Intelligent Energy Management Strategy and Sizing Methodology for Hybrid Systems in Isolated Regions

Abstract

In this study, a comprehensive approach is presented for the sizing and management of hybrid renewable energy systems (HRESs) that incorporate a variety of energy sources, while emphasizing the role of artificial neural networks (ANNs) in system management. For optimal sizing of an HRES, the monthly average method wherein historical weather data are used to calculate the monthly averages of solar irradiance and wind speed, offering a well-balanced strategy for system sizing. This ensures that the HRES is appropriately scaled to meet the actual energy requirements of the specified location, avoiding the pitfalls of over- and under-sizing, and thereby enhancing the operational efficiency. Furthermore, the study details a cutting-edge strategy that employs ANNs for managing the inherent complexities of HRESs. It elaborates on the design, modeling, and control strategies for the HRES components by utilizing Matlab/Simulink for implementation. The findings demonstrate the proficiency of the ANN -based power manager in determining the operational modes guided by a specifically designed flowchart. By integrating ANN-driven energy management strategies into an HRES, the proposed approach marks a significant advancement in system adaptability, precision control, and efficiency, thereby maximizing the effective utilization of renewable resources.