Isolated power systems,such as those on islands,face acute challenges in balancing energy demand with limited generation resources,making them particularly vulnerable to disruptions.This paper addresses these challeng...Isolated power systems,such as those on islands,face acute challenges in balancing energy demand with limited generation resources,making them particularly vulnerable to disruptions.This paper addresses these challenges by proposing a novel control and simulation framework based on a holonic multi-agent architecture,specifically developed as a digital twin for the Mayotte island grid.The primary contribution is a multi-objective optimization model,driven by a genetic algorithm,designed to enhance grid resilience through intelligent,decentralized decisionmaking.The efficacy of this architecture is validated through three distinct simulation scenarios:(1)a baseline scenario establishing nominal grid operation;(2)a critical disruption involving the failure of a major power plant;and(3)a localized fault resulting in the complete disconnection of a regional sub-grid.The major results demonstrate the system’s dual resilience mechanisms.In the plant failure scenario,the top-level holon successfully managed a global energy deficit by optimally reallocating shared resources,prioritizing grid stability over complete demand satisfaction.In the disconnection scenario,the affected holon demonstrated true autonomy,transitioning seamlessly into a selfsufficient islanded microgrid to prevent a cascading failure.Collectively,these findings validate the holonic model as a robust decision-support tool capable of managing both systemic and localized faults,thereby significantly enhancing the operational resilience and stability of isolated smart grids.展开更多
基金funding for this study from the European Union’s Horizon 2020 research and innovation programme under grant agreement No.957843The research was supported by the MAESHA project.More information can be found at the Horizon 2020 website:https://cordis.europa.eu/project/id/957843/fr(accessed on 28 September 2025).
文摘Isolated power systems,such as those on islands,face acute challenges in balancing energy demand with limited generation resources,making them particularly vulnerable to disruptions.This paper addresses these challenges by proposing a novel control and simulation framework based on a holonic multi-agent architecture,specifically developed as a digital twin for the Mayotte island grid.The primary contribution is a multi-objective optimization model,driven by a genetic algorithm,designed to enhance grid resilience through intelligent,decentralized decisionmaking.The efficacy of this architecture is validated through three distinct simulation scenarios:(1)a baseline scenario establishing nominal grid operation;(2)a critical disruption involving the failure of a major power plant;and(3)a localized fault resulting in the complete disconnection of a regional sub-grid.The major results demonstrate the system’s dual resilience mechanisms.In the plant failure scenario,the top-level holon successfully managed a global energy deficit by optimally reallocating shared resources,prioritizing grid stability over complete demand satisfaction.In the disconnection scenario,the affected holon demonstrated true autonomy,transitioning seamlessly into a selfsufficient islanded microgrid to prevent a cascading failure.Collectively,these findings validate the holonic model as a robust decision-support tool capable of managing both systemic and localized faults,thereby significantly enhancing the operational resilience and stability of isolated smart grids.
