Adaptive Neuro-Fuzzy Control-Based Multi- Objective Energy Management for Solar- Integrated Battery–Supercapacitor Electric Vehicles

Abstract: Integrating electric vehicles with renewable energy is a core direction for building sustainable transportation systems, and it is also the central research focus in the current new energy transportation sector. This paper targets solar electric vehicles that integrate on-board photovoltaics and are equipped with a hybrid energy storage system (HESS) composed of power batteries and supercapacitors, and proposes a novel adaptive neuro-fuzzy inference system (ANFIS) energy management strategy. This hybrid energy storage system operates based on the complementary characteristics of its two component types: power batteries provide continuous basic power supply via their high energy density, while supercapacitors, relying on their high-power density, meet the fast charging and discharging demands of vehicle acceleration, deceleration, and braking energy recovery. As an auxiliary energy source, on-board photovoltaics can effectively extend driving range and reduce the whole vehicle’s reliance on the public power grid. The ANFIS controller in this paper combines the independent learning capability of neural networks and the interpretability of fuzzy logic. It can respond in real time to three core dynamic working condition variables: driving mode, solar irradiance level, and state of charge, and simultaneously achieve four optimization goals: minimizing battery degradation through intelligent power allocation, maximizing solar energy collection efficiency, optimizing the whole vehicle’s equivalent fuel economy, and maintaining the supercapacitor’s state of charge within a compliant operating range. This paper uses the MATLAB/Simulink platform to conduct simulation verification with three standard driving cycles: UDDS, HWFET, and US06. The proposed strategy is compared with two baseline strategies, namely a traditional rule-based controller and a pure fuzzy logic controller, and its practicality is also tested using real-world scenarios extracted from a natural driving database. The results show that compared with the baseline strategies, the proposed scheme reduces battery current stress by 23%, increases energy efficiency by 18%, lifts solar energy utilization by 31%, and extends the estimated battery cycle life by 35%. Meanwhile, it exhibits good robustness under different irradiance and temperature conditions, achieves fast convergence in controller training, and has outstanding cross-condition generalization ability. This study advances the development of next-generation intelligent energy management systems for solar-powered electric vehicles, and puts forward a highly robust adaptive scheme that can balance multiple conflicting objectives and adapt to the inherent uncertainties in two types of scenarios.

Keywords: Adaptive Neuro-Fuzzy Inference System (ANFIS), Electric Vehicles, Solar-Assisted EVs, Hybrid Energy Storage System, Multi-Objective Optimization, Energy Management System (EMS).