Description

This Master Prompt focuses on the integration of artificial intelligence, machine learning, statistical modelling, and data-driven methodologies into electrical engineering research. It examines electrical fault prediction, equipment condition monitoring, electrical parameter preprocessing, predictive maintenance, remaining useful life estimation, degradation classification, and intelligent performance forecasting. Research applications include induction motor fault diagnosis, AC contactor life prediction, magnetic hysteresis modelling, semiconductor degradation analysis, and electrical load forecasting. The theme incorporates supervised and unsupervised learning, deep neural networks, convolutional architectures, recurrent neural networks, transfer learning, hybrid optimization algorithms, physics-informed modelling, and explainable artificial intelligence. It emphasizes data acquisition, signal preprocessing, feature extraction, model training, hyperparameter optimization, uncertainty quantification, cross-validation, and performance benchmarking. Particular attention is given to combining physical engineering knowledge with data-driven prediction to improve model reliability and interpretability. The objective is to provide comprehensive methodologies for developing intelligent electrical diagnostics, predicting component failures, estimating service life, and supporting evidence-based engineering decisions.