基于GA优化支持向量机的变压器故障诊断

Fault diagnosis of transformer based on genetic algorithm optimization-based support vector machine

针对变压器故障诊断中出现的多故障分类问题,为提高支持向量机的多故障分类的准确率,利用遗传算法(GA)对支持向量机的相关参数进行了优化。将利用遗传算法优化的支持向量机(GA-SVM)应用于变压器故障诊断中,并与利用粒子群算法优化的支持向量机(PSO-SVM)的识别结果进行比较。对比试验结果可以看出,GA-SVM算法能够更为有效地选择支持向量机的相关参数,在很大程度上提高了变压器多故障分类的准确性。

Aiming at the problems that the fault diagnosis of the transformer appear fault classification, in order to improve the Support Vector Machine (SVM) fault classification accuracy, the Genetic Algorithm (GA) to optimize the parameters of SVM is utilized. Finally, the Genetic Algorithm to optimize the Support Vector Machine (GA-SVM) algorithm is applied to the fault diagnosis of transformer, and compared with the PSO algorithm to optimize the Support Vector Machine (GA-SVM) algorithm identification results. The results show that GA-SVM algorithm can obtain better classification result than the PSO-SVM algorithm, which has a significant guidance on the fault diagnosis of transformer.