A decentralized bi-level framework for activating flexibility services from energy storage systems, electric vehicle batteries, and integrated demand response in multi-energy networks

Energy hubs are emerging as key enablers of flexibility in modern multi-energy systems, particularly as the integration of energy storage technologies expands across residential, commercial, and industrial sectors. Their ability to coordinate distributed resources, including storage, demand response, and vehicle-to-grid (V2G) assets, offers a scalable pathway to support system reliability and local balancing needs. Realizing this potential, however, requires decentralized coordination schemes that allow hubs and system operators to collaborate without extensive data exchange, a growing necessity in competitive and privacy-sensitive environments. This study develops a decentralized bi-level optimization framework that explicitly links the operational decisions of energy hubs with the real-time management of coupled electricity and gas networks. At the upper level, the system operator supervises both infrastructures, addresses energy imbalances, and specifies the flexibility requirements to be procured from upstream networks and downstream hubs. Intra-day uncertainties in demand and renewable generation are modeled using white noise with Gaussian, Beta, and Weibull probability distributions to capture realistic operational fluctuations. At the lower level, diverse hubs autonomously schedule their storage units, controllable loads, and V2G resources to maximize the flexible services they can offer. To enhance coordination performance, an adaptive Alternating Direction Method of Multipliers (ADMM) is introduced, which reduces communication exchanges and achieves a 63.27% faster convergence compared to the classical formulation. The proposed framework is implemented in GAMS and solved with GUROBI, using a test system that couples a 118-bus electrical distribution network with a 65-node gas distribution network. Results demonstrate that the model effectively activates multi-energy flexibility from integrated demand response, energy storage systems, and electric vehicles. This coordinated flexibility increases economic benefits for hubs, reduces the system operator's operational costs by up to 19.33%, and lowers total system losses by 6.12%.