APPLICATION OF MACHINE LEARNING IN GRID CONNECTED DFIG BASED WIND ENERGY SYSTEMS FOR FAULT DIAGNOSIS AND CONTROL: A REVIEW
Anju Kumari, Asif Jamil Ansari, Monauwer Alam
Department of Electrical Engineering, Integral University, India
Abstract: Environmental climate change has increased the demand for pollution-free electrical energy generation systems. Among renewable energy technologies, wind turbines (WTs) have emerged as one of the most efficient and environmental friendly power generation sources due to their low operational emissions, scalability and relatively low installation costs. However, the reliability and maintenance of wind turbine systems remain critical challenges because mechanical and electrical faults can significantly reduce energy production and increase operational costs. Recent advances in Artificial Intelligence (AI), particularly Machine Learning (ML) have demonstrated strong potential for improving fault diagnosis and condition monitoring in wind turbines. Techniques such as Artificial Neural Networks (ANN), Adaptive Neuro-Fuzzy Inference Systems (ANFIS), and deep learning models including Convolutional Neural Networks (CNN) have been widely applied to detect early-stage faults and predict failures in turbine components such as generators, gearboxes and induction motors. Hybrid frameworks including spatiotemporal pattern networks (STPN) integrated with deep CNN architectures have further improved diagnostic accuracy by capturing complex temporal and spatial fault patterns. Despite these advancements, challenges remain in handling incomplete, noisy or missing sensor data, which can affect model training and real-time monitoring performance. This review critically analyzes existing state-of-the-art of ML and hybrid approaches for wind turbine fault diagnosis and highlights emerging research directions for automated, data-driven fault detection and intelligent control strategies aimed at improving reliability, predictive maintenance, and operational efficiency in modern wind energy systems.
Keywords: DFIG, Artificial Intelligence, Faults and Diagnosis, Wind Turbines, CNN
VOLUME 10 ISSUE 03 2026: 160 – 179