An Intelligent Genetic Fuzzy Classifier for Transformer Faults

ABSTRACT

Classification of faults in transformers with high accuracy is fundamental to ensuring good power quality with least interruptions. Our current work develops an intelligent genetic algorithm (GA)-tuned fuzzy classifier for transformer fault identification. The proposed classifier is able to segregate all fault types using dissolved gas analysis (DGA) samples from real power transformers of HPSEB (India) and other sources. DGA samples have been pre-processed using the J48 algorithm. We propose to replace the conventional action selection procedure of reinforcement learning by a GA-based optimizer. The classifier is able to garner very high classification accuracy which is higher than the one obtained with benchmark fuzzy Q learning (FQL) and other conventional classifiers. With our approach, the average fault classification rate achieved is 91.85% (FQL) and 97.51% genetic algorithm fuzzy Q-learning (GAFQL) though with a slightly higher computational complexity over the FQL. Our proposed classifier could serve as an important tool in ensuring the healthy operation of power transformers.

KEYWORDS:

• Fuzzy Q learning
• Genetic algorithm fuzzy Q learning
• Genetic algorithm
• J48 algorithm
• Power transformers
• Reinforcement learning

Additional information

Notes on contributors

Amit Kukker

Amit Kukker obtained his BTech in electrical engineering from JECRC, RTU, Jaipur and MTech in ICE from SLIET University, Punjab. He is doing his PhD in instrumentation and control engineering from Netaji Subhas Institute of Technology, Delhi University, Delhi.

Rajneesh Sharma

Rajneesh Sharma obtained his BE in electrical engineering from Delhi College of Engineering, Delhi University and ME in control & instrumentation from Delhi College of Engineering, Delhi University. He obtained his PhD in intelligent control of non-linear system from Indian Institute of Technology, Delhi. Thereafter, he carried out post-doctoral research at Institute for Systems and Robotics, IST, Lisbon, Portugal for one year. He has worked in BHEL, Department of Telecommunications as IES and Indian Railways as IES officer prior to joining NSIT in 2001. He has published more than 40 research papers in reputed refereed journals and conferences. Email: [email protected]

Hasmat Malik

Hasmat Malik obtained his MTech from National Institute of Technology, Hamirpur and PhD in the area of power system from Indian Institute of Technology, Delhi. His major areas of interest are condition monitoring and fault diagnosis (CMFD), noise and vibration analysis, signal processing of power system & machines, intelligent techniques for condition monitoring and control of power system and power quality studies, renewable energy and high voltage engineering. He is working as an assistant professor in NSIT but currently doing post-doctoral research from Singapore. Email: [email protected]