Study On Fault Diagnosis Of Transformer Based On Dissolved Gas-in-oil Analysis

After the introduction of power transformer typical faults, the principle of gas generation in transformer is analysed, the traditional methods and artificial intelligence methods for transformer fault diagnosis based on dissolved gas-in-oil analysis (DGA) are introduced. The three type's aritificail neural network (ANN) is introduced, after then, and they are implemented in MATLAB for training and diagnosis of transformer. Consideration the complementary of various ANN diagnosis results, a method integrating various ANN diagnosis results by Dempster-Shafer (D-S) evidence theory is presented, which can improve the accuracy and effectivity of power transformer fault diagnosis result based on DGG, and it is verified by transformer fault cases.