A fuzzy logic dispatching method based on model predictive control for adaptively addressing uncertain operation scenarios in multi-energy systems

The integrated energy system (IES) effectively combines wind, solar, natural gas, and other resources to efficiently meet the diverse energy demands of cooling, heat, and electricity. Given the inherent randomness and intermittency of multi-energy sources and loads, achieving optimal conversion and allocation between different energy flows during IES operation remains a significant challenge. A sophisticated energy management approach that achieves long-term optimization under operating uncertainties is imperative. In this paper, a two-stage IES real-time dispatching method based on fuzzy logic and model predictive control (MPC) framework is proposed. In the day-ahead stage, utilizing the potential operating scenario constructed by the interval prediction boundary curve, the knowledge parameters of evolutionary fuzzy inference system (EFIS) are optimized using economic cost as the objective. This day-ahead stage can obtain the set of EFIS models tailored to different scenarios for IES operation dispatching. In the intra-day stage, based on the rolling prediction information in MPC framework, the real-time decision model is dynamically alternated from the pre-obtained EFIS models to enhance adaptability to varying operating patterns under multi-uncertainties. Compared with the three benchmark methods, the proposed method can save 14.1 %, 7.1 % and 5.3 % of the system operating costs, respectively. The simulation results demonstrate that the proposed method can effectively address the dispatching of FIS among multiple energy flows, thereby enhancing the operational efficiency of the system under multi-uncertainties.