Power-System Transient Stability Assessment Based on High-Level Sample Feature Extraction and Model Updating

Neural Network-Driven Fault Classification for HVAC Transmission Systems: A Comparative Evaluation of Voltage, Current, and Phase Angle Inputs

The results indicate that phase-angle information provides supplementary and class dependent discriminative value, but does not consistently improve all fault classes, whereas conventional voltage and current measurements alone represent a simpler and more stable alternative, whereas phase-angle measurements may be incorporated when synchronized phasor information is already available.

Forced Oscillation Detection Using Hybrid Knowledge–Deep Learning Features

A Hybrid Knowledge-DL network (HKD-SVM) that utilizes Support Vector Machine (SVM) as the classifier, making the network well-suited for nonlinear, high-dimensional classification scenarios with limited training samples, which are common in power system applications.

Adaptive Weighted Feature Mode Decomposition and Its Application in Bearing Fault Diagnosis

A copula entropy enhanced mRMR feature selection method for fault diagnosis of hydropower units

To address the challenges of conventional feature extraction methods in capturing nonlinear dependencies within fault signals and reducing feature redundancy during hydropower units fault diagnosis, this paper proposes a feature selection framework integrating the minimum redundancy maximum relevance (mRMR) criterion with copula entropy (mRMR-CE). This framework utilizes CE to capture both linear and nonlinear dependencies in vibration signals. Combined with the mRMR criterion to suppress feature redundancy, it achieves stable selection of highly discriminative features. In noisy environments and with various classifiers, the method shows strong performance and stability. To validate the effectiveness of the proposed frame, seven feature selection approaches—CE, mRMR, mRMR-CE, Pearson, Principal Component Analysis, Hibert-Schmidt independence criterion—Lasso, and concrete autoencoder —were applied to the training samples during the feature selection stage. Lastly, the chosen features were input into four different types of classifiers for training and testing: support vector machine, Random Forest, multi-layer perceptron, and extreme gradient boosting. Experimental results demonstrate that the proposed method exhibits outstanding performance on both the Case Western Reserve University (CWRU) bearing fault dataset and the Unit 3 dataset from a hydropower plant. On the CWRU dataset, it achieved an average precision of 99.76% and an average F1 score of 96.57%, while on the Unit 3 dataset, it attained a 98.96% average accuracy and an average F1 score of 98.97%. These results significantly outperform traditional feature selection methods while demonstrating high stability and robustness.

Arc-Fault Detection Using Stage-Wise Alignment and Feature Fusion of Dual Learnable Time–Frequency Representations

A progressive three-stage time–frequency learning framework that identifies series arc faults directly from normalized current waveforms and provides robust discrimination across unseen measurement sessions within the evaluated load categories and operating conditions is presented.

High Voltage Circuit Breaker Fault Detection Using Vibration and Acoustic Signals Analysis and Machine Learning

Reliable fault detection in high-voltage circuit breakers is essential for ensuring power system availability and reducing maintenance-related downtime. This paper proposes a hybrid machine learning framework for binary fault detection using vibration and acoustic signals acquired during circuit breaker operations. The framework integrates signal processing and feature engineering techniques, including Fast Fourier Transform (FFT), signal decomposition methods (VMD and EMD), and dimensionality reduction techniques (Random Projection, PCA, and Kernel PCA). The extracted features are classified using a Neural-Adaptive Tabu Search (NATS) model, which combines the nonlinear learning capability of artificial neural networks with the optimization capability of Adaptive Tabu Search. Experimental studies were conducted under normal and multiple fault conditions. Comparative results demonstrate that FFT-based feature extraction combined with Random Projection (RP) provides the most effective feature representation, while the proposed NATS classifier achieves binary classification accuracies exceeding 92% across all operating scenarios. Compared with VMD- and EMD-based approaches, the FFT-RP framework offers superior diagnostic performance with lower computational complexity. The results indicate that the proposed hybrid AI framework provides an effective and practical solution for real-time condition monitoring and intelligent fault detection of high-voltage circuit breakers, supporting predictive maintenance strategies in modern power systems.