Model Predictive Control for Cooperative Energy Management at City-District Level

This work presents the design and application of Model Predictive Control (MPC) for energy management at city-district level. MPC is applied to find an optimal combination of buildings with dispatchable heating systems in order to compensate schedule and forecast deviations of demand and generation. The method combines global objectives, such as integration of renewabies and tracking of a negotiated schedule, together with local objectives, such as individual preferences regarding frequency of switching and thermal storage limitations. Here we present the MPC design with the corresponding system model, objective function and constraints. Based on a test scenario, we assess performance indicators, such as deviation remainder, number of switching events and effects on the individual schedules, and we demonstrate the eligibility of MPC for the application at aggregator level for energy management in city districts.