Artificial immune system (AIS) is a new kind of intelligent computation methods, has been a hot point of computational intelligence. Fault diagnosis can be considered as a pattern recognition problem, and it is an important application area of AIS. An accurate assessment of power transformer state analysis and its fault diagnosis technology has important theoretical significance and practical application value.This paper analyzed and summarized relevant theory and method of AIS and pattern recognition. The self-organization antibody net (soAbNet) and the antibody generation algorithm have been tested on some benchmark data sets from the UCI repository, test results show that the method is feasibility and effectiveness. Faults diagnosis of power transformers is discussed based on the antibody generation algorithm and the soAbNet.
|