摘要
首先,介绍了传统神经网络,在其基础上引出了改进的卷积神经网络;然后,搭建了转子和滚动轴承的动力学模型,对转子和轴承的裂纹模型进行分析研究;最后,实现了联合收割机裂纹转子与滚动轴承故障诊断系统。实验结果表明:基于卷积神经网络的诊断模型达到稳定识别精度的迭代次数更少,且识别精度更高,效果更好,证明了系统的可行性和可靠性。
It first introduces the traditional neural network,and then leads to the improved convolution neural network.Then,it built the dynamic model of the rotor and rolling bearing,and analyzed the crack model of the rotor and bearing.Finally,it realized the fault diagnosis system of the cracked rotor and rolling bearing of the combine.The experimental results showed that it has less iterations to achieve stable recognition accuracy the diagnostic model based on convolution neural network.And it has higher recognition accuracy and better effect,which proves the feasibility and reliability of the system.
作者
詹宝容
庾锡昌
Zhan Baorong;Yu Xichang(Department of Information Engineering,Guangdong Innovative Technical College,Dongguan 523960,China;Dongguan Branch,China Mobile Group Guangdong CO.,LTD,Dongguan 523129,China)
出处
《农机化研究》
北大核心
2024年第5期187-191,共5页
Journal of Agricultural Mechanization Research
基金
广东省普通高校特色创新项目(2022KTSCX384)
广东创新科技职业学院校级科研项目(2022ZDYY01)。
关键词
联合收割机
卷积神经网络
转子
滚动轴承
裂纹
故障诊断
combine harvester
convolution neural network
rotor
rolling bearing
crackle
fault diagnosis
