摘要
深度学习在特征提取与模型拟介方而显示了其潜力和优势。对于特征提取精度要求高的故障诊断领域,引入深度学习具有重要的意义。特征提取的精度决定着故障状态辨识的结果。许多学者将深度学习应用在故障诊断领域,并取得不少的成果。本文介绍了深度学习在故障诊断中的研究现状,总结了深度学习在故障诊断中应用的研究现状和技术难点,最后对深度学习在故障诊断中的研究进行了展望。
Deep learning shows its potential and advantages in feature extraction and model quasi mediation.For high precision of feature extraction,it is of great importance to introduce depth learning in the field of fault diagnosis.The accuracy of feature extraction determines the results of fault state identification.Many scholars have applied deep learning to the field of fault diagnosis,and have achieved many results.This paper introduces the research status quo in fault diagnosis of deep learning,summarizes the research status of deep learning and technical difficulties in the application of fault diagnosis,finally the prospects of the study on deep learning in fault diagnosis.
作者
刘林凡
LIU Lin-fan(Hunan University of Technology,Zhuzhou Hunan,412007)
出处
《新型工业化》
2017年第4期45-48,61,共5页
The Journal of New Industrialization
基金
湖南工业大学研究生创新基金资助项目(CX1707)
关键词
深度学习
故障诊断
特征提取
状态辨识
综述
Deep learning
Fault diagnosis
Feature extraction
State identification
Overview