tests [6]. Then, the overall health index (which is between 1 and 100) is calculated through (4). It is obvious that the more the health index, the better the transformer condition K 1 ihifi K 4 ihif i = i = i HI=40% + 60% K 4K i = 1 i i = 4 Where tests i=1 to i=3, are related to tap-changer and other tests are related to transformer itself. Since about 40% of the transformer failures are related to the tap-changer [7], a coefficient equal to 0.4 has been considered for the tests which are related to tap-changer. D. Prioritization It is not feasible to carry out the condition assessment for all of the power transformers in the network due to the large number of them and the high cost of this work; so, it is better to detect the transformers with critical condition. Since it is too complicated to prioritize the transformers, it is essential to have a software to carry out the prioritization. In TransDiag- 1 software, transformers can be prioritized through three different methods: based on health index, based on importance index and based on risk assessment curve. Prioritization based on health index is not appropriate because of neglecting the importance of the transformers. Also, Prioritization based on importance index is not suitable because of neglecting the health condition of the transformers. So, by considering both health and important indices, a curve is formed which is called risk assessment curve. In this curve, the most critical point is assigned to a transformer with an importance index equal to 100 and a health index equal to zero. The distance from this reference point has been considered as a criterion for prioritization the transformers. Risk assessment curve is shown in Fig.2. i (4) 2

3 Importance Index In order to calculate the degree of polymerization (DP), 4 methods have been used. In these methods, the value of DP is calculated by using the results from chemical tests. The formulations of calculating the DP through Burton, Vuarchex, Chendong and Depablo are shown in (5) to (8) respectively: 2.5 log[ f ] DP = (5) log[ f ] DP = (6) Fig. 2. risk assessment curve Refractive index DP 1.51 log[ f ] = (7) This prioritization is independent. Also TransDiag-1 software can compare a sample transformer with another one. In the proportional prioritization, the reference point (the most critical point) has the most importance index and the least health index which exists in transformers' data. E. Condition Assessment One of the steps of assessing the power transformers condition is to analyze the results of condition assessment tests. Due to vast variety of condition assessment tests and difficulty of analyzing some of these tests such as DGA (Dissolved Gas Analysis), utilizing a software to analyze the tests' results and to know how to resolve some of these probable failures is necessary; so, one of the output parts of the TransDiag-1 software is allocated to condition assessment. User can select the desired test and see the tests interpretation. In the Condition Assessment section of the TransDiag-1 software, calculation, interpretations and procedures to resolve the failures for common tests have been done based on IEEE and IEC standards [8]-[12]. User can obtain a good assessment of desired transformer condition based on credible standards. F. Remaining Lifetime Estimation Estimating the remaining lifetime of power transformers is one of the concerns of the equipment owners. This concern becomes more important when a transformer is under continual or periodic overload condition. Factors such as electrical tensions, mechanical tensions and/or thermal ones can increase the aging and lead to destruction of insulation system. In TransDiag-1, different methods have been used to estimate the transformer lifetime based on insulation system condition. Three methods have been introduced in this software for estimating the transformers' lifetime: 1) Monte Carlo method 2) Thermal-Electrical Tension method 3) Neural Network method 1850 DP = f III. CASE STUDY In order to represent the operation of the TransDiag-1 software, a sample transformer data have been considered as input data. Fig. 3 shows the general information which should be inputted into the software Input data section which includes input data related to condition assessment tests, repairing history, loading history and economical information. This section consists of 22 parts. For example, inputted data for DGA test can be observed in Fig. 4. Fig. 3. general information which should be inputted into the software. (8) 3

4 Clicking on the Condition Assessment button opens a window which is shown in Fig. 6. User selects the desired test between different electrical, mechanical, chemical or any other type of tests. For example, the interpretation of the DGA's results based on the inputted information which are presented in Fig. 4 is like what is shown in Fig. 7. Software analyzes the inputted data using Duval triangle method presented in IEC standard and represents the results. Hence, user can choose the analyzing procedure between different methods from IEC and IEEE standards. Based on the calculations and results, software suggests user to use Duval4 and Duval5 methods too; if user decide to use these methods, Fig. 8 will be shown. Fig. 6. Condition Assessment Fig. 4. Input data for DGA test After importing the data, user can see the results by clicking on any part of the Result section. If user wants to know about the transformer health condition, software shows Fig. 5 as output. As it is seen, health index value is equal to It should be noted that in HI section, if user doesn't import the information related to a specific test, software calculates the HI considering 'good condition' for this test. Also, in the HI window, it is possible to modify the data, create a new file and surveying the save data. Fig. 7. Duval method for interpretation of DGA test Fig. 5. Health Index in TransDiag-1 Fig. 8. Duval 4 and Duval 5 4

