An affine arithmetic-model predictive control approach for optimal economic dispatch of combined heat and power microgrids

Abstract This paper presents a novel approach for the optimal economic dispatch of Combined Heat and Power (CHP) Microgrids (MGs), which incorporates an Affine Arithmetic-based Economic Dispatch (AAED) problem into a Model Predictive Control (MPC) framework. The proposed algorithm solves an AAED problem each Δ t minutes (e.g. 15 min) with time steps of Δ t minutes over a time horizon T (e.g. 24 h). It uses the available forecast and the current state of the system to provide the schedule and the affine forms that represent the operation intervals of the generators and Energy Storage Systems (ESS) for the next time interval [ t , t + Δ t ] . Online set points for generators and ESS are then obtained by computing the noise symbols values of the affine forms, based on the most updated information of electricity and heat demands and available renewable energy power. A theoretical CHP-based MG, comprising PVs, a gas boiler, a CHP unit, a battery, and a thermal tank, is used to assess the performance of the AA-MPC approach in both connected and isolated operation modes. The method is also compared with a deterministic MPC approach. Results show the ability of the method to better address forecasting errors, resulting in more cost-effective solutions without considerably affecting the computation performance.