Artificial Intelligence–driven cyber–physical energy resilience framework for secure and sustainable smart distribution networks

This paper presents an Artificial Intelligence-driven Cyber-Physical Energy Resilience Framework for next-generation smart distribution networks, responding to the critical demand for secure, adaptive, and sustainable grid operation amid complex cyber-physical disruptions. The framework combines cybersecurity-aware control with renewable energy sources to make it possible for systems to heal themselves, stay stable, and use energy in a way that is good for the environment in changing and unpredictable situations. It is tested on modified IEEE 33-bus and 118-bus test networks that include distributed energy resources (DERs) like photovoltaic systems, wind turbines, battery energy storage systems, and battery electric vehicles. The proposed model's main purpose is to allow secure inter-microgrid coordination (SIMC), adaptive network reconfiguration, and dynamic microgrid formation. This will make sure that critical loads are restored and faults are quickly isolated during disruptive events. A multi-objective optimization function (MOF) is created to maximize resilience, cybersecurity strength, and trading revenue at the same time, while minimizing energy not delivered (END), operational cost, and energy losses. Adaptive weighting coefficients dynamically prioritize objectives across three operational scenarios: (A) renewable uncertainty and economic operation; (B) fault recovery and cybersecurity restoration; and (C) multi-threat coordination with balanced objectives. To solve the MOF quickly, an AI-enhanced metaheuristic optimization mechanism is created by combining the Artificial Gorilla Troops Optimization (AGTO) algorithm with the Grey Wolf Optimizer (GWO) in the SIMC layer. The AGTO is nearly 30% faster than the baseline Hunter–Prey Optimization Algorithm (HPOA) when it comes to adaptive convergence. It also improves MOF by more than 20%, END by 85%, and critical load restoration by 95% in all cases. Comparative results show that the suggested AGTO–SIMC framework gives cyber-secure, resilient, and sustainable energy management a scalable, smart, and self-healing operational base. This supports the strategic growth of AI-enabled cyber–physical resilience in smart energy distribution systems.