基于精英蜂群算法的变压器故障诊断

Transformer Fault Diagnosis Based on Elite Swarm Algorithm

针对变压器故障类型的复杂难辨的问题,为提高其故障信息的利用有效率,达到提高变压器的故障诊断的准确率效果,在结合DGA技术的基础上,提出了一种改进精英蜂群算法优化支持向量机的变压器油中溶解气体故障诊断的方法(EABC-SVM)。通过实验仿真验证,EABC-SVM故障诊断方法在变压器故障诊断实例中诊断准确率高达94.08%,比Rand-SVM、RandF、ABC-SVM、GA-SVM算法模型在变压器故障诊断实例中准确分别高了12.8%、12.8%、6.75%、2.6%。证明了EABC-SVM变压器故障诊断法在实际应用中的可行性,该算法模型泛化能力强、计算简单、分类精度高。

Transformer plays an irreplaceable role in power system,Once the fault occurs,it will have a great impact on the stable operation of the whole power system,Therefore,it is very important to accurately judge the type of transformer fault.Aiming at the complex and difficult identification of transformer fault types,it is necessary to improve the utilization of fault information and the accuracy of transformer fault diagnosis.Based on DGA technology,a fault diagnosis method of dissolved gas in transformer oil based on improved elite bee colony algorithm and support vector machine is proposed(EABC-SVM).It is verified by experiment and simulation,Fault diagnosis method of EABC-SVM,In the transformer fault diagnosis example,the diagnosis accuracy rate is as high as 94.08%,Compared with Rand SVM,RandF,ABC-SVM and GA-SVM,the accuracy in transformer fault diagnosis is improved by 12.8%,12.8%,6.75%and 2.6%respectively.The feasibility of EABC-SVM transformer fault diagnosis method in practical applicationand the algorithm model has strong generalization ability,simple calculation and high classification accuracy are proved.