Title: Air compressor fault diagnosis through statistical feature extraction and random forest classifier
Authors: S. Aravinth; V. Sugumaran
Addresses: Department of Mechanical Engineering, Rajalakshmi Institute of Technology, Chembarambakkam, Chennai, India ' School of Mechanical and Building Sciences (SMBS), Vellore Institute of Technology, Chennai Campus, Vandalur-Kelambakkam Road, Chennai 600127, India
Abstract: Fault occurrence and machine downtime in work area is one of the major concerns in many industries which lead to severe economic losses and causalities. The main causes behind these problems is nothing but in-avoidance of regular checking and periodical inspection of working environment. Here is one of the similar cases, where failures in compressor system lead to several losses in industrial aspect due to its enormous application. So monitoring and diagnosis of faults in compressor systemic proposed in this study to avoid regular breakdown and idle time of machineries in industrial and domestic applications. Out of several faults in compressor, five major and common faults were taken in this study and vibration parameters for each condition is measured using accelerometer sensor. Further signals were extracted and classified through machine learning approach for the efficient diagnosis and detection of faults in compressor system.
Keywords: condition monitoring; air compressor; statistical features; machine learning; vibration signals.
Progress in Industrial Ecology, An International Journal, 2018 Vol.12 No.1/2, pp.192 - 205
Received: 16 Nov 2017
Accepted: 13 Jun 2018
Published online: 25 Oct 2018 *