基于IFOA优化SVM的油浸式变压器故障诊断方法

摘要

In view of the small sample data of oil-immersed transformer fault diagnosis, correct factor was applied to Fruit Flying Optimization Algorithm, oil-immersed transformer fault diagnosis method was proposed to optimize SVM based on IFOA. Penalty factor C and the kernel function parameter g of the SVM were optimized by IFOA, realizing small sample data of oil-im-mersed transformer fault diagnosis. In order to verify the validity and reliability of this algorithm, IFOA-SVM and Grid Search-SVM, FOA-SVM, SVM algorithm were compared. Experimental results show that IFOA-SVM has higher accuracy than Grid Search-SVM, FOA-SVM and SVM ,The proposed algorithm is more suitable for oil-immersed transformer fault diagnosis.%针对小样本数据的油浸式变压器故障诊断,将修正因子引入果蝇优化算法,提出了一种基于IFOA优化SVM的油浸式变压器故障诊断方法.通过IFOA优化SVM的惩罚因子C和核函数参数g,实现小样本数据的油浸式变压器故障诊断.为了验证该算法的有效性和可靠性,将IFOA-SVM和Grid Search-SVM,FOA-SVM,SVM等算法进行比较.实验结果表明,IFOA-SVM比Grid Search-SVM,FOA-SVM和SVM具有更高的准确率,更加适合油浸式变压器的故障诊断.