Decentralized coordination of the operation of residential heating units

This work presents an approach to decentralized coordination of electro-thermal heating systems, like Combined Heat and Power systems (CHP), Heat Pumps (HP), or Electric Heating systems (EH), each equipped with thermal storage, within a group or cluster of residential buildings. The flexibility of these thermal systems can be exploited for coordinating the heating units for a higher-level objective, like balancing power fluctuations of Renewable Energy Sources (RES). In order to assess the decentralized coordination algorithm (DCA) presented in this paper, its performance is compared with that of a centralized approach. In the centralized approach, a Mixed Integer Program (MIP) is formulated and solved, obtaining the optimal schedules for each heating system within the building cluster. In the decentralized approach, an MIP is first executed for each heating system, providing a set of schedules, which all satisfy the local constraints as well as local objectives of the building. Next, one schedule per set is selected in a way that the combination of the selected schedules facilitates the higher-level objective. The results presented in this work show that the proposed DCA can in principle balance power fluctuations as global objective, while still enabling local optimization. However, the results appear to be strongly dependent on the flexibility of the local optimization.