摘要
光电探测设备是一种具有实际应用价值的设备,其工作状态受外界干扰比较大,出现故障的概率高,针对传统光电探测设备故障的误诊率、虚警率高的难题,以准确对光电探测设备故障进行识别和诊断,设计了粒子群算法(PSO)优化神经网络的光电探测设备故障诊断模型。首先分析国内外对光电探测设备故障研究现状,指出了传统光电探测设备故障方法的局限性,然后采集光电探测设备工作状态信号,提取可以描述故障类型的特征向量,然后采用RBF神经网络建立光电探测设备故障诊断分类器,对RBF神经网络参数难以确定的难题,引入粒子群算法解决该难题,并与其它光电探测设备故障诊断模型进行了对比测试,所提方法的光电探测设备故障诊断正确率超过90%,光电探测设备故障的误诊率、虚警率低于10%,光电探测设备故障效果远优于对比模型。
Photoelectric detection equipment is a kind of equipment with practical application value.Its working state is greatly interfered by external environment,and the probability of failure is high.Aiming at the problems of high misdiagnosis rate and missed diagnosis rate of traditional photoelectric detection equipment,in order to accurately identify and diagnose the failure of photoelectric detection equipment,a fault diagnostic model of photoelectric detection equipment optimized by particle swarm optimization(PSO)neural network is designed.Firstly,the current situation of the research on the fault of photoelectric detection equipment at home and abroad is analyzed,and the limitations of traditional fault methods of photoelectric detection equipment are pointed out.Then,the working state signals of photoelectric detection equipment are collected,and the feature vectors that can describe fault types are extracted.The fault diagnosis classifier of photoelectric detection equipment is established by using neural network,and the problems determined by the parameters of RBF neural network are solved by the introduced particle swarm optimization algorithm.Compared with other fault diagnosis models of photoelectric detection equipment.The accuracy of fault diagnosis of photoelectric detection equipment is more than 90%in the proposed model,the misdiagnosis rate and missed diagnosis rate of photoelectric detection equipment are less than 10%,and the fault effect of photoelectric detection equipment is far better.
作者
吕德深
梁承权
LV Deshen;LIANG Chengquan(College of Mechanical and Electrical Engineering,Nanning University,Nanning 530299,China)
出处
《激光杂志》
北大核心
2020年第9期216-220,共5页
Laser Journal
基金
广西高校中青年教师基础能力提升项目(No.2019KY0934)
南宁学院科研团队项目(No.2018KYTD05)。
关键词
光电探测设备
工作状态
特征向量
粒子群优化算法
虚警率
photoelectric detection equipment
working state
eigenvector
particle swarm optimization algorithm
false alarm rate
