To address the uncertainty of intermittent energy sources and improve the economic efficiency and operation performance of microgrids

a novel three-layer coupled microgrid dispatch model is proposed based on model predictive control. The model is optimized and solved using an improved dung beetle algorithm. First

by comprehensively considering time-sharing tariffs and pollution control costs

the model mitigates the impact of day-ahead uncertainties and performs online correction of the optimal control variables

thereby forming a new three-layer scheduling structure. Second

improvements are made to the traditional dung beetle algorithm in aspects such as population initialization

rolling behavior

and foraging behavior. The feasibility of the algorithm is verified by simulation

showing improved accuracy and convergence speed. Subsequently

the improved dung beetle algorithm is used to optimize the economic performance of the scheduling layer and determine the optimal control actions in rolling control. Finally

comparative analyses in terms of economic performance

rolling scheduling

and control effect demonstrate that the proposed model and algorithm can effectively enhance the environmental and economic performance of microgrids while improving system controllability and stability.