<p>In the integrated energy systems (IESs), multiple energy sources are coupled, and their spatiotemporal characteristics are different, making the optimal scheduling of the IES extremely difficult. Considering the impact of the randomness of wind power and photovoltaic output on the scheduling plan, an optimal scheduling method of day-ahead, intra-day, and real-time correction for IES is proposed. Firstly, random scenarios of wind power and photovoltaic output are generated based on kernel density estimation and copula function. Secondly, under the optimal scenario, the day-ahead optimal scheduling model is established with the lowest total operating cost of IES as the objective function. For intra-day scheduling, the objective function is to minimize the sum of penalty costs for wind and photovoltaic power abandonment, energy storage equipment, each equipment power change, and the change of power supply. Moreover, considering the difference in response speed of cooling, heating, and power, the power-type energy storage is used to realize short-time power dispatching, and the optimization model of real-time correction is established. Finally, the improved coati optimization algorithm (COA) is used to solve the problem. The simulation and experimental results validate the effectiveness and feasibility of the proposed strategy.</p>