Abstract: Combined cooling, heating, and power（CCHP） systems have the characteristics of energy saving, environmental protection, economical efficiency, and so on, which bring it a good prospect and great application value. In order to track the fluctuation of the renewable energy and load in real time, a dynamic optimal scheduling strategy based on the model predictive control was proposed in this paper. This scheduling strategy linearizes the efficiency curves of the devices, takes the day-ahead dispatch plan as the reference, establishes the prediction model for the daily renewable energy and load, and then obtains smooth output results of the devices based on the multi-step rolling optimization. Through the case study, the model of linearizing the efficiency curve of the gas turbine can not only reflect the actual operation conditions of the combined cooling heating and power micro-grid, but also meets the calculation of online scheduling. In the case, the rolling period in the case are 4 hours, which can reach a stable and a quick response speed. The proposed model is able to depress the effect of uncertainties to the system dispatching, and ensures the system running with high economic efficiency and security.