基于ISSA-SVM的煤矿变压器故障诊断方法研究

Research on Fault Diagnosis Method of Coal Mine Transformer Based on ISSA-SVM

为有效提高煤矿变压器故障诊断精度,通过分析变压器油中溶解气体与故障类型的联系,提出基于ISSA-SVM的煤矿变压器故障诊断新方法。采用核主成分分析(KPCA)对煤矿变压器数据进行特征提取;采用Logistic混沌映射和高斯柯西-变异算子对传统麻雀算法(SSA)进行改进,基准测试函数实验结果表明ISSA寻优能力和收敛速度均有较大提高。通过ISSA优化SVM的参数建立煤矿变压器故障诊断方法模型,实验结果表明:ISSA-SVM、PSO-SVM、SSA-SVM诊断精度分别为94.91%、80.84、86.33%,ISSA-SVM有效提高煤矿变压器的诊断精度。

In order to effectively improve the fault diagnosis accuracy of coal mine transformer,a new method of coal mine transformer fault diagnosis based on ISSA-SVM is proposed by analysing the connection between dissolved gases in transformer oil and fault types.Kernel Principal Component Analysis(KPCA)is used to extract features from coal mine transformer data;Logistic chaotic mapping and Gaussian Cosi-variance operator are used to improve the traditional Sparrow Algorithm(SSA),and the experimental results of the benchmarking test function show that the ISSA optimisation seeking ability and convergence speed have been greatly improved.The parameters of SVM are optimised by ISSA to establish the coal mine transformer fault diagnosis method model,and the experimental results show that the diagnostic accuracies of ISSA-SVM,PS0-SVM,and SSA-SVM are 94.91%,80.84,and 86.33%,respectively,and that ISSA-SVM is effective in improving the diagnostic accuracies of coal mine transformer.