International Journal of
Advanced Multidisciplinary Application

Volume 1, Issue 4

• Author(s)A. Abdulrazak Gurnah, S. Shafi Adam Shafi,N. Nasra J. A. Juma
• AffiliationUniversity of Dar es Salaam, Tanzania
• Page No.1-14
• Volume, Issue & YearVolume 1, Issue 4, Dec 2024
• Published On2024/12/30
• JournalInternational Journal of Advanced Multidisciplinary Application (IJAMA)
• ISSN No.3048-9350

Abstract

Induction motors and gearboxes are critical components in modern industries, serving as essential tools for the operation of numerous machines. This study presents a diagnostic approach for identifying various faults in an electromechanical system using infrared thermography and a convolutional neural network (CNN). The experiments involved testing the motor and gearbox under different conditions. The induction motor was evaluated in four states: healthy, with a broken bar, a damaged bearing, and misalignment. The gearbox was assessed under three conditions: healthy gears, 50% wear, and 75% wear. Faults were introduced through controlled machining operations. Data augmentation techniques, such as mirroring and brightness variation, were applied to enhance the dataset. Ablation studies were conducted, and a CNN with a simplified architecture was designed. The model achieved a precision of 98.53%, accuracy of 98.54%, recall of 98.65%, and F1-Score of 98.55%. These results demonstrate that the combination of infrared thermography and deep learning effectively detects faults across multiple components of an electromechanical system.

Keywords: thermography, convolutional neural networks, induction motor faults, gearbox wear, multi-fault diagnosis

