Recent research led by Abhishek Pratap Singh from the Department of Electrical Engineering at Maulana Azad National Institute of Technology in Bhopal, India, presents an innovative approach to electric vehicle charging stations (EVCS) that leverages artificial intelligence. This study, published in the Ain Shams Engineering Journal, introduces an adaptive interaction artificial neural network (AI-ANN)-based power management controller (PMC) designed for DC microgrids that integrate solar photovoltaic systems, storage batteries, electric vehicles, and the main power grid.
The proposed system is particularly relevant for residential buildings and offices where electric vehicles may be parked for extended periods. It operates in two distinct modes: in the first mode, the electric vehicle acts as a power source, supplying energy back to the grid (vehicle-to-grid or V2G). In the second mode, the vehicle serves as a load, drawing energy from the grid (grid-to-vehicle or G2V). This dual functionality enhances the efficiency of energy use and supports the stability of the local energy grid.
One of the standout features of this AI-ANN-based controller is its ability to manage power effectively. If the solar photovoltaic system and storage battery do not provide enough energy to meet demand, the system can draw power from the electric vehicle. Should all other sources fall short, the controller will then source power from the grid. This flexibility is crucial as it allows for better energy management and reduces reliance on any single source.
The research highlights significant improvements in performance metrics compared to conventional controllers. Specifically, the AI-ANN-based PMC reduces the overshoot of the DC bus voltage from 9.6% to 0% and decreases settling time from 1.18 seconds to 0.52 seconds. Additionally, the rise time is minimized from 0.27 seconds to 0.25 seconds. These enhancements indicate a more stable and responsive energy management system, which is vital for maintaining consistent power supply and quality.
The commercial implications of this research are substantial. As electric vehicle adoption continues to rise, the need for robust charging infrastructure becomes increasingly critical. This intelligent power management system could pave the way for more efficient and sustainable charging stations, ultimately lowering operational costs and enhancing the user experience. Furthermore, the ability to utilize electric vehicles as energy sources can create new revenue streams for vehicle owners and charging station operators alike.
In Singh’s words, “The suggested power management controller tested for two different modes i.e., V2G and G2V using MATLAB Simulink software.” This indicates a rigorous testing process that adds credibility to the findings and suggests readiness for real-world applications.
As the energy sector seeks innovative solutions to integrate renewable energy sources and improve grid reliability, this research represents a significant step forward. The development of AI-based controllers for EV charging stations not only supports the transition to electric mobility but also enhances the overall efficiency of energy systems. The potential for commercial deployment of such technologies could lead to a more resilient and sustainable energy future.
