基于CPSO方法的伺服系统故障树诊断分析  

Fault tree diagnosis of forging press servo system based on CPSO method

作　　者：汤瑞 杨峥 TANG Rui;YANG Zheng(Department of Automotive Engineering,Kaifeng Technician College,Kaifeng 475000,Henan China)

机构地区：[1]开封技师学院汽车工程系,河南开封475000

出　　处：《锻压装备与制造技术》2024年第5期150-152,共3页China Metalforming Equipment & Manufacturing Technology

摘　　要：采用单一诊断方法无法达到合理的故障诊断率,综合故障树分析(FTA)与混沌粒子群(CPSO)方法相结合的方式可较好实现故障诊断。从实际测试信号角度出发,并与人工神经网络(ANN)与粒子群(PSO)方法结果进行对比。结果表明:CPSO方法故障诊断准确率获得比ANN高出7.55%,比PSO高出5.18%,实现了准确率的显著提升。CPSO方法 200次迭代后达到稳定适应值,可以在最短时间内完成迭代,有效避免产生局部极值问题,大幅提升了诊断精度。该研究可以拓展到其他机械信号测试领域,具有很好的应用价值。A single diagnosis method cannot achieve a reasonable fault diagnosis rate.Comprehensive fault tree analysis(FTA)and chaotic particle swarm(CPSO)method are combined to achieve fault diagnosis.From the point of view of actual test signal,the results are compared with those of ANN and PSO methods.The results show that the fault diagnosis accuracy of CPSO method is 7.55%higher than that of ANN and 5.18%higher than that of PSO,achieving a significant improvement in accuracy.After 200 iterations,the CPSO method reaches the stable adaptation value,which can complete the iteration in the shortest time,effectively avoid the problem of local extreme value,and greatly improve the diagnostic accuracy.This research can be extended to other mechanical signal testing fields and has good application value.

关 键 词：故障诊断 故障树分析(FTA) 混沌粒子群方法(CPSO) 

分 类 号：TM921.541[电气工程—电力电子与电力传动]