Search

Insufficient transformer fault diagnosis rate has always been a key problem restricting the safety and low efficiency of power grid operation. To solve this problem, a transformer fault diagnosis method based on improved seagull optimization algorithm support vector machine (ISOA-SVM) was proposed. Firstly, the fault diagnosis model of dissolved gas analysis in oil based on SVM was constructed, and the data in oil was processed by kernel principal component analysis (KPCA). Secondly, the optimal kernel function parameters and penalty coefficient of SVM were found by ISOA. Finally, the data was normalized into the ISOA-SVM model for diagnosis, and the operational state of the transformer was judged. The results were compared with other algorithm optimization models. The simulation results show that the fault detection method of the model is significantly superior to other models in fault identification speed and accuracy, which helps to ensure the stable operation of the transformer. [ABSTRACT FROM AUTHOR]