Bridging Mobility and Clean Energy: A Machine Learning Perspective on Plug-in Electric Vehicles and Smart Energy Networks

• Review cutting-edge machine learning applications in PHEV energy management and smart charging • Compare optimization techniques including reinforcement learning, neural networks, and MPC • Explore ML’s role in predictive maintenance and battery health enhancement • Assess ML-enabled solutions for Shared Autonomous Electric Vehicles (SAEVs) and dynamic mobility • Evaluate integration of ML-based charging models with renewable energy systems • Identify challenges and propose adaptive frameworks for real-world deployment of ML in PHEVs . As the urgency to combat climate change intensifies, the convergence of clean energy, electric transportation and artificial intelligence emerges as key to a sustainable future. Plug-in Hybrid Electric Vehicles (PHEVs), empowered by Machine Learning (ML), hold immense promise in redefining the transportation market. This paper reviews cutting-edge advancements in ML-integrated energy management strategies (EMS), predictive maintenance, and smart charging infrastructure for PHEVs. By inferring insights from over 200 peer-reviewed sources, it categorizes and compares optimization techniques, including reinforcement learning, neural networks, and model predictive control, highlighting their role in improving vehicle efficiency, battery longevity, and environmental impact. The review further delves into the transformative effects of Shared Autonomous Electric Vehicles (SAEVs) services, highlighting ML’s capability to analyse real-time data streams for intelligent mobility. Charging optimization models, dynamic pricing mechanisms, and integration with renewable energy systems are explored to underscore how ML bridges the gap between energy efficiency and grid resilience. The paper also investigates the challenges of implementing such an optimization model and proposes adaptive algorithms and collaborative frameworks to overcome real-world limitations. Hence, our work provides a unified, cross-domain exploration of PHEV charging infrastructure, EMS optimization, SAEV systems, and challenges of ML integration, into a comprehensive, well-structured framework that extends the focus of PHEV research. Thus, this review synthesizes current breakthroughs and charts a visionary path for future innovation by blending technology, infrastructure, and sustainability toward ML-enabled PHEVs to shift global attention toward cleaner transportation.