基于IBBOA优化BP神经网络的变压器故障诊断  

Fault diagnosis of transformer based on BP neural network optimized by IBBOA

作　　者：赵玲玲 王群京[2,3] 陈权[1,4] 汪伟[5] ZHAO Ling-ling;WANG Qun-jing;CHEN Quan;WANG Wei(School of Electrical Engineering and Automation,Anhui University,Hefei 230601,China;National Engineering Laboratory of Energy-saving Motor&Control Technique,Anhui University,Hefei 230601,China;Power Quality Engineering Research Center,Ministry of Education,Anhui University,Hefei 230601,China;Anhui Key Laboratory of Industrial Energy-Saving and Safety,Anhui University,Hefei 230601,China;New Smart City High-Quality Power Supply Joint Laboratory of China Southern Power Grid(Shenzhen Power Supply Co.,Ltd.),Shenzhen 518020,China)

机构地区：[1]安徽大学电气工程与自动化学院,安徽合肥230601 [2]高节能电机及控制技术国家地方联合工程实验室,安徽大学,安徽合肥230601 [3]教育部电能质量工程研究中心,安徽大学,安徽合肥230601 [4]工业节电与用电安全安徽省重点实验室,安徽大学,安徽合肥230601 [5]南方电网公司新型智慧城市高品质供电联合实验室(深圳供电局有限公司),广东深圳518020

出　　处：《电工电能新技术》2021年第9期39-46,共8页Advanced Technology of Electrical Engineering and Energy

基　　金：国家自然科学基金项目(51637001);安徽省自然科学基金(1808085ME113);南方电网公司科技项目(090000KK52190169/SZKJXM2019669)。

摘　　要：为提高变压器故障诊断准确率,提出基于IBBOA优化BP神经网络的变压器故障诊断模型。在IBBOA-BP模型中,引入自适应权重,协调BBOA的全局和局部搜索能力;增加变异算子,提高蝴蝶种群的多样性,避免蝴蝶个体陷入局部最优。通过IBBOA优化BP神经网络的权值和阈值,避免BP神经网络出现易早熟问题,提高变压器故障诊断模型的准确性。且利用测试函数,通过与PSO、BOA对比,证明IBBOA算法具有更好的全局寻优能力、精确度和收敛速度。仿真计算表明,IBBOA-BP模型的变压器故障诊断正确率比PSO-BP和BOA-BP诊断模型正确率分别提高10.4477%和5.9701%。In order to improve the accuracy of the transformer fault diagnosis model,a transformer fault diagnosis model based on BP neural network optimized by intervention bidirectional butterfly optimization algorithm(IBBOA)is proposed.In IBBOA-BP,the introduction of adaptive weight can better coordinate the global and local search ability of bidirectional butterfly optimization algorithm(BBOA);the introduction of mutation operator improves the diversity of butterfly population and avoids butterfly individual falling into local optimum.By IBBOA to optimize the weights and thresholds of the BP neural network,the BP neural network is prevented from falling into the local optimum,and the reliability of the transformer fault diagnosis model is improved.In addition,the test function is used in the article,and compared with the particle swarm optimization algorithm(PSO)and the standard butterfly optimization algorithm(BOA).It is verified that the global optimization ability,accuracy and convergence speed of the IBBOA algorithm have been improved.The simulation calculation proves that the diagnostic accuracy of the IBBOA-BP fault diagnosis model is 10.4477%and 5.9701%higher than that of the PSO-BP and BOA-BP diagnosis models,respectively.

关 键 词：变压器 故障诊断 蝴蝶优化算法 自适应权重 变异算子 

分 类 号：TM41[电气工程—电器]