A hybrid technique for optimal sizing and performance analysis of hybrid renewable energy sources

This article proposes an optimum size and performance analysis of hybrid renewable energy sources using the atomic orbital search-recalling-enhanced recurrent neural network technique. The system consists of energy sources, like wind turbines and solar panels, energy storage systems, or battery-bank and biodiesel generator methods. The proposed optimization method is the integration of atomic orbital search and recalling-enhanced recurrent neural network and hence it is known as atomic orbital search-recalling-enhanced recurrent neural network technique. The major goal of the proposed method is to reduce the life-cycle cost of the hybrid renewable energy sources while taking into account certain limits by altering decision variables, such as the swept area of wind turbines, solar area, fuel consumption of the biodiesel generator, and battery quality. Solar radiation, wind speed, load demand, and temperature variations are the inputs for the design of hybrid renewable energy sources. In the proposed method, the recalling-enhanced recurrent neural network uses the historical dataset to estimate the load demand in the best possible way. The atomic orbital search method offers an ideal hybrid renewable energy sources setup based on the anticipated load requirement. The proposed method also optimizes a number of other factors, including system overall cost, power generation from different sources, the contribution of different sources, supply continuity for the load demand, and unmet load. The proposed method optimizes the characteristics of hybrid renewable energy sources, and it offers a reliable solution. The proposed method is done in MATLAB and its performance is examined by using the comparative analysis with existing methods.