Machine Learning and Artificial Intelligence Techniques for Intelligent Control and Forecasting in Energy Storage‐Based Power Systems

ABSTRACT The tremendous penetration of renewable energy sources and the integration of power electronics components increase the complexity of the operation and power system control. The advancements in Artificial Intelligence and machine learning have demonstrated proficiency in processing tasks requiring computational, perceptual, and cognitive capabilities. Efficient and secure operation of power systems and effective enhancement of power system resilience require integrating massive amounts of real‐time data from intelligent electronic devices, wide area measurement systems, and modern machine learning technologies. Consistent with the ongoing development of AI techniques, electric power systems are experiencing substantial technical transformations aimed at speeding up computation, lowering and ensuring the consistent functioning of electrical power networks, as well as utility and customer expenses. Traditional methods of power system management have their uses, but they are often fraught with problems including inefficiency, high costs, poor accuracy, and adaptability to the wide variety of power grids in use today. The operational efficacy, planning, control, and forecasting capabilities of power systems are enhanced by artificial intelligence. More accurate forecasts and efficient use of resources allow control activities to be performed automatically for power grids. Artificial intelligence methods outperform numerical optimisation methods in processing vast volumes of data. These capabilities allow AI methods to automate power systems even more and boost their performance. The review looks at over 150 recent studies on machine learning, deep learning, and optimisation methods in power systems, with a focus on performance metrics like improved stability and adaptability in reinforcement learning‐based control strategies, solution time reductions of 30%–50% in deep learning–enhanced distributed multi‐objective optimisation, and forecasting accuracy improvements of up to 15% with LSTM and GRU models. This paper provides a thorough description of various AI methods towards forecasting, planning, control, and operation of power systems, which in turn reduces the computational time by giving a precise performance.