Optimizing hybrid renewable energy systems with integrated electric vehicle using a hybrid approach

This paper presents a hybrid technique for hybrid renewable energy system sizing optimization in the presence of electric vehicles. The proposed hybrid strategy is based on the cooperation of the light spectrum optimizer (LSO) and deep attention dilated residual convolutional neural network. Commonly it is named as LSO-DADRCNN technique. The proposed method's primary goal is to maximizing the system reliability, minimizing system costs, and ensuring sufficient power supply for both EVs and other electrical loads. The LSO method is utilized to optimize the sizing of the renewable energy system, taking into parameters such as load demand, photovoltaic power, wind turbine power, and battery capacity. The DADRCNN technique is utilized to predict the future energy demand of electric vehicles. The performance of the suggested method is put into practice on the MATLAB platform and contrasted with methods that are already in use. From the result, it concludes that the efficiency of the proposed methods becomes 96 %. In the existing methods, like heap-based optimizer (HBO), particle swarm optimization (PSO), and wild horse optimizer (WHO) the efficiency becomes 79 %, 85 %, and 72 %. The findings demonstrate that, in comparison to other existing methods, the proposed solution lowers costs by 1.7 dollars. Comparing the proposed approach to other existing methods, it achieves a high accuracy of 99.5%. The greatest result, 2996.348, the mean, 3000.100, the worst result, a standard deviation of 1.160348 and a median of 2998.816 are all performance indicators for the proposed approach that exceed those of other existing methods.