Abstract:Dissolved gas analysis is an effective and important method for power transformer fault diagnosis.Here，the artificial immune algorithm is led into the analysis of gas dissolved in transformer oil.A power tranformer fault diagnosis method is proposed，which can get more characterization fault memory antibody characteristics of the sample collection and classification through the study of the fault samples by increasing the use of antigens，antibodies and memory types of information and so on.From the Matlab experimental data and the comparison of the results and the IEC of three rations，we can get the conclusion that this algorithm can get higher accuracy of diagnosis.