Transformer Fault Detection Method Based on RCMDE and POA-LSSVM

DOI: https://doi.org/10.62517/jike.202504208

Author(s)

Zhi Wang*, Hao Wang, Yuan Wang

Affiliation(s)

Hohhot Power Supply Branch, Inner Mongolia Power (Group) Co., Ltd, Hohhot, Inner Mongolia, China *Corresponding Author

Abstract

To improve the feature extraction capability and fault detection accuracy for transformers, a new transformer fault detection approach is introduced. This approach combines refined composite multiscale dispersion entropy (RCMDE) for feature extraction with pelican optimization algorithm (POA)-optimized least squares support vector machine (LSSVM) for classification tasks. Specifically, RCMDE is employed to capture the sound signals of transformers under various operating conditions and calculate their corresponding entropy features. Given that the classification performance of LSSVM is highly sensitive to parameter settings, POA is utilized to optimize these parameters, thereby establishing a robust POA-LSSVM model. The fault features extracted are then classified by utilizing the optimized model. Experimental outcomes indicate that the proposed method attains a comprehensive fault detection accuracy of 96.67%.

Keywords

RCMDE; POA; LSSVM; Rransformer; Fault Detection

References

[1]Ren S, Lou X. Rolling Bearing Fault Diagnosis Method Based on SWT and Improved Vision Transformer. Sensors, 2025, 25(7):2090-2090. [2]Zhang Z, Xu M ,Wang S , et al. Sequence-Aware Vision Transformer with Feature Fusion for Fault Diagnosis in Complex Industrial Processes. Entropy, 2025, 27(2):181-181. [3]Cabral W T, Gomes V F, Lima D R E, et al. Kolmogorov–Arnold Network in the Fault Diagnosis of Oil-Immersed Power Transformers. Sensors, 2024, 24(23):7585-7585. [4]Lv L H. Study on Fault Diagnosis of Power Transformer with Reduction Method of Attribute Significance. Advanced Materials Research, 2014, 3593(1049-1050):665-668. [5]Liu L, iu Z, ian X. Rolling bearing fault diagnosis based on generalized multiscale mean permutation entropy and GWO‐LSSVM. IET Science, Measurement & Technology, 2023, 17(6):243-256. [6]Zhiming Z. Bearing Fault Diagnosis Based on Refined Composite Multi‐Scale Dispersion Entropy and Extenics. IEEJ Transactions on Electrical and Electronic Engineering, 2021, 17(3):479-485. [7]Guo L, Xu C, Ai X, et al. A Combined Forecasting Model Based on a Modified Pelican Optimization Algorithm for Ultra-Short-Term Wind Speed. Sustainability, 2025, 17(5):2081-2081.