AI-Enhanced Power Management Strategies for Next-Generation Hybrid Electric Vehicles: Deep Reinforcement Learning and Predictive Analytics

This study presents a comprehensive investigation into next-generation Power Management Strategies (PMS) for Hybrid Electric Vehicles (HEVs), focusing on artificial intelligence-enhanced approaches, bidirectional energy flow management, and emerging quantum computing applications. Recent advances in deep reinforcement learning, including Auto-Tune Soft Actor-Critic (ATSAC) and offline RL methods, demonstrate significant improvements in real-time adaptability and fuel efficiency. The integration of Vehicle-to-Grid (V2G) technologies opens new paradigms for HEV energy management, enabling vehicles to serve as mobile energy storage units. Furthermore, quantum computing emerges as a promising solution for complex optimization problems beyond classical computing capabilities. By synthesizing recent publications, this study provides a roadmap for intelligent, sustainable, and grid-integrated PMS frameworks for future HEV systems.