5 To see how the software prioritizes the transformers, health and importance indices of 60 transformers have been imported to the software (Fig. 9). User can see the prioritization of the transformers by clicking the Show Result button. Fig. 10 shows the prioritization. If user wants to see results proportional to each other, he/she can click on the "Show Result 2" button; in this case, Fig. 11 is shown. As it is seen in this figure, the most critical point of the curve has the maximum importance index (related to transformer number 3) and the minimum health index (related to transformer number 50). Importance Index i=30 i=31 i=12 i= Fig. 11. Proportional prioritization Refractive index Fig. 9. health and importance indices of 60 transformers calculated by software User can see the remaining lifetime of the transformer and best replacement time by clicking on the "Remaining Lifetime Estimation" button. For example, if user imports the information presented in Fig. 12 to the TransDiag-1 software, by selecting the Monte Carlo and Burton method, output results are like Fig. 13. TransDiag-1 software has estimated that the remaining lifetime of the transformer is equal to 21 years, and the best time to replace the transformer is 6 years i=12 i=3 Importance Index Refractive index Fig. 10. Independent prioritization Fig. 12. Input data for Remaining Lifetime Estimation section 5

6 the software have been demonstrated through figures and the information of sample transformer have been used to prove the validity of this software's features. Fig. 13. Remaining Lifetime Estimation IV. CONCLUSION In this paper, TransDiag-1 software has been introduced. This software is designed for power transformer condition assessment and lifetime management, because this equipment is one the most important devices in the transmission network. Having a good assessment of the network's transformers is necessary to increase network's reliability. This assessment needs too much time because of the large number of the transformers; so, this software has been prepared to carry out this assessment. TransDiag-1 allows user to create classified databases, analyze the condition assessment tests' results, estimate the remaining lifetime of the transformers and etc. Also, transformers with critical condition can be detected using risk assessment curve. Computational algorithms and operational procedures of the software have been represented. Finally, all the features of REFERENCES [1] DRM Associates (2002). "New Product Development Glossary". Retrieved [2] Richard H. Thayer, Barry W. Boehm (1986). Tutorial: software engineering project management. Computer Society Press of the IEEE. p.130 [3] English, J.R.; Li Yan; Landers, T.L., "A modified bathtub curve with latent failures," Reliability and Maintainability Symposium, [4] Carneiro, J. C.; Jardini, J.A.; Brittes, J. L P, "Substation power transformer risk management: Reflecting On reliability centered maintenance and monitoring," Transmission and Distribution: Latin America Conference and Exposition (T&D-LA), 2012 Sixth IEEE/PES, vol., no., pp.1,8, 3-5 Sept [5] Abu-Elanien, A.E.B.; Salama, M. M A, "Survey on the Transformer Condition Monitoring," Power Engineering, 2007 Large Engineering Systems Conference on, vol., no., pp.187,191, Oct [6] Jahromi A., Piercy R., Cress S., Service J. and Fan W., "An approach to power transformer asset management using health index," IEEE Electrical Insulation Magazine, vol. 25, pp , [7] CIGRE Working Group 05, An international survey of failures in large power transformers in service, Electra, no. 88, pp , May 1983 [8] Guide for the interpretation of gases generated in oil-immersed transformers. IEEE std C [9] Mineral Oil-Impregnated Electrical Equipment in Service - Guide to the Interpretation of Dissolved and Free Gases Analysis. IEC Publication ( ). [10] Guide for diagnostic filed testing of electric power apparatus-part1: oil filled Power transformers, regulators, and reactors. IEEE std [11] Guide for acceptance and maintenance of insulating oil in equipment. IEEE std C [12] Mineral insulating oils in electrical equipment-supervision and maintenance guide. IEC Publication (2005). 6