基于QPSO-LSSVM模型的变压器匝间短路故障诊断  

Diagnosis of Transformer Inter-Turn Short Circuit FaultBased on QPSO-LSSVM Model

作　　者：韦金国 Wei Jinguo(Datang Guanyinyan Water and Electricity Development Co.,Ltd.,Panzhihua 617012,China)

机构地区：[1]大唐观音岩水电开发有限公司,四川攀枝花617012

出　　处：《黑龙江科学》2025年第8期34-37,共4页Heilongjiang Science

摘　　要：为提高变压器匝间短路故障诊断的准确率,确保变压器正常运行,提出基于量子粒子群优化算法(Quantum Particle Swarm Optimization, QPSO)和最小二乘支持向量机(Least Squares Support Vector Machine, LSSVM)相结合的变压器匝间短路故障诊断模型,分析现有变压器匝间短路故障诊断模型存在的问题,利用QPSO算法优化LSSVM模型的惩罚因子C和核参数g,构建QPSO-LSSVM故障诊断模型。以110 kV变压器匝间短路故障数据为算例样本,分别采用BP、LSTM、LSSVM、PSO-LSSVM和QPSO-LSSVM五种模型进行分析,分析结果表明,QPSO-LSSVM实现了变压器匝间短路故障的智能诊断,具有较好的泛化能力和较高的故障诊断精度,为电气设备的状态检测与故障诊断提供了一种有效的分析方法。To enhance the diagnosis accuracy of transformer inter-turn short circuit fault and ensure their normal operation,a diagnosis model based on the combination of Quantum Particle Swarm Optimization Algorithm(QPSO)and Least Squares Support Vector Machine(LSSVM)is proposed.The problem in existing transformer inter-turn short circuit fault diagnosis models is analyzed,so the QPSO algorithm is utilized to optimize the penalty factor C and the kernel parameter g in the LSSVM model,and a QPSO-LSSVM fault diagnosis model is constructed.Through taking a 110 kV transformer inter-turn short circuit fault data as a case sample,the performance of BP,LSTM,LSSVM,PSO-LSSVM and QPSO-LSSVM is analyzed.The diagnostic results show that the QPSO-LSSVM has achieved intelligent diagnosis of transformer inter-turn short circuit faults,with superior generalization ability and high fault diagnosis accuracy.This provides an effective analysis method for the condition monitoring and fault diagnosis of electrical equipment.

关 键 词：变压器 匝间短路 故障诊断 量子粒子群 最小二乘支持向量机 

分 类 号：TM407[电气工程—电器]