基于麻雀算法优化BP神经网络诊断数控机床故障  

Faults diagnosis of CNC machine tool by using BP neural network optimized by sparrow algorithm

作　　者：王舒玮 WANG Shuwei(College of Mechanical and Electrical Engineering,Shanxi Datong University,Datong 037000,China)

机构地区：[1]山西大同大学机电工程学院,山西大同037000

出　　处：《沈阳工业大学学报》2023年第5期546-551,共6页Journal of Shenyang University of Technology

基　　金：山西省基础研究计划项目(202103021224313);山西大同大学2021年度科研专项课题项目(2021YGZX53).

摘　　要：针对数控机床的同一故障引发因素不同,甚至存在多因素共同作用的问题,提出了数控机床故障诱因分析方法,准确判断并查找故障位置,从而解决数控机床出现的故障.采用麻雀搜索算法改善BP神经网络性能,进而诊断数控机床服役中的常见故障,采集不同状态的故障信号作为BP神经网络样本,利用经过麻雀搜索算法优化后的BP算法识别机床的故障状态.结果表明:诊断结果发生误判的概率仅为2.29%,说明麻雀搜索算法优化BP神经网络在检测数控机床故障诊断问题中能够进行推广应用.Aiming at the problem that the same fault of computer numerical control(CNC)machine tool may be caused by different factors,or even by multiple factors working together,an inducing factor analysis method for CNC machine tool was proposed to accurately judge and find the fault location and solve the faults of CNC machine tool.The sparrow search algorithm(SSA)was used to improve the performance of BP neural network for diagnosing the common faults of CNC machine tools in service.The fault signals of different states were collected as the samples of BP neural network,and the fault state of machine tool was identified by the BP algorithm optimized by SSA.The results show that the misdiagnosis probability of diagnosing results is only 2.29%,which indicates that the SSA optimized BP neural network can be widely used for the fault detection and diagnosis of CNC machine tool.

关 键 词：麻雀搜索算法 神经网络 数控机床 故障诊断 优化处理 仿真 误差处理 训练样本 

分 类 号：TH165[机械工程—机械制造及自动化]