摘要
为了提高滚动轴承故障诊断的精度,提出了自适应遗传算法优化BP神经网络故障诊断模型(Adaptive Genetic Algorithm to Optimize BP,AGA-BP)。首先,针对遗传算法易早熟收敛、易陷于局部最优解的问题,通过自动调整交叉概率和变异概率的操作对遗传算法进行改进,并用自适应遗传算法优化BP神经网络的权值和阈值。其次,利用优化的BP网络对轴承正常、内圈故障、外圈故障和滚动体故障4种工况进行故障诊断。最后,将AGA-BP网络与BP、GA-BP网络进行测试对比,验证滚动轴承诊断的有效性。结果表明:AGA-BP网络诊断精度达97.83%,能够有效且准确地对滚动轴承进行故障诊断,且其诊断精度及误差收敛均优于其他诊断模型。提出的AGA-BP诊断模型更适用于滚动轴承的故障诊断。
In order to improve the accuracy of rolling bearing fault diagnosis,an adaptive genetic algorithmis is proposed to optimize BP neural network fault diagnosis model(Adaptive Genetic Algorithm to Optimize BP,AGA-BP).Firstly,aiming at the problem that the genetic algorithm is easy to prematurely converge and is trapped in the local optimal solution,the genetic algorithm is improved by automatically adjusting the crossover probability and the mutation probability,and the weight and threshold of the BP neural network are optimized by the adaptive genetic algorithm.Secondly,the optimized BP network is used to diagnose the normal conditions of rolling bearings,inner ring faults,outer ring faults and rolling element faults.Finally,the test is compared with the BP and GA-BP networks to verify the effectiveness in the diagnosis of rolling bearings.The results show that the AGA-BP network can diagnose the rolling bearing fault effectively and accurately with the accuracy of 97.83%,and its diagnostic accuracy and error convergence are better than other convergence models.The AGA-BP diagnosis model is more suitable for fault diagnosis of rolling bearings.
作者
刘鹏
胡超
陈聪
刘申君
Liu Peng;Hu Chao;Chen Cong;Liu Shenjun(Jiangsu Aviation Technical College,College of Aeronautical Engineering,Zhenjiang,Jiangsu 212134,China)
出处
《机电工程技术》
2023年第11期237-239,245,共4页
Mechanical & Electrical Engineering Technology
基金
2022年江苏航空职业技术学院院级重点课题资助项目(JATC22010104)。
关键词
BP神经网络
自适应遗传算法
滚动轴承
故障诊断
BP neural network
adaptive genetic algorithm
rolling bearing
fault diagnosis
