为了提高变压器故障诊断正确率,笔者提出一种基于改进秃鹰(improved bald eagle search,IBES)算法优化最小二乘支持向量机(least squares support vector machine,LSSVM)的变压器故障诊断方法.利用高斯-柯西变异算子对最优秃鹰个体进行变异,使IBES算法能够及时局部最优,提高了IBES算法的收敛精度.采用IBES算法对LSSVM的核参数和惩罚参数进行优化,建立基于IBES-LSSVM的变压器故障诊断模型,并与BES-LSSVM、GWO-SVM和GA-BP模型进行仿真实验对比.结果表明,IBES-LSSVM模型的诊断正确率为 98.33%,比上述对比模型分别提高了 3.50%、7.27%和 9.26%,且计算时间最短,验证了该文所提变压器故障诊断方法的正确性和实用性.
Transformer Fault Diagnosis Based on LSSVM Optimized by Improved Bald Eagle Search Algorithm
In order to improve the accuracy of transformer fault diagnosis,a transformer fault diagnosis method based on improved bald eagle search(IBES)algorithm optimized least squares support vector machine(LSSVM)is proposed.The gaussian-cauchy mutation operator is used to mutate the optimal bald eagle individual,so that the IBES algorithm can be locally optimal in time,and the convergence accuracy of the IBES algorithm is improved.IBES algorithm is used to optimize the kernel parameters and penalty parameters of LSSVM,and a transformer fault diagnosis model based on IBES-LSSVM is established.The simulation experiments are compared with BES-LSSVM,GWO-SVM and GA-BP models.The results show that the diagnostic accuracy of IBES-LSSVM model is 98.33%,which is 3.50%,7.27%and 9.26%higher than that of the above comparison model,and the calculation time is the shortest,which verifies the correctness and practicability of the transformer fault diagnosis method proposed in this paper.
transformerfault diagnosisimprove bald eagle search algorithmleast squares support vector machinegaussian-cauchy mutation
国家科技期刊平台
NSTL
万方数据
