Machine learning for machinery fault diagnosis


EOI: 10.11242/viva-tech.01.05.0MECH_12

Download Full Text here



Citation

Swapnil Raut, Hritik Pawar, Sagar Pawar, Omkar Rewale , "Machine learning for machinery fault diagnosis", VIVA-IJRI Volume 1, Issue 7, Article MECH_12, pp. 1-6, 2024. Published by Computer Engineering Department, VIVA Institute of Technology, Virar, India.

Abstract

Machinery fault diagnosis is vital in maintaining industrial machinery, focusing on identifying issues through various tools to prevent breakdowns and ensure efficient operation. In machine learning, accurately detecting and predicting faults poses challenges such as data quality, imbalanced datasets, and interpretability. Addressing issues like feature selection, generalization, early fault detection, data labeling, and real time processing is crucial for predictive maintenance. Advanced techniques like machine learning and data analytics analyze data, identify patterns, and predict potential failures. However, ongoing challenge persist in interpreting results and implementing effective maintenance strategies, emphasizing the complexity of machine fault diagnosis in industrial settings.

Keywords

Tri-axial accelerometer, AIML, Vibration, Algorithms, Diagnosis.

References

  1. JTauheed Mian, Anurag Choudhary, Shahab Fatima, “Multi-sensor fault diagnosis for misalignment and unbalance detection using machine learning.” IEEE Transactions on Industry Applications ( Volume: 59, Issue: 5, Sept.-Oct. 2023) 5749 – 5759, 2023.
  2. Cem Ekin Sunal, Vladimir Dyo, Vladan Velisavljevic, “Review of Machine Learning based fault detection for centrifugal pump induction motor.” IEEE Access (Volume: 10) 71344 – 71355, 2022.
  3. Jan Vrba, Matous Cejnek, Jakub Steinbach, Zuzana Krbcova, Entropy 2021, 23(9), 1130; https://doi.org/10.3390/e23091130, 2021.
  4. Md Junayed Hasan , Akhand Rai, Zahoor Ahmad ,And Jong-Myon Kim, “A fault diagnosis framework forcentrifugal pumps by scalogram based imaging and deep learning.” IEEE Access,Volume:9, 2021. Antonio J. Marques Cardoso.Andre S Barcelocs. "Current based bearing fault diagnosis using deep learning" Energies 2021, 14/9),2509, https://doi.org/10.3390/en14092509, 2021.
  5. RN Toma, AE Prosvirin, JM Kim, School of Electrical, Electronics and Computer Engineering, University of Ulsan, Ulsan 44610, Sensors 2020, 20(7), 1884, 2020.
  6. Antonio J. Marques Cardoso.Andre S Barcelocs. "Current based bearing fault diagnosis using deep learning" Energies 2021, 14/9),2509, https://doi.org/10.3390/en14092509, 2021.
  7. Zhe Hua, Yanca iXiao, Jiadong Cao, “Misalignment Fault Prediction of Wind Turbines Based on Improved Artificial Fish Swarm Algorithm. “Entropy 2021, 23(6), 692; https://doi.org/10.3390/e23060692, 2020.
  8. Sarmah N, Tiwari R, “Dynamic analysis and identification of multiple fault parameters in a crack rotor system equipped with active magnetic bearings.” Volume 28,2020- Issue 8, 1103-1134, 2020.
  9. P B Mallikarjuna EMAIL logo , M Sreenatha , S Manjunath and Niranjan C Kundur, From the “journal Journal of Intelligent Systems https://doi.org/10.1515/jisys-2019-0237.” 2020.
  10. Muhammad Sohaib Cheol Hong Kim and Jong-Myon Kim, "A hybrid feature model I and deep learningbased bearing” Sensors 2017, 17(12). 2876, https://doi.org/10.3390/17122876, 2017.
  11. Guan Z, "Vibration analysis of shaft misalignment and diagnosis method of structure faults for rotating machinery." Int J Performability Eng» 2017. Vol 13 Issue (4). 337-347, 2017
  12. Lixiang Duan, Menuyun Xie. Tangbo Bai, Jmpang Wang Expert Systems, "A new support vector data description method for machinery fault diagnosis with unbalanced datasets." Volume 64, Pages 239-246, 2016.
  13. P. Pennacchi and A "Vania, Model-based diagnostic techniques can be used to identify the faults that affect rotating machines." Volume 20, Issue 8 November 2006, Pages 2112 2147, 2004.
  14. BLI. M-Y.Chow, Y Tipsuwan, JC Hung. Neural network based on motor rolling bearing fault diagnosis." IEEE Transactions on Industrial Electronics (Volume 47, Issue: 5, October 2000) 1060-1069, 2000