Bees Algorithm and PSO-Optimized Hybrid Models for Accurate Power Transformer Fault Diagnosis: A Real-World Case Study
Open Access
Abstract
This paper introduces an intelligent fault-diagnosis framework for power transformers that integrates hybrid machine-learning models with nature-inspired optimization. Current signals were acquired from a laboratory-scale three-phase transformer under both healthy and various fault conditions. A suite of 41 discriminative features was engineered from time–frequency and sparse representations generated via Discrete Wavelet Transform (DWT) and Matching Pursuit (MP). The resulting dataset of 2400 labeled segments was used to develop four hybrid models, PSO-SVM, PSO-RF, BA-SVM, and BA-RF, wherein Particle Swarm Optimization (PSO) and the Bees Algorithm (BA) served as wrapper optimizers for simultaneous feature selection and hyperparameter tuning. Rigorous evaluation with 5-fold and 10-fold cross-validation demonstrated the superior performance of Random Forest-based models, with the BA-RF hybrid achieving peak performance (98.33% accuracy, 99.09% precision). The results validate the proposed methodology, establishing that the fusion of wavelet- and MP-based feature extraction with metaheuristic optimization constitutes a robust and accurate paradigm for transformer fault diagnosis.Keywords
Funding Information
- Cardiff University, School of Engineering
This publication has 38 references indexed in Scilit:
- Adaptive fault diagnosis of railway vehicle on-board controller with large language modelsApplied Soft Computing, 2025
- Integrative fault diagnostic analytics in transformer windings: Leveraging logistic regression, discrete wavelet transform, and neural networksHeliyon, 2025
- Enhanced detection of power transformer winding faults through 3D FRA signatures and image processing techniquesElectric Power Systems Research, 2025
- Power transformer health index and life span assessment: A comprehensive review of conventional and machine learning based approachesEngineering Applications of Artificial Intelligence, 2024
- Fuzzy Self-tuning Bees Algorithm for designing optimal product linesApplied Soft Computing, 2024
- Setting-Less Differential Protection of Power Transformers Based on Wavelet TransformIranian Journal of Science and Technology, Transactions of Electrical Engineering, 2024
- Adversarial training-based robust lifetime prediction system for power transformersElectric Power Systems Research, 2024
- Machine learning based multi-method interpretation to enhance dissolved gas analysis for power transformer fault diagnosisHeliyon, 2024
- Enhancing reliability and efficiency of grid-connected solid-state transformer through fault detection and classification using wavelet transform and artificial neural networkElectrical Engineering, 2023
- A review of the applications of machine learning in the condition monitoring of transformersEnergy Systems, 2022