Abstract: With complete infrastructure and numerous energy storage devices， the industrial parks have the potential to participate in day-ahead peak-load regulation and day-ahead real-time tracking control. There may be many factory buildings in the industrial park， and the types of production devices are different. By adjusting the traditional control method of tie-line power real-time controlled by gas turbine power output， it is difficult to take into account the relationship between various interest subjects under different assessment schemes. Therefore， from the perspective of the park， by using of disposable energy storage devices in the park to implement real-time control， can effectively coordinate the economic interests of all parties， and obtain good control effect. Mathematics model of gas storage， cooling storage， electricity storage and some other storage devices and the thermal model of the room are established in this paper. In response to the tie-line power fluctuations caused by forecast error of the photovoltaic， wind power and other clean energy， and at the same time to ensure a variety of energy storage devices to meet operating constraints and energy balance constraints of cooling， heat， electricity， gas load， this paper proposes an intra-day rolling optimization correction strategy on the basis of MPC. According to the results of day-ahead economic operation， operators can adjust the operating power of energy storage devices to eliminate the influence of uncertain factors in the microgrid， and to ensure the implementation of scheduling results of tracking. Taking an industrial park for battery production in southern China as an example， the effectiveness of the proposed optimization model and algorithm is verified by numerical analysis.

Key words: micro-grid, model predictive control（MPC）, intra-day real-time control, thermal load, tie-line assessment scheme