多策略融合改进AO优化SVM的变压器故障诊断研究  

Research on Transformer Fault Diagnosis Based on Multi-strategy Fusion and Improved AO-optimized SVM

作　　者：谢国民[1] 齐晓亮 XIE Guomin;QI Xiaoliang(School of Electrical and Control Engineering,Liaoning Technical University,Huludao 125000,China)

机构地区：[1]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125000

出　　处：《控制工程》2024年第11期2000-2009,共10页Control Engineering of China

基　　金：国家自然科学基金资助项目(51974151);辽宁省教育厅重点实验室基金资助项目(LJZS003)。

摘　　要：针对变压器故障诊断精度不高的问题,提出了一种多策略融合改进天鹰优化器(IAO)优化支持向量机(SVM)的变压器故障诊断模型。首先,采用核主成分分析(KPCA)方法对高维度据进行降维,减少数据中的稀疏性对结果的影响;其次,利用Tent混沌映射、动态扰动因子策略、点对称策略改善其寻优能力和收敛速度,通过算法寻优能力测试验证了其优越性;最后,利用IAO对SVM的参数寻优,克服SVM参数选择不良的弊端,建立变压器故障诊断模型。结果显示,与AO、WOA、GWO优化SVM相比,IAO优化SVM的诊断正确率分别提升了7.08%、9.74%、15.93%,同时,也优于最小二乘支持向量机(LSSVM)、BP神经网络(BPNN)、随机森林(RF)典型分类模型,验证了所建立的变压器故障诊断模型的优越性,并具有较强的泛化能力。To solve the problem of low accuracy of transformer fault diagnosis,a transformer fault diagnosis model based on multi-strategy fusion and improved aquila optimizer(IAO) optimization support vector machine(SVM) is proposed.Firstly,kernel principal component analysis(KPCA) is used to reduce the dimensionality of high-dimensional data to reduce the impact of sparsity on the results.Secondly,it uses Tent chaotic mapping,dynamic disturbance factor strategy and point symmetry strategy to improve its optimization ability and convergence speed,and verifies its superiority through algorithm optimization ability test.Finally,IAO is used to optimize SVM parameters to overcome the malpractice of poor SVM parameter selection,and a transformer fault diagnosis model is established.The results showed that compared with SVM optimized by AO,WOA and GWO,the diagnostic accuracy of SVM optimized by IAO is improved by 7.08%,9.74% and 15.93%,respectively.At the same time,it is superior to the typical classification model of least squares support vector machine(LSSVM),BP neural network(BPNN) and random forest(RF),which verifies the superiority of the proposed transformer fault diagnosis model and has strong generalization ability.

关 键 词：变压器 故障诊断 油中溶解气体分析 算法改进 支持向量机 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]