AI-Enabled Smart Grid: Integrating Electrical Power Systems, Computer Intelligence, and Wireless Communication Technologies

Abstract

The transition of conventional electrical networks into smart, flexible, and reliable energy systems requires more than digitalizing individual grid elements. Modern power systems comprise renewable energy sources, distributed resources, electric vehicles, smart meters, storage systems, smart substations, and controllable consumers. These grid assets produce vast amounts of heterogeneous data that need reliable communication infrastructure for appropriate utilization in day-ahead scheduling, real-time monitoring, control, and maintenance. This study focuses on electrical power systems, artificial intelligence (AI), edge-cloud computing, and wireless communications in the context of AI-enabled smart grids. An integrated research approach was undertaken by surveying the state-of-the-art methods in power system operation, machine learning, communication networks, interoperability, edge intelligence, and cybersecurity. Particular emphasis was put on unifying power and communication infrastructures through cross-layer concepts applied in the AI-enabled smart grid framework. The framework includes power, sensing and devices, communication, edge-intelligence, cloud-intelligence, and application/control layers interconnected via continuous cyber-physical feedback. The role of AI algorithms, including machine learning, deep learning, reinforcement learning, federated learning, anomaly detection, and digital twins was evaluated in power forecasting, renewable energy prediction, grid fault diagnosis, voltage control, demand response, maintenance, electric vehicle management, and cybersecurity. The communication-related aspects of the smart grid framework were examined by considering the requirements of 5G, WiFi, LPWAN, wireless mesh networks, fiber, and hybrid communication systems. The study concluded with recommendations on how power systems should transition from separate optimization of energy, computation, and communication domains towards communication-aware AI and grid-aware networking. Edge intelligence, federated learning, interoperability, quality of service adaptation, and cyber-secure control were identified as five key enablers of future smart grids. The study provides a cross-disciplinary insight into the design of next-generation communication-integrated smart energy systems.