摘要
数字化装备具有结构复杂、技术密集、信息化程度高等特点,传统的故障诊断方法需要拆装的部件多、故障定位准确率低,而深度学习能够从装备原始数据中挖掘有价值且敏感的特征,适合用于数字化装备的智能故障诊断;为此,首先进行了部队数字化装备故障诊断的现实困境和挑战分析,阐述了国内外数字化装备维修保障的研究现状,而后总结了装备故障诊断的主要方法和研究应用进展,重点将深度学习在装备故障诊断领域的研究成果进行了梳理,最后结合实际提出了基于深度学习方法实现数字化装备故障诊断的3种研究思路。
Digital equipment has the characteristics of complex structure,intensive technology,and high information level.Traditional fault diagnosis methods require multiple components to be disassembled and have low accuracy in fault localization.But deep learning can extract valuable and sensitive features from the raw data of equipment,it is suitable for intelligent fault diagnosis of digital equipment.For this purpose,this paper first analyzes the practical difficulties and challenges of digital equipment fault diagnosis in the military,expounds the research status of digital equipment maintenance support at home and abroad,and then summarizes the main methods and research progress of equipment fault diagnosis,emphatically sorting out the deep learning research results of in the equipment fault diagnosis field;Finally,three research ideas for implementing digital equipment fault diagnosis based on deep learning methods are proposed in combination with practical applications.
作者
刘奥林
古平
赵张鹏
LIU Aolin;GU Ping;ZHAO Zhangpeng(Shijiazhuang Campus,Army Engineering University,Shijiazhuang 050000,China)
出处
《计算机测量与控制》
2024年第5期1-7,共7页
Computer Measurement &Control
基金
全军军事类研究生资助课题(JY2021C093)。
关键词
数字化装备
维修保障
故障诊断
深度学习
研究综述
digital equipment
maintenance support
fault diagnosis
deep learning
research overview
