The Applicability of Machine Learning Algorithms in Predictive Modeling for Sustainable Energy Management

Abstract

AbstractThe world’s energy sector is having difficulties governing the best management synthesis because of challenges such as a request in production supply and demand design changes. Mapping data of the energy sector to machine learning (ML) can effectively alleviate the problem. ML algorithms can analyze equipment data, build predictive models and solve issues regarding sustainability. Innovative areas designed with ML algorithms can naturally react to fluctuations in power costs and control energy utilization. Frameworks dependent on ML can help energy providers to get ready to stay up with fluctuating sustainable power supplies through predicting energy demand, forecasting the maintenance period of pieces of equipment in energy plants such as sunlight based PVs, wind power and hydrogen sources enabling to eliminate the applicability limits of these renewable energy sources around the world. Designing smart grids in combination with advanced control techniques, such as model predictive control (MPC) enables to comfort satisfaction of consumers while handling constraints needed to meet sustainability. This paper is devoted to use of ML algorithms in different renewable energy sources and bridging ML with MPC to achieve sustainable energy management .KeywordsMachine learningPredictive modelingSustainable energyModel predictive control