Mansouri, Seyedamir Nematbakhsh, Emad Ramos Galán, Andrés Chaves Ávila, José Pablo García González, Javier Aguado Sánchez, José Antonio 2026-01-27T08:38:07Z 2026-01-27T08:38:07Z 2026-03-15 0306-2619 https://doi.org/10.1016/j.apenergy.2025.127320 Artículos en revistas Smart prosumers with Distributed Generation (DGs) and controllable loads can provide cost-effective grid services. However, realizing this potential requires distributed optimization mechanisms that ensure market efficiency, participant privacy, and compliance with electricity market regulations. This paper presents a bi-level distributed optimization mechanism to maximize flexibility services from industrial parks and Internet Data Centers (IDCs) in distribution-level Congestion Management (CM) markets. The upper-level models the Distribution System Operator (DSO), which identifies congested lines using linear AC power flow analysis on pre-settled energy market results and sends corrective signals to prosumers. The lower level allows prosumers to adjust their operations accordingly and communicate updated transactions back to the DSO. A novel proxy-driven algorithm is proposed to facilitate service-sharing among geo-distributed IDCs, considering congestion issues. Additionally, an adaptive Alternating Direction Method of Multipliers (ADMM) algorithm enables decentralized coordination among market agents, achieving 74.52 % faster convergence than the standard ADMM. A real-world case study from Spain demonstrates that the proposed mechanism enables the grid operator to maximize grid services from prosumers, reducing congestion alleviation costs by 35.27 %. Moreover, IDCs reduced daily costs by 11.07 % through service-sharing and task-shifting aligned with CM market signals, while industrial parks achieved a 13.68 % cost reduction by aligning material production processes with CM market signals, both enabled by the proposed bi-level mechanism. Smart prosumers with Distributed Generation (DGs) and controllable loads can provide cost-effective grid services. However, realizing this potential requires distributed optimization mechanisms that ensure market efficiency, participant privacy, and compliance with electricity market regulations. This paper presents a bi-level distributed optimization mechanism to maximize flexibility services from industrial parks and Internet Data Centers (IDCs) in distribution-level Congestion Management (CM) markets. The upper-level models the Distribution System Operator (DSO), which identifies congested lines using linear AC power flow analysis on pre-settled energy market results and sends corrective signals to prosumers. The lower level allows prosumers to adjust their operations accordingly and communicate updated transactions back to the DSO. A novel proxy-driven algorithm is proposed to facilitate service-sharing among geo-distributed IDCs, considering congestion issues. Additionally, an adaptive Alternating Direction Method of Multipliers (ADMM) algorithm enables decentralized coordination among market agents, achieving 74.52 % faster convergence than the standard ADMM. A real-world case study from Spain demonstrates that the proposed mechanism enables the grid operator to maximize grid services from prosumers, reducing congestion alleviation costs by 35.27 %. Moreover, IDCs reduced daily costs by 11.07 % through service-sharing and task-shifting aligned with CM market signals, while industrial parks achieved a 13.68 % cost reduction by aligning material production processes with CM market signals, both enabled by the proposed bi-level mechanism. en-GB Revista: Applied Energy, Periodo: 1, Volumen: online, Número: , Página inicial: 127320-1, Página final: 127320-20 Instituto de Investigación Tecnológica (IIT) Internet data centers and industrial parks as flexibility providers in modern power systems: An ADMM-based coordination mechanism info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion info:eu-repo/semantics/openAccess Industrial parks; Internet data Centers; Congestion management; Distributed optimization; Smart prosumers; Smart grids Industrial parks; Internet data Centers; Congestion management; Distributed optimization; Smart prosumers; Smart grids