Enhanced Strategies of Electric Vehicle Fast Charging Stations and Reliability Assessment in Distribution Networks With Solar-Based Distributed Generation

Abstract

Industry insiders who want to lower the greenhouse gas emissions linked to conventional fuel cars consider electric vehicles (EVs) as a practical alternative for mobility. EVs are a potential problem even though their performance is limited by their low battery power, long service charging times, and high resource costs. To improve the EV performance, this manuscript presents the hybrid technique for the optimal position of electric vehicles fast-charging stations (EVFCSs) in the distribution network. The proposed scheme is a joined execution of Wild Horse Optimizer (WHO) and Gradient Boosting Decision Tree (GBDT), which is commonly named the WHO-GBDT technique. The primary goal of the research is to decrease loss of power and voltage deviation. The optimal position for an electric vehicle charging station (EVCS) is determined using the WHO method. GDBT is used to predict the load demand. The proposed WHO-GDDT regulates the placement of EVCS, balancing their integration with distributed generation while enhancing the sustainability and reliability of distribution networks. The proposed WHO-GBDT algorithm is actualized in the MATLAB platform and compared their performance with various existing strategies like the Forensic Investigation Algorithm, Archimedean Optimization Algorithm (FBIAOA), Tunicate Swarm Algorithm (TSA), and Cuttlefish Algorithm (CA). The simulation findings of the proposed scheme are validated under three cases in the IEEE 33 bus system, like load 1, load 2 and load 3. From the result, the proposed method effectively reduced loss of power and voltage variation by 58.24% and 90.47%, respectively.