11, August 2025

Bayesian Optimization for Machine Learning–Based Diagnosis of Soft Faults in DC–DC Converters Using ICEEMDAN

VOLUME-11; ISSUE-8; AUG-2025

Author(s): Mabia Khatun, Md. Ariful Islam, Merina Akter

Authors Affiliations:

  1. Masters of Electrical Engineering, Anhui University of Science and Technology, Huainan, Anhui, China
  2. Masters of Mechanical Engineering, Changan University, xian, China
  3. Masters of Computer Science and Technology, Changan University, xian, China

DOIs:10.2015/IJIRMF/202508008     |     Paper ID: IJIRMF202508008


Abstract
Keywords
Cite this Article/Paper as
References

Soft faults in DC–DC converters gradually degrade their performance; however, conventional threshold-based methods often fail to detect these subtle issues. This study proposes a framework that leverages advanced signal processing and machine learning to diagnose soft-faults. The converter voltage signal is decomposed using Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) into intrinsic mode functions (IMFs). These IMFs were then denoised using wavelet thresholding to remove high-frequency noise. Time-domain statistical features (mean, root-mean-square, skewness, and kurtosis) are extracted from the denoised IMFs. The resulting feature vectors were used to train three classifiers: Support Vector Machine (SVM), Extreme Learning Machine (ELM), and XGBoost. The hyperparameters of each classifier were optimised using Bayesian optimisation. The ICEEMDAN-based features achieved a classification accuracy of 95.6 %(without tuning), which was significantly higher than the 88.9% accuracy achieved with the basic EMD. With Bayesian-tuned XGBoost, the accuracy further improves. The proposed integrated framework (ICEEMDAN decomposition → denoising → feature extraction → Bayesian-optimised classification) effectively exposes soft faults by mitigating the noise and mode mixing. This enhanced fault detection is valuable for critical applications in which undetected defects can lead to costly failures.

Bayesian optimization; DC–DC converter; fault detection; ICEEMDAN; machine learning; soft faults; wavelet denoising.

Mabia Khatun,  Md. Ariful Islam,  Merina Akter (2025); Bayesian Optimization for Machine Learning–Based Diagnosis of Soft Faults in DC–DC Converters Using ICEEMDAN, International Journal for Innovative Research in Multidisciplinary Field, ISSN(O): 2455-0620, Vol-11, Issue-8, Pp. 54-64.         Available on –   https://www.ijirmf.com/

  1. Mohan N., Undeland T. M., Robbins W. P., (2003): Power Electronics: Converters, Applications, and Design(3rd ed.). Wiley, Hoboken, NJ, USA.
  2. Kazerani M., (2020): A review of DC-DC converter topologies for renewable energy systems. IEEE Transactions on Power Electronics, 35(2), 984–1002.
  3. Blaabjerg F., Yang Y., Yang D., Wang X., (2018): Power electronics for renewable energy systems—Status and trends. IEEE Transactions on Industrial Applications, 54(6), 5814–5823.
  4. Patel M. R., Naikan V. N. A., (2018): Fault diagnosis of DC-DC converters using signal processing and machine learning. IEEE Transactions on Industrial Electronics, 65(7), 5726–5736.
  5. Cheng L., Qiu Z., Sripad K., (2019): A survey on fault diagnosis of power electronic circuits. IEEE Access, 7, 120985–121003.
  6. Huang N. E. et al., (1998): The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society A, 454(1971), 903–995.
  7. Torres M. E. et al., (2011): A complete ensemble empirical mode decomposition with adaptive noise. IEEE Transactions on Signal Processing, 60(5), 1056–1069.
  8. Colominas M. A., Schlotthauer G., Torres M. E., (2014): Improved complete ensemble EMD: A suitable tool for biomedical signal processing. Biomedical Signal Processing and Control, 14, 19–29.
  9. Donoho D. L., (1995): De-noising by soft-thresholding. IEEE Transactions on Information Theory, 41(3), 613–627.
  10. Yan R., Gao R. X., (2007): Wavelet packets and their applications in machine fault diagnosis. Mechanical Systems and Signal Processing, 21(2), 793–805.
  11. Randall R. B., (2021): Vibration-Based Condition Monitoring: Industrial, Aerospace, and Automotive Applications. Wiley, Hoboken, NJ, USA.
  12. Lei Y. et al., (2020): Applications of machine learning to machine fault diagnosis: A review and roadmap. Mechanical Systems and Signal Processing, 138, 106587.
  13. Cortes C., Vapnik V., (1995): Support-vector networks. Machine Learning, 20(3), 273–297.
  14. Huang G. B., Zhu Q. Y., Siew C. K., (2006): Extreme learning machine: Theory and applications. Neurocomputing, 70(1–3), 489–501.
  15. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., 2016, pp. 785–794.
Post Views: 60

Download Full Paper

Download PDF No. of Downloads:6 | No. of Views: 60