Sivianes, Manuel , VELARDE RUEDA, PABLO ANIBAL, Zafra-Cabeza, Ascensión , Maestre, José M. , Bordons, Carlos
No
Energy Systems Integration for Multi-Energy Systems
Capítulo de un Libro
Científica
13/05/2025
2-s2.0-105005409848
This work formulates a hierarchical and distributed optimization framework for energy communities operating under uncertainties. This framework aims to eliminate the need for a centralized coordinator by utilizing a smart contract deployed on a blockchain network. This scheme consists of two levels. At the higher level, households employ stochastic model predictive controllers to compute hour-by-hour scheduling for an entire day independently. Concurrently, a distributed optimization process facilitated by the smart contract acting as a distributed coordinator on a blockchain platform enables households to reach a consensus. At the lower level, model predictive controllers are responsible for tracking the references imposed by the higher level and derived from the globally agreed solution achieved through consensus within the smart contract. In this level, the control variables are generated with a lower sampling time than in the higher level. Unlike existing approaches that often rely on auctions and continuous blockchain transactions for energy exchanges, our methodology takes a different approach. We compute a reliable one-hour control sequence at the higher level, considering perturbations. This is followed by a much lower sampling time low-level controller that operates independently and does not rely on blockchain. Extensive simulations validate the framework’s effectiveness in optimizing economic costs associated with energy utilization and control efforts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
Stochastic control systems; Stochastic models; Block-chain; Distributed framework; Distributed optimization; Energy; Hierarchical optimization; Model predictive controllers; Model-predictive control approach; Optimization framework; Sampling time; Stochastic model predictive controls; Predictive control systems