Feature Extraction and Analysis of Bearing Faults: A Review

Year : 2025 | Volume : 15 | Issue : 02 | Page : 20 28
    By

    Varun Sharma,

  1. Assistant Professor, Department of Electrical & Electronics Engineering, KIET Group of Institution, Ghaziabad, Uttar Pradesh, India

Abstract

document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_abs_200995’);});Edit Abstract & Keyword

One of the most important steps in identifying bearing problems is feature extraction. In order to provide a more meaningful dataset, it entails locating and extracting pertinent features from raw bearing vibration signals. Tasks involving categorization and prediction can then make use of these attributes. In many practical applications, such as monitoring rotating machinery or electronic components, the raw signals collected (e.g., vibration, current, temperature) are often complex, high-dimensional, and noisy. Feature extraction helps reduce this complexity by isolating key characteristics such as statistical parameters, frequency components, or time-frequency patterns that are sensitive to faults. These extracted features enhance the capability of diagnostic models to detect, classify, and predict faults with greater accuracy and reliability. Moreover, by concentrating on the most relevant aspects of the data, feature extraction improves computational efficiency, making real-time monitoring and automated fault detection more practical. It also aids in distinguishing between different fault types and severities, which is essential for effective condition-based maintenance. In the experimental set up, there are four test bearings placed on a single shaft. First of all, data is fetched from all the four bearings and then various features, Max, Min, Mean, Standard deviation, RMS, Skewness, Kurtosis, Crest factor, and Form factor are calculated for all the four bearings and then a comparison is made among all of these features. All these features can be further used for fault classification.

Keywords: Fault diagnosis, feature extraction, bearing, test rig, vibration

[This article belongs to Trends in Mechanical Engineering & Technology ]

How to cite this article:
Varun Sharma. Feature Extraction and Analysis of Bearing Faults: A Review. Trends in Mechanical Engineering & Technology. 2025; 15(02):20-28.
How to cite this URL:
Varun Sharma. Feature Extraction and Analysis of Bearing Faults: A Review. Trends in Mechanical Engineering & Technology. 2025; 15(02):20-28. Available from: https://journals.stmjournals.com/tmet/article=2025/view=0


document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_ref_200995’);});Edit

References

  1. Hai Qui, Jay Lee, Jing Lin, Gang Yu. Wavelet filter based weak signature detection method and its application on rolling element bearing prognostics. Elsevier, J Sound Vib. 2006Feb; 289(4–5): 1066–1090.
  2. Rene Jaros, Radek Byrtus, Jakub Dohnal, Lukas Danys, Jan Baros, Jiri Koziorek, Petr Zmij, Radek Martinek. Advanced Signal Processing Methods for Condition Monitoring. Arch Computat Methods Eng. 2023; 30: 1553–1577.
  3. Masoud Jalayer, Amin Kaboli, Carlotta Orsenigo, Carlo Vercellis. Fault Detection and Diagnosis with Imbalanced and Noisy Data: A Hybrid Framework for Rotating Machinery. Machines, MDPI. 2022 Mar; 10(4): 237.
  4. Minh-Quang Tran, Meng-Kun Liu, Quoc-Viet Tran, Toan-Khoa Nguyen. Effective Fault Diagnosis Based on Wavelet and Convolutional Attention Neural Network for Induction Motors. IEEE Trans Instrum Meas. 2021 Dec; 71: 3501613(13p).
  5. Rajvardhan Jigyasu, Vivek Shrivastava, Sachin Singh. Hybrid multi-model feature fusion based vibration monitoring for rotating machine fault diagnosis. J Vib Eng Technol. 2023 May; 12: 2791–2810.
  6. Mustafa MO, Nikolakopoulos G, Gustafsson T. Faults classification scheme for three phase induction motor. Int J Syst Dyn Appl. 2014 Jan; 3(1): 1–20.
  7. Navarro-Navarro A, Ruiz-Sarrio Jose E, Biot-Monterde V, Antonino-Daviu Jose A, Sven Urschel. Application of Transient Analysis Techniques to Fault Diagnosis in Low- and Medium-Power Synchronous Machines. Machines. 2023 Feb; 11(2): 288.
  8. Andrea Bonci, Marina Indri, Renat Kermenov, Sauro Longhi, Giacomo Nabissi. Comparison of PMSMs motor current signature analysis and motor torque analysis under transient conditions. IEEE 19th International Conference on Industrial Informatics (INDIN). 2021 Sep; 1–6.
  9. Redon P, Picazo Rodenas MJ, Antonino-Daviu Jose A. Development of a diagnosis tool, based on deep learning algorithms and infrared images, applicable to condition monitoring of induction motors under transient regime. IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society. 2020 Oct; 2505–2510.
  10. Alok Kumar Verma, Haja Kuthubudeen, Viswanathan Vaiyapuri, Sivakumar Nadarajan, Narasimalu Srikanth, Amit Kumar Gupta. Bearings Fault Diagnosis in Electromechanical System Using Transient Motor Current Signature Analysis. In Conference on Advances in Condition Monitoring and Structural Health Monitoring. Singapore: Springer. 2021 Feb; 439–448.
  11. Bilal Asad, Toomas Vaimann, Anouar Belahcen, Ants Kallaste, Anton Rassolkin, Hamidreza Heidari. The Low Voltage Start-up Test of Induction Motor for the Detection of Broken Bars. IEEE 2020 International Conference on Electrical Machines (ICEM). 2020 Dec; 1481–1487.
Regular Issue Subscription Case Study
Volume 15
Issue 02
Received 20/05/2025
Accepted 10/06/2025
Published 20/06/2025
Publication Time 31 Days
37

[last_name]

My IP

PlumX Metrics