Artificial Immune Algorithm And Its Application In Fault Diagnosis Of Transformer

In the last few years we could perceive a great increase in interest in studying biologically inspired systems and the great development of the construction of artificial systems by drawing inspiration from biologically inspired systems. Among these, we can emphasize artificial neural networks, evolutionary computation, DNA computation, and now artificial immune systems- a relatively new field. The immune system is a complex of cells, molecules and organs which has powerful information processing capabilities, such as feature extraction, pattern recognition, learning, memory, diversity, distributed detection, and so on. Based on immunological principles, many immune-inspired models and artificial immune algorithms are being developed and have become a promising computational approach to solving problems in complex application domains, such as computer network security, data analysis and so on. However, although lots of productions about the theory and application research of immune algorithm have been obtained, the applied research of immune algorithm in fault diagnosis is on the initial stages.Research work in this paper is mainly concentrated on design of immune algorithm, and its application in pattern recognition and fault diagnosis. The main work in this paper is: Begin with reading a lot of documents and thinking deeply, summarize the important and useful mechanisms of natural immune system, summarize the development and principle of immune algorithms, put forward a common frame of immune algorithm based on the former work, discuss the design of immune algorithm in detail, summarize the development and principle of transformer fault diagnosis, put forward several fault diagnosis approaches using immune algorithm. By simulationexperiments, the performance of the immune algorithms is studied, and the validity of these immune algorithms in solving transformer fault diagnosis is assessed.