Optimizing the Selection of Intermediate Charging Stations in EV Routing Through Neuro-Fuzzy Logic

We propose a comprehensive Electric Vehicle (EV) routing algorithm to find the optimal set of intermediate charging stations (CSs) present between a given source and destination. Each intermediate charging station is selected to maximize efficiency by considering three crucial parameters: distance to reach the destination from the selected CS, waiting time at the CS, and energy consumed to reach the selected CS along the route. Unlike existing algorithms, that focus solely on energy or distance, this algorithm integrates all three factors to generate an efficient path. Machine Learning (ML) is employed to predict vehicle range using data provided by the user, ensuring that the selected route avoids the risk of battery depletion midway. This predicted range is then used to determine CSs that can be reached from current location. Furthermore, the algorithm utilizes Breadth-First Search (BFS) to identify CS nodes with the least cost, enhancing routing accuracy. The cost of reaching each CS node is calculated using Neuro-Fuzzy Logic, which effectively handles uncertain inputs, which is common in EV routing scenarios. Comparative analysis against a recently proposed route planning algorithm (EV-RPA) reveals superior performance of the proposed approach, particularly as the number of CSs increases. It excels in all three aspects: distance covered, waiting time, and energy consumed, highlighting its effectiveness in optimizing EV routing.