Aug 2026· Noise & Vibration Worldwide· 0 citations· 23 references
Abstract
The diagnosis of faults in power transformers (PT) plays a crucial role in ensuring the reliability and stability of power systems. However, traditional fault detection techniques are prone to low detection accuracy and are not stable enough under complex operating conditions. To address these challenges, this paper proposes a novel hybrid Variational Mode Decomposition (VMD), Graph Attention Network (GAN) with Adaptive Extreme Learning Machine (AELM) method for a transformer fault diagnosis framework. VMD can efficiently extract discriminative frequency bands of the transformer signals, GAN can dynamically learn the importance and relationship of the extracted feature, and AELM can classify rapidly and accurately with low complexity. The dataset was obtained from simulations of various fault conditions in the Matlab/Simulink. The experimental results demonstrate that the proposed method has an accuracy of 99.5%, a precision of 99.66%, a recall of 99.33%, and an F1-score of 99.5% compared to existing methods. The proposed hybrid framework enables efficient classification capability, feature learning with attention, and adaptive feature extraction, which contributes to improved performance.
This work presents an innovative approach for fault classification in power transformers, combining advanced wavelet transform with machine learning techniques. The proposed method stands out for its robustness against data missing conditions, which is a critical challenge in fault diagnosis for these systems. The approach employs machine learning algorithms for fault classification, even when data quality and availability are compromised due to transmission failures or possible interference in the connection circuit between the current transformers and the transformer protective relay. Its ability to adapt to data loss makes it highly suitable for real-world industrial applications, aligning with Industry 4.0 principles, particularly in environments where data integrity is essential for real-time analysis and decision-making. The approach’s effectiveness is validated through a comprehensive evaluation of a diverse dataset covering several critical faults. When compared with an existing threshold-based fault classificator, the proposed method demonstrated outperformance in fault classification both in scenarios with all data available and in scenarios with missing data, reaching expressive success rates of 100% and 92.9%, respectively, against 42.5% and 36.9% obtained by the conventional one for a signal-to-noise ratio of 40 dB. The proposed method’s robustness in challenging high-noise and missing-data scenarios reinforces its viability in industrial operations, where measurement infrastructure may be compromised by hardware malfunctions.
I. A. Dantas, R. P. Medeiros, F. Costa et al.· IEEE Access· 0 citations
Transformers play an indispensable role in any power system. The health condition of these devices should be the top priority. Early fault detection of these devices is essential to have sustainable power flow. There are many routinary transformer tests like winding test, furan analysis, insulation resistance tests, but dissolved gas analysis stands to be one of the most critical tests among others. This is to the fact that the DGA test can evaluate the major condition of the transformer. Dissolved Gas Analysis (DGA) methods, while widely used, often struggle with accuracy and scalability under complex fault scenarios. This paper proposed a novel ML-based DGA framework that integrates the IEEE standard with Principal Component Analysis (PCA) and Gradient Boosting Machine (GBM) to enhance transformer fault diagnosis. PCA captures 95% of the variance with five principal components. The framework showed a test accuracy of 87.5% and a cross-validation accuracy of 86.05%, outperforming traditional methods such as the Duval Triangle (83.08%) and IEC Ratio Method (82.05%), as well as other machine learning models, including Random Forest (77%) and Support Vector Machines (37%). These findings demonstrate the effectiveness of the framework as a soft sensor in Transformer diagnostics.
Apolinario Awit, Joseph Jay Brañanola, Maria Vina Presbitero et al.· international journal of eng...· 0 citations
The continuous complexities and the requirement of reliable power transmission have led to the necessity for advanced fault detection systems in order to ensure stability and safety in electrical grids. In this study, an ensemble learning-based technique has been proposed for fault detection in power transmission lines by using Deep Q-Networks (DQN) in conjunction with standard classifiers such as Naive Bayes, Multilayer Perceptron (MLP), Logistic Regression, and Deep Forest. The proposed algorithm uses reinforcement learning for optimizing classifier weights. Voltage and current waveforms are used as input data for extracting features and performing classification. The performance of the proposed technique is highly improved, providing 99.10% accuracy, 99.40% precision, 99.70% sensitivity, and 99.50% F1-score, which is far better than the existing techniques. Moreover, the proposed framework provides reduced processing time (12-13 milliseconds) and low memory consumption (150-152 megabytes).
Sandeep Godhade, Jayendra Kumar· International Journal of Ele...· 0 citations
To address the insufficient diagnostic precision and weak anti-interference capability of the Probabilistic Neural Network (PNN) in identifying transformer faults, an enhanced diagnostic approach integrating the Sparrow Search Algorithm (SSA) with PNN is developed. The SSA is applied to search for the optimal smoothing factor of PNN, and the tuned parameter is then fed into the PNN framework for model training, thereby constructing a high-performance fault identification model. Several conventional approaches are also implemented for benchmarking purposes. Experimental outcomes reveal that the developed SSA-PNN model outperforms Particle Swarm Optimization (PSO)-PNN, Grey Wolf Optimizer (GWO)-PNN, and the standard PNN by margins of 7.1%, 10.7%, and 28.6% in overall diagnostic accuracy, respectively. Under data perturbation conditions, the accuracy degradation of the developed model remains minimal among all compared methods, which confirms that SSA can substantially strengthen the anti-disturbance capability of PNN, thereby providing reliable technical support for power transformer fault diagnosis and offering certain reference value for enhancing the operational reliability of power grid equipment.
Bo Liu, Jianghong Dong, Shuyu Ren et al.· Journal of Physics, Conferen...· 0 citations
The stable detection of faults in smart power grids is essential when focusing on the stable operation and the reduction of the downtime. In this paper, the author suggests a hybrid deep learning-based model that combines convolutional neural networks (CNN) and long short-term memory (LSTM) with explainable artificial intelligence (XAI) to detect and classify faults accurately and interpretably. The model aims at capturing the spatial and time-varying attributes of multivariate electrical signatures such as voltage, current, and frequency changes. A combination of real time sensor measurements and simulated fault conditions are used which includes several fault classes including LG, LL, LLG, and three phase faults. Experimental evaluation demonstrates that the proposed model achieves a classification accuracy of 97.84%, with an F1-score of 97.08%, outperforming conventional CNN and LSTM models by more than 3%. The framework is also able to work in a noisy environment with a stable performance of less than 2% performance decrease, and inference latency of 18 ms, which can be deployed in real-time. Besides, the combination of SHAP and attention processes improves the interpretability through the detection of the crucial features that lead to fault prediction. The findings suggest that the suggested solution is a powerful, scalable, and clear solution when it comes to intelligent fault management in a contemporary smart grid.
Udit Mamodiya, Divyanshu Sinha, I. Kishor et al.· Scientific Reports· 0 citations
The diagnosis of faults in rotating machinery is essential for maintaining reliability and operational efficiency. Shaft misalignment and bearing unbalance represent two of the most prevalent defects in such systems. This study investigates the effectiveness of a model-driven approach
compared with machine learning (ML) techniques for the automatic detection of these faults, using vibration features extracted from vibration signals as diagnostic inputs. The model-driven method is based on a Bayesian framework (Acoem Accurex), while the ML approaches include a fully connected
neural network and extreme gradient boosting (XGBoost). Experimental results indicate that XGBoost achieves the highest accuracy (70%) in identifying unbalance, outperforming the neural network (66%) and the Bayesian model (58%). For misalignment detection, however, the
three methods exhibit comparable performance, underscoring the limitations of ML models that rely solely on vibration indicators for this fault type.
A. Marzougui, A. Hachem, T. Mazoyer· Insight - Non-Destructive Te...· 0 citations