Abstract
The prediction of the dissolved gases content in an oil-immersed power transformer is very important for early fault detection. However, it is quite difficult to obtain accurate predictions due to the non-linearity of gas data. Different machine learning technics have been used to solve this problem, but they neither consider the relationship of different gases nor the sampling errors. In this paper, we propose to use Grey relational analysis (GRA) to calculate grey relational coefficients for gas feature selection and a Gaussian process regression (GPR) to predict dissolved gas value. In this method, both the relationship of gas features and sampling errors are considered. Four algorithms of ANN, SVM, LSSVM and GPR are used in gas prediction. We conducted experiments on eight dissolved gas datasets. The comparison results have shown that the GRA method is effective in selecting good gas features. The performance of prediction of gas values is significantly improved.
Similar content being viewed by others
References
(2009) IEEE guide for the interpretation of gases generated in oil-immersed transformers, pp 1–36
Rogers RR (1978) IEEE and IEC codes to interpret incipient faults in transformers, using gas in oil analysis. IEEE Trans Electr Insul 13(5):349–354
Mirowski P, LeCun Y (2012) Statistical Machine Learning and Dissolved Gas Analysis: A Review. IEEE Trans Power Deliv 27(4):1791–1799
Lin CE, Ling JM, Huang CL (1993) An expert ssystem for transformer fault diagnosis using dissolved gas analysis. IEEE Trans Power Deliv 8(1):231–238
Su Q, Lai LL, Austin P (2000) A fuzzy dissolved gas analysis method for the diagnosis of multiple incipient faults in a transformer. Int Conf Adv Power Syst Control Oper Manag 2:343–348
Richardson ZJ, Fitch J, Tang WH, Goulermas JY, Wu QH (2008) A probabilistic classifier for transformer dissolved gas analysis with a particle swarm optimizer. IEEE Trans Power Deliv 23(2):751–759
Dong LX, Xiao DM, Liang YS, Liu YL (2008) Rough set and fuzzy wavelet neural network integrated with least square weighted fusion algorithm based fault diagnosis research for power transformers. Electr Power Syst Res 78(1):129–136
Yang Z, Tang WH, Shintemirov A, Wu QH (2009) Association rule mining-based dissolved gas analysis for fault diagnosis of power transformers. IEEE Trans Syst Man Cybern Part C 39(6):597–610
Li, J., Zhang Q, Wang K, Wang J, Zhou T, Zhang Y (2016) Optimal dissolved gas ratios selected by genetic algorithm for power transformer fault diagnosis based on support vector machine. IEEE Tran Dielectr Electr Insul 23(2):1198–1206
Shintemirov A, Tang WH, Wu QH (2009) Power transformer fault classification based on dissolved gas analysis by implementing bootstrap and genetic programming. IEEE Trans Syst Man Cybern Part C 39(1):69–79
Ghoneim SSM, Taha IBM, Elkalashy NI (2016) Integrated ANN-based proactive fault diagnostic scheme for power transformers using dissolved gas analysis. IEEE Trans Dielectr Electr Insul 23(3):1838–1845
Samaher AJ, Sarvesh R, Ahmed P, Ibrahim AS (2015) Design and evaluation of a hybrid system for detection and prediction of faults in electrical transformers. Int J Electr Power Energy Syst 67:324–335
Wang MH (2003) A novel extension method for transformer fault diagnosis. IEEE Trans Power Deliv 18(1):164–169
Wang MH, Hung CP (2003) Novel grey model for the prediction of trend of dissolved gases in oil-filled power apparatus. Electr Power Syst Res 67(1):53–58
Wang MH (2004) Grey-extension method for incipient fault forecasting of oilimmersed power transformer. Electr Power Comp Syst 32(10):959–975
Fei SW, Sun Y (2008) Forecasting dissolved gases content in power transformer oil based on support vector machine with genetic algorithm. Electr Power Syst Res 78(3):507–514
