摘要
针对刀具故障诊断信号信噪比低、诊断结果不准确等问题,采用局域均值分解(LMD)结合排列熵(PE)来处理采集到的刀具加工时的振动信号,然后将提取到的特征向量输入到训练好的长短期记忆神经网络(LSTM)中得到诊断结果,为了提高LSTM的诊断效率,结合卷积神经网络(CNN)对LSTM进行了改造;试验表明,文章提出的方法诊断准确率比BP神经网络提高了将近12%,改进LSTM网络比传统LSTM的诊断时间缩短了50%。
Aiming at the problems of low signal-to-noise ratio of tool fault diagnosis signals and inaccurate diagnosis results,local mean value decomposition(LMD)combined with permutation entropy(PE)is used to process the collected vibration signals during tool processing,and then the extracted feature vectors are inputted to the trained long and short-term memory neural network(LSTM)to get the diagnosis result.In order to improve the diagnosis efficiency of LSTM,combined with convolutional neural network(CNN)to transform LSTM.Experiments show that the diagnostic accuracy of the method proposed in this paper is nearly 12%higher than that of the BP neural network,and the improved LSTM network reduces the diagnostic time of the traditional LSTM by 50%.
作者
杨瑞元
尹晨
何建樑
王禹林
Yang Ruiyuan;Yin Chen;He Jianliang;Wang Yulin(Machinery Institute,Nanjing University of Science and Technology,Nanjing 210094,China)
出处
《计算机测量与控制》
2021年第6期25-29,共5页
Computer Measurement &Control
基金
南京理工大学机械工程学院重点科技项目(2018ZX04002001-008)。
关键词
刀具故障诊断
局域均值分解
排列熵
长短期记忆神经网络
卷积神经网络
tool fault diagnosis
local mean decomposition
permutation entropy
long and short-term memory neural network
convolutional neural network
