Optimal dispatching method for integrated energy system based on robust economic model predictive control considering source-load power interval prediction

Effective source-load prediction and reasonable dispatching are crucial to realize the economic and reliable operations of integrated energy systems (IESs). They can overcome the challenges introduced by the uncertainties of new energies and various types of loads in the IES. Accordingly, a robust optimal dispatching method for the IES based on a robust economic model predictive control (REMPC) strategy considering source-load power interval prediction is proposed. First, an operation model of the IES is established, and an interval prediction model based on the bidirectional long short-term memory network optimized by beetle antenna search and bootstrap is formulated and applied to predict the photovoltaic power and the cooling, heating, and electrical loads. Then, an optimal dispatching scheme based on REMPC is devised for the IES. The source-load interval prediction results are used to improve the robustness of the REPMC and reduce the influence of source-load uncertainties on dispatching. An actual IES case is selected to conduct simulations; the results show that compared with other prediction techniques, the proposed method has higher prediction interval coverage probability and prediction interval normalized averaged width. Moreover, the operational cost of the IES is decreased by the REMPC strategy. With the devised dispatching scheme, the ability of the IES to handle the dispatching risk caused by prediction errors is enhanced. Improved dispatching robustness and operational economy are also achieved.