摘要
近二十年来,特别是近十年来,稀疏分解的理论与应用都取得了重大的发展,在故障诊断领域也有大量文献发表。针对已发表的文献,对稀疏分解方法的发展进行综述,在简要介绍稀疏分解方法的基础上,针对其存在的问题总结了现存的改进算法,包括改进原子库、改进计算速度,其中快速算法主要介绍智能稀疏分解方法,并总结了稀疏分解方法在旋转机械故障诊断中的应用,此外,还讨论了一些新的研究趋势,包括人工智能稀疏分解、针对特定信号的稀疏分解以及有利于故障诊断技术发展的复合故障稀疏分解方法等。总结得到,稀疏分解方法在旋转机械故障诊断领域的发展具有广阔的前景,未来会有大量结合人工智能的新算法出现。
In the recent two decades, especially in the last ten years, the theory and application of sparse decomposition have made great progress, and a great deal of literature has also been published in the field of fault diagnosis. Based on the published literature, the development of sparse decomposition is reviewed. Based on the brief introduction of sparse decomposition method, the existing improved algorithms are summarized according to the existing problems, including improving the atomic library and improving the computing speed. The fast algorithm mainly introduces the intelligent sparse decomposition method. The application of sparse decomposition in rotating machinery fault diagnosis is summarized. Some new research trends are also discussed, including artificial intelligence sparse decomposition, sparse decomposition for specific signal and compound fault sparse decomposition method which is helpful to the development of fault diagnosis teehnology. It is eoneluded that the sparse deeomposition method has a broad prospect in the development of rotating machinery fault diagnosis. In the future, a large number of new algorithms combining artificial intelligence will appear.
作者
胡年炜
杨建伟
姚德臣
Hu Nianwei;Yang Jianwei;Yao Dechen(School of Machine-electricity and Automobile Engineering,Beijing University of Civil Engineering Architecture,Beijing 100044,China;Beijing Key Laboratory of Performance Guarantee on Urban Rail Transit Vehicles,Beijing University of Civil Engineering Architecture,Beijing 100044,China)
出处
《现代制造工程》
CSCD
北大核心
2018年第11期155-161,共7页
Modern Manufacturing Engineering
基金
国家自然科学基金资助项目(51605023)
国家十三五重点研发计划项目(2016YFB1200402)
2017年度创新基地培育与发展专项项目(Z171100002217087)
关键词
稀疏分解
旋转机械
故障诊断
人工智能
sparse decomposition
rotating machinery
fault diagnosis
artificial intelligence