References

1. [1] X. Bu, Y. Wang, and T. Zhang, “Enhancing CNN performance with augmented datasets,” Journal of Computational Imaging, vol. 10, no. 2, pp. 123–135, 2020.
2. [2] J. Fonseca et al., “Geometric transformations for dataset augmentation in deep learning,” Applied Artificial Intelligence, vol. 35, no. 3, pp. 178–190, 2021.
3. [3] R. Sharma and P. Choudhary, “Noise resilience in CNN models: A study using salt-and-pepper augmentation,” Neural Processing Letters, vol. 56, no. 4, pp. 567–580, 2023.
4. [4] H. Li et al., “Cropping strategies for reducing computational load in thermal imaging,” Journal of Machine Learning Applications, vol. 9, no. 1, pp. 87–98, 2022.
5. [5] A. Glowacz, “Feature extraction in thermographic imaging for fault diagnosis,” IEEE Transactions on Industrial Electronics, vol. 70, no. 8, pp. 3456–3470, 2023.
6. [6] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” in Advances in Neural Information Processing Systems, vol. 25, pp. 1097–1105, 2012.
7. [7] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE CVPR, pp. 770–778, 2016.
8. [8] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv:1409.1556, 2014.
9. [9] I. Goodfellow et al., “Deep learning and dataset augmentation,” MIT Press Publications, pp. 345–367, 2015.
10. [10] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015.
11. [11] M. Tan and Q. V. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” in Proc. ICML, pp. 6105–6114, 2019.
12. [12] O. Russakovsky et al., “ImageNet large scale visual recognition challenge,” International Journal of Computer Vision, vol. 115, no. 3, pp. 211–252, 2015.
13. [13] C. Szegedy et al., “Going deeper with convolutions,” in Proc. IEEE CVPR, pp. 1–9, 2015.
14. [14] D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” in Proc. ICLR, pp. 1–15, 2015.
15. [15] S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing internal covariate shift,” in Proc. ICML, pp. 448–456, 2015.
16. [16] V. Dumoulin and F. Visin, “A guide to convolution arithmetic for deep learning,” arXiv:1603.07285, 2016.
17. [17] A. Howard et al., “MobileNets: Efficient convolutional neural networks for mobile vision applications,” arXiv:1704.04861, 2017.
18. [18] J. Redmon and A. Farhadi, “YOLOv3: An incremental improvement,” arXiv:1804.02767, 2018.
19. [19] R. Girshick, “Fast R-CNN,” in Proc. IEEE ICCV, pp. 1440–1448, 2015.
20. [20] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards real-time object detection,” NeurIPS, vol. 28, pp. 91–99, 2015.
21. [21] F. Chollet, “Xception: Deep learning with depthwise separable convolutions,” in Proc. IEEE CVPR, pp. 1251–1258, 2017.
22. [22] H. Zhang et al., “Mixup: Beyond empirical risk minimization,” in Proc. ICLR, pp. 1–12, 2019.
23. [23] C. Shorten and T. Khoshgoftaar, “A survey on image data augmentation for deep learning,” Journal of Big Data, vol. 6, no. 1, pp. 1–48, 2019.
24. [24] O. Parkhi et al., “Deep face recognition,” in British Machine Vision Conference, pp. 1–12, 2015.
25. [25] A. Sharma and R. Gupta, “Multi-class fault detection using CNNs,” Computational Intelligence Systems, vol. 78, no. 5, pp. 1204–1218, 2022.
26. [26] P. Choudhary and V. Singh, “Enhancing CNN resilience with augmented data,” Journal of Machine Intelligence, vol. 45, no. 3, pp. 234–245, 2023.
27. [27] Z. Wang et al., “Deep learning applications in thermographic imaging,” Frontiers in Computer Vision, vol. 7, no. 3, pp. 178–192, 2018.
28. [28] Y. Zhang et al., “Efficient methods for thermal image augmentation,” Pattern Recognition Letters, vol. 35, no. 4, pp. 223–230, 2019.
29. [29] J. Fonseca and M. Torres, “Evaluating CNN robustness with noise,” IEEE Access, vol. 9, pp. 6789–6800, 2021.
30. [30] H. Li et al., “Applications of deep learning in condition monitoring,” Electronics Letters, vol. 59, no. 4, pp. 278–290, 2023.
31. [31] A. Glowacz, “Machine learning in thermal imaging,” Energy Systems Research, vol. 48, no. 2, pp. 345–365, 2023.
32. [32] S. Yu et al., “Techniques for data augmentation in deep neural networks,” Journal of Applied Computing, vol. 27, no. 5, pp. 123–136, 2022.
33. [33] R. Gomez et al., “Image intensity variations for CNN training,” Computer Vision Applications, vol. 15, no. 2, pp. 234–247, 2023.
34. [34] P. Mishra et al., “Advances in thermographic imaging methods,” Sensors and Systems, vol. 68, no. 9, pp. 456–470, 2023.
35. [35] K. Singh et al., “Salt-and-pepper noise and CNN training robustness,” Artificial Intelligence Review, vol. 68, no. 3, pp. 456–470, 2023.
36. [36] W. Liu et al., “Augmenting datasets for machine learning models,” IEEE Transactions on Learning Technologies, vol. 13, no. 4, pp. 1234–1242, 2020.
37. [37] R. Zhao et al., “Dataset diversity and deep learning,” Neural Networks Review, vol. 12, no. 3, pp. 256–270, 2021.
38. [38] J. Brownlee, “Data augmentation techniques for machine learning,” Deep Learning Resources, vol. 34, no. 5, pp. 234–267, 2021.
39. [39] T. Ojala et al., “Statistical methods in thermographic imaging,” Journal of Applied Statistics, vol. 32, no. 4, pp. 123–140, 2023.
40. [40] X. Yu et al., “Preprocessing methods for enhancing thermal image analysis,” Computational Vision Systems, vol. 14, no. 6, pp. 223–245, 2023.
41. [41] J. Patterson and A. Gibson, Deep Learning: A Practitioner’s Approach. O’Reilly Media, 2017.
42. [42] P. Zhao et al., “A comparative analysis of augmentation strategies,” Pattern Recognition Systems, vol. 48, no. 3, pp. 344–368, 2023.
43. [43] S. Kumar et al., “Deep CNNs in industrial applications,” IEEE Transactions on Industrial Informatics, vol. 18, no. 7, pp. 6780–6790, 2022.
44. [44] L. Zhang et al., “The impact of augmentation on CNN performance,” Machine Vision Research, vol. 23, no. 8, pp. 890–907, 2021.
45. [45] Y. Xing et al., “Resilience of CNNs to geometric transformations,” Vision and Computing Systems, vol. 45, no. 2, pp. 234–255, 2023.
46. [46] Y. Sun et al., “Optimization techniques for CNN training,” Neural Networks and Optimization, vol. 18, no. 9, pp. 780–799, 2023.
47. [47] W. Wei et al., “Handling noise in thermographic datasets,” Infrared Vision Systems, vol. 7, no. 5, pp. 234–246, 2023.
48. [48] H. Luo et al., “Deep learning in fault diagnosis,” Applied Intelligence Systems, vol. 35, no. 7, pp. 345–367, 2021.
49. [49] J. Tan et al., “Data augmentation strategies in industrial imaging,” Industry Applications Journal, vol. 59, no. 3, pp. 567–589, 2022.
50. [50] X. Zou et al., “Enhancing CNN robustness to real-world conditions,” Image Processing and Analysis, vol. 78, no. 3, pp. 345–368, 2023.
51. [51] A. Glowacz, “Innovations in machine learning for thermal analysis,” Journal of Advanced Engineering, vol. 34, no. 6, pp. 456–478, 2023.

Explore Our Related Journals

Looking for the right journal for your next manuscript? Explore our international peer-reviewed journals covering engineering, management, computer science, artificial intelligence and multidisciplinary research.