A novel model predictive control strategy for multi-time scale optimal scheduling of integrated energy system
Keyong Hu,Ben Wang,Shihua Cao,Wenjuan Li,Lidong Wang
DOI: https://doi.org/10.1016/j.egyr.2022.05.184
IF: 5.2
2022-01-01
Energy Reports
Abstract:Model predictive control is one of the key technologies to realize the multi-time scale optimal scheduling of integrated energy system. However, the traditional centralized model predictive control has the high model order, a large amount of online calculation and is not easy to expand, it is not suitable for the optimal scheduling of an integrated energy system with many distributed units. In this paper, a multi-time scale optimal scheduling method of integrated energy system based on distributed model predictive control is proposed, which realizes the flexible scheduling of the integrated energy system through the coordination and cooperation of various subsystems. Firstly, the detailed models of various generation equipment are established according to the four energy forms of cold, heat, electricity and gas. Then, a multi-time scale optimal scheduling is divided based on three scales: day-ahead long time scale scheduling in 1 h, intra-day predictive control in 15 min and real-time adjustment in 5 min. Next, during the day-ahead scheduling and intra-day rolling optimization, we establish an optimization model based on the best economic operation of the system, the daily operating cost of the system and the minimum penalty cost of start-up and shutdown. During the real-time adjustment, a distributed model predictive control method is proposed to decompose the overall optimization problem of the integrated energy system. Each subsystem estimates the state according to the input sequence of other subsystems at the previous time and optimizes its own performance index. Finally, the case shows that the model predictive control strategy proposed in this paper can increase the speed of optimal operation by about 15% and reduce the cost of optimal operation by about 3.8% compared with the traditional centralized model predictive control method, which not only improves the control performance of system operation, but also improves the economy of system operation. (C) 2022 The Author(s). Published by Elsevier Ltd.