A Game-Theoretic Hybrid Optimization Framework for Dynamic Energy Management of Electric Vehicles

The energy management of electric vehicles can be formulated as a dynamic multi-agent optimization problem, where power sources (i.e., battery and supercapacitor) act as strategic decision-makers with competing objectives. Traditional methods fail to address nonlinear coupled dynamics and real-time adaptation. This paper proposes a game-theoretic learning framework combining improved fuzzy control with a Hybrid Whale Optimization Algorithm-Grey Wolf Optimizer (WOA-GWO). The fuzzy system is modeled as a cooperative game with optimized membership functions and rules, while WOA-GWO employs competitive hunting dynamics (whale swarm vs. wolf pack) to auto-tune parameters, achieving Nash equilibrium in search efficiency. Experimental results revealed Pareto-optimal outcomes: urban/highway scenarios showed 7.6%/5.2% battery loss reduction (non-cooperative baseline: 12.4%/9.8%), 18.3 kWh/15.4 kWh total consumption (22.1 kWh/18.9 kWh baseline), and 92.6%/94.8% efficiency (88.3%/91.2% baseline). Supercapacitor utilization reached 27.4%/29.7%, demonstrating evolutionary stable strategies for resource allocation. The core contribution of this article lies in a novel hierarchical game framework that utilizes competitive optimization WOA-GWO to automatically design a collaborative control strategy using fuzzy logic. This strategy has been formally proven to be an evolutionarily stable strategy for an energy storage system for robust energy allocation.