A Comprehensive Review of Optimal Scheduling Techniques for PV-Battery-Electric Vehicle Loads Using Scalable Quantum Non-Local Neural Networks

Abstract

The transition toward decentralized and renewable-based energy systems has increased the complexity of energy management and scheduling in smart grids. The integration of photovoltaic (PV) systems, battery energy storage systems (BESS), and electric vehicle (EV) charging infrastructures introduces dynamic load patterns, intermittent generation, and complex interactions. Traditional optimization methods are insufficient to address these challenges, necessitating advanced artificial intelligence and machine learning approaches.  This paper reviews recent advancements in intelligent energy scheduling using deep learning and hybrid optimization techniques. It highlights the role of non-local neural networks in capturing long-range dependencies within complex energy systems, enabling improved decision-making for load scheduling and demand response. Additionally, the integration of quantum-inspired optimization methods is explored for solving high-dimensional scheduling problems efficiently. The study examines various frameworks across smart homes, microgrids, and community energy systems, utilizing datasets such as smart meter data and IEEE benchmark systems. Hybrid approaches combining neural networks, reinforcement learning, and metaheuristic algorithms demonstrate superior performance in cost reduction, renewable energy utilization, and system stability. Key challenges, including scalability, uncertainty, and computational complexity, are discussed, along with future directions toward developing adaptive, efficient, and intelligent energy scheduling systems.