基于数据驱动的微小故障诊断方法综述

发布时间：2018-04-17 11:25

本文选题：微小故障诊断 + 数据驱动　；参考：《自动化学报》2016年09期

【摘要】：能否及时诊断出微小故障是保障系统安全运行并抑制故障恶化的关键,本文针对微小故障幅值低、易被系统扰动和噪声掩盖等特点,从数据驱动的角度对现有研究进行综述.并将其分为三大类:基于统计分析的微小故障诊断技术、基于信号处理的微小故障诊断技术和基于人工智能的微小故障诊断技术,进而对不同方法的基本研究思想、研究进展、应用以及局限性予以介绍.最后不仅指出复杂系统微小故障诊断研究中的现存问题,而且从增加新的信息、挖掘未利用的隐含信息和采用新的数学工具三个角度进行展望,提出基于关联性分析、基于多源信息融合、基于机器学习和基于时频分析四个值得探究的微小故障诊断思想.
[Abstract]:Timely diagnosis of small faults is the key to ensure the safe operation of the system and inhibit the deterioration of the fault, the fault is small low amplitude, disturbance and noise masking characteristics of the system, summarizes the existing research from the perspective of data driven. It is divided into three categories: small fault diagnosis technology based on statistical analysis. Based on the small signal processing and fault diagnosis technology based on micro fault diagnosis technology of artificial intelligence, research progress and basic research ideas, different method, application and limitation are introduced. Finally the existing problems not only refers to the study of small complex fault diagnosis system, and the increase of new information, hidden information and by three a view of new mathematical tools in mining exploitation, put forward the correlation analysis based on multi-source information fusion based on machine learning based and based on time frequency Analysis of four small fault diagnosis ideas worthy of inquiry.

【作者单位】：杭州电子科技大学自动化学院;浙江大学电气工程学院;
【基金】：国家自然科学基金(U1509203,61333005,61490701,61273170) 浙江省自然科学基金(LZ15F030001)资助~~
【分类号】：TP277;TP18

【相似文献】