Research on intelligent fault diagnosis of gears using EMD, spectral features and data mining techniques

14th International Conference on Science, Engineering and Technology

自然科学;工业技术

Sagar, M.^1 ; Vivekkumar, G.^1 ; Reddy, Mallikarjuna^1 ; Devendiran, S.^1 ; Amarnath, M.^2

School of Mechanical Engineering, VIT University, Vellore, Tamil Nadu

632014, India^1

Department of Mechanical Engineering, IIIT DM, Jabalpur, India^2

关键词: Classification algorithm;    Data acquisition system;    Empirical Mode Decomposition;    Feature extraction and classification;    Intelligent fault diagnosis;    Operating condition;    Prediction model;    Signal processing technique;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/263/6/062047/pdf
DOI  :  10.1088/1757-899X/263/6/062047

来源: IOP

【 摘 要 】

In this present work aims to formulate an automated prediction model using vibration signals of various gear operating conditions by using EMD (empirical mode decomposition) and spectral features and different classification algorithms. In this present work empirical mode decomposition (EMD) is a signal processing technique used to extract more useful fault information from the vibration signals. The proposed method described in following parts gear test rig, data acquisition system, signal processing, feature extraction and classification algorithms and finally identification. Meanwhile, in order to remove the redundant and irrelevant spectral features and classification algorithms, data mining is implemented and it showed promising prediction results.