MFLRS‐RDF technique for optimal sizing and performance analysis of HRES

Abstract

AbstractAn optimum HRES size and performance using upgraded optimization technique is proposed in this paper. The upgraded optimization technique is the joint execution of moth flame optimization (MFO) with the integration of local random search (LRS) called MFLRS and random decision forest (RDF), and hence it is named as MFLRS‐RDF technique. Minimization of the life‐cycle cost of the HRES subject to some constraints by adjusting decision variables namely, photovoltaic area, swept area of wind turbines, fuel consumption of the biodiesel generator, and a number of batteries is the main objective of the proposed approach. The inputs for the design of HRES are solar radiation, wind speed, demand for load, and temperature that vary in time. In the proposed technique, the RDF optimally predicts the load demand based on the historical dataset. In light of the forecasted load demand, the MFLRS technique provides an optimal configuration of HRES. For enhancing the updating function of MFO, the LRS technique is utilized. Furthermore, the proposed technique optimizes the various parameters such as system total cost, various sources power generation, the contribution of various sources, continuity of supply to the unmet load, and load demand. HRES' parameters are optimized with the proposed strategy, and the proposed strategy provides a reliable solution. The proposed strategy is implemented on the working platform of MATLAB/Simulink. The HRES performance is evaluated using comparative analysis with existing techniques.