基于改进激活函数的一维卷积神经网络电机轴承故障诊断的研究  

Research on fault diagnosis of motor bearings based on one-dimensional convolutional neural network based on improved activation function

作　　者：任大卫 周舒昊 伦淑娴[1] 李明[1] REN Dawei;ZHOU Shuhao;LUN Shuxian;LI Ming(College of Control Science and Engineering,Bohai University,Jinzhou 121013,China)

机构地区：[1]渤海大学控制科学与工程学院,辽宁锦州121013

出　　处：《渤海大学学报（自然科学版）》2024年第1期74-80,共7页Journal of Bohai University:Natural Science Edition

摘　　要：提出了一种基于改进激活函数的一维卷积神经网络的电机轴承故障诊断的方法,该方法首先介绍了一维卷积神经网络的结构,然后详细说明了激活函数的改进点,最后通过仿真试验依次采用三种一维卷积神经网络对电机轴承故障进行分类,通过对比发现,此方法具有诊断准确率高、收敛速度快、无需人为提取故障特征等优点。This paper introduces a fault diagnosis method for motor bearings based on a one-dimensional convolutional neural network with an improved activation function.Firstly,the structure of one-dimensional convolutional neural network is introduced.Then the improvement of the activation function is introduced.Finally,three kinds of one-dimensional convolutional neural networks are used to classify the faults of motor bearings through simulation experiments.

关 键 词：激活函数 一维卷积神经网络 电机轴承故障诊断 

分 类 号：TH133[机械工程—机械制造及自动化] TP183[自动化与计算机技术—控制理论与控制工程]