Liao RJ, Zheng HB, Grzybowski S, Yang LJ, Tang C, Zhang YY (2011) Fuzzy information granulated particle swarm optimisation-support vector machine regression for the trend forecasting of dissolved gases in oil-filled transformers. IET Electr Power Appl 5(2):230–237
Liao RJ, Zheng HB, Grzybowski S, Yang LJ (2011) Particle swarm optimization-least squares support vector regression based forecasting model on dissolved gases in oil-filled power transformers. Electr Power Syst Res 81(12):2074–2080
Liao RJ, Bian JP, Yang LJ, Grzybowski S, Wang YY, Li J (2012) Forecasting dissolved gases content in power transformer oil based on weakening buffer operator and least square support vector machine—Markov. IET Gener Transm Distr 6(2):142–151
He YL, Wang XZ, Huang JZX (2016) Fuzzy nonlinear regression analysis using a random weight network. Inf Sci 364–365:222–240
He YL, Liu JNK, Hu YH, Wang XZ (2015) OWA operator based link prediction ensemble for social network. Expert Syst Appl 42(1):21–50
Wei CH, Tang WH, Wu QH (2014) Dissolved gas analysis method based on novel feature prioritisation and support vector machine. IET Electr Power Appl 8(8):320–328
Wang XZ, He YL, Wang DD (2014) Non-naive Bayesian classifiers for classification problems with continuous attributes. IEEE Trans Cybern 44(1):21–39
Wang XZ, Ha MH (1992) Fuzzy linear regression analysis. Fuzzy Sets Syst 51(2):179–188
Chen WG, Long ZZ (2013) Experimental research on air-gap partial discharge properties in transformer oil-paper insulation. IEEE conference on electrical insulation and dielectric phenomena, vol 9, pp 1274–1277
Liu S, Forrest J, Yang Y (2011) A brief introduction to grey systems theory. IEEE international conference on grey systems and intelligent services, vol 2, pp 2403–2408
Yiyo K, Taho Y, Guan WH (2008) The use of grey relational analysis in solving multiple attribute decision-making problems. Comput Ind Eng 55(1):80–93
Truongm DQ, Ahn KK, Trung NT (2013) Design of an advanced time delay measurement and a smart adaptive unequal interval grey predictor for real-time nonlinear control systems. IEEE Trans Ind Electron 60(10):4574–4589
Brasel E (2000) A new concept for on-line transformer gas monitoring. CIGRE Session, Paris
Tenbohlen S, Schafer M, Wang Z, Atanasova I (2008) Investigation on sampling, measurement and interpretation of gas-in-oil analysis for power transformers. CIGRE Session, Paris, pp A2–A102
Williams CKI, Rasmussen CE (1996) Gaussian processes for regression. Adv Neural Inf Process Sys 8(6):514–520
MacKay DJC (1998) Introduction to Gaussian Processes. Neural Netw Mach Learn 168:133–165
Rasmussen CE, Williams CKI (2006) Gaussian processes for machine learning. MIT Press, Cambridge
Li ST (2012) Study of dissolved gas analysis under electrical and thermal stresses for natural esters used in power transformers. University of Manchester, Manchester
Zheng RR, Zhao JY, Wu BC (2009) Transformer oil dissolved gas concentration prediction based on genetic algorithm and improved gray verhulst model. International conference on artificial intelligence and computational intelligence, vol 4, pp 575–579
Li JQ, Wang DY, Yong J (2011) Forecast of mass concentration of dissolved gas in transformer oil based on combinative forecasting model. Guangdong Electric Power 24(9):19–23
Acknowledgements
All authors contributed equally the same to this article which is supported by National Natural Science Foundations of China (61503252 and 61473194), China Postdoctoral Science Foundation (2016T90799), Guangdong Province Fund(2014GKXM054) and Shenzhen—Hong Kong Technology Cooperation Fund(SGLH20161209101100926).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Lu, S.X., Lin, G., que, H. et al. Grey relational analysis using Gaussian process regression method for dissolved gas concentration prediction. Int. J. Mach. Learn. & Cyber. 10, 1313–1322 (2019). https://doi.org/10.1007/s13042-018-0812-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13042-018-0812-y