MATLAB-based vibration signal processing for fault diagnosis

• Main Authors: Motakabber, S. M. A., Hoque, Md. Mominul, Islam, Md. Rafiqul, Ihsan, Sany Izan, Islam, Gazi Zahirul, Zahirul Alam, A. H. M.
• Format: Article
• Language: English
• Published: AlamBiblio Publishers 2023
• Subjects: T10.5 Communication of technical information
• Online Access: http://irep.iium.edu.my/107261/1/107261_MATLAB-based%20vibration.pdf; http://irep.iium.edu.my/107261/; https://alambiblio.com/ojs/index.php/ajoeee/article/view/52
• Institution: Universiti Islam Antarabangsa Malaysia

Traditionally, vibration signal processing has been performed using analog and digital signal analyzers or writing code in intermediate and high-level computer languages. However, the advent of higher-level interpretive-based signal processing software products such as MATLAB has added a new dimension to vibration signal processing. This paper presents a method for analyzing motor vibration data using MATLAB. The method first pre-processes the vibration data to remove noise and baseline wander. Then, the frequency spectrum of the vibration signal is calculated using the Fourier transform. The frequency spectrum is then used to identify the dominant frequencies in the vibration signal. These dominant frequencies can be used to identify potential problems with the motor, such as bearing defects or misalignment. The method was studied on a set of vibration data collected from open source online data of a real motor. The results showed that the method was able to identify the dominant frequencies in the vibration signal accurately. The method was also able to identify the potential problems with the motor. This paper demonstrates the effectiveness of using MATLAB for analyzing motor vibration data. The method presented in this paper can be used to improve the reliability and efficiency of motor maintenance.