Type of Document	Dissertation
Author	Wang, Zhenyuan
URN	etd-08102000-21510032
Title	Artificial Intelligence Applications in the Diagnosis of Power Transformer Incipient Faults
Degree	PhD
Department	Electrical and Computer Engineering
Advisory Committee	Advisor Name Title Yilu Liu Committee Chair Hugh VanLandingham Committee Member Jaime De La Ree Lopez Committee Member Maury Nussbaum Committee Member Robert Broadwater Committee Member
Keywords	Neural Network Dissolved Gas-in-oil Analysis (DGA) Artificial Intelligence Fault Diagnosis Power Transformer Fuzzy System Expert System
Date of Defense	2000-08-08
Availability	unrestricted
Abstract This dissertation is a systematic study of artificial intelligence (AI) applications for the diagnosis of power transformer incipient fault. The AI techniques include artificial neural networks (ANN, or briefly neural networks - NN), expert systems, fuzzy systems and multivariate regression. The fault diagnosis is based on dissolved gas-in-oil analysis (DGA). A literature review showed that the conventional fault diagnosis methods, i.e. the ratio methods (Rogers, Dornenburg and IEC) and the key gas method, have limitations such as the “no decision” problem. Various AI techniques may help solve the problems and present a better solution. Based on the IEC 599 standard and industrial experiences, a knowledge-based inference engine for fault detection was developed. Using historical transformer failure data from an industrial partner, a multi-layer perceptron (MLP) modular neural network was identified as the best choice among several neural network architectures. Subsequently, the concept of a hybrid diagnosis was proposed and implemented, resulting in a combined neural network and expert system tool (the ANNEPS system) for power transformer incipient diagnosis. The abnormal condition screening process, as well as the principle and algorithms of combining the outputs of knowledge based and neural network based diagnosis, were proposed and implemented in the ANNEPS. Methods of fuzzy logic based transformer oil/paper insulation condition assessment, and estimation of oil sampling interval and maintenance recommendations, were also proposed and implemented. Several methods of power transformer incipient fault location were investigated, and a 7´21´5 MLP network was identified as the best choice. Several methods for on-load tap changer (OLTC) coking diagnosis were also investigated, and a MLP based modular network was identified as the best choice. Logistic regression analysis was identified as a good auditor in neural network input pattern selection processes. The above results can help developing better power transformer maintenance strategies, and serve as the basis of on-line DGA transformer monitors.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    Appendix.pdf 102.30 Kb 00:00:28 00:00:14 00:00:12 00:00:06 < 00:00:01   Chapter0.pdf 102.97 Kb 00:00:28 00:00:14 00:00:12 00:00:06 < 00:00:01   Chapter04.pdf 30.75 Kb 00:00:08 00:00:04 00:00:03 00:00:01 < 00:00:01   Chapter1.pdf 86.25 Kb 00:00:23 00:00:12 00:00:10 00:00:05 < 00:00:01   Chapter2.pdf 146.37 Kb 00:00:40 00:00:20 00:00:18 00:00:09 < 00:00:01   Chapter3.pdf 181.63 Kb 00:00:50 00:00:25 00:00:22 00:00:11 < 00:00:01   Chapter4.pdf 174.48 Kb 00:00:48 00:00:24 00:00:21 00:00:10 < 00:00:01   Chapter5.pdf 123.91 Kb 00:00:34 00:00:17 00:00:15 00:00:07 < 00:00:01   Chapter6.pdf 238.22 Kb 00:01:06 00:00:34 00:00:29 00:00:14 00:00:01   Chapter7.pdf 103.85 Kb 00:00:28 00:00:14 00:00:12 00:00:06 < 00:00:01   Chapter8.pdf 146.53 Kb 00:00:40 00:00:20 00:00:18 00:00:09 < 00:00:01   Chapter9.pdf 69.32 Kb 00:00:19 00:00:09 00:00:08 00:00:04 < 00:00:01   Reference.pdf 106.52 Kb 00:00:29 00:00:15 00:00:13 00:00:06 < 00:00:01

If you have questions or technical problems, please Contact DLA.