Fault Diagnosis of Power Transformer Using Dynamic Clustering Algorithm

A novel dynamic clustering algorithm based on artificial immune network is proposed in this paper. Firstly artificial immune network can get memory polls, which effectively represent the characteristics of fault samples, using the ability of immune memory and learning. Then genetic algorithm is used to dynamically optimize and select the best memory cells as initial clustering centers of kernel-based possibility clustering algorithm. A lot of fault samples are analyzed by this algorithm, and the results are compared with those obtained by BPNN. The results indicate that the samples can effectively be classified through the algorithm and precision of fault diagnosis can be improved.