Abstract: With the widespread application of renewable energy, the operational complexity and uncertainty of integrated energy systems have increased significantly. Traditional scheduling methods are insufficient to efficiently address issues such as multi-energy flows and strong coupling. Therefore, an integrated energy system model encompassing photovoltaic panels, wind turbines, hydrogen storage tanks, batteries, heat pumps, and fuel cells was developed. By combining the deep deterministic policy gradient (DDPG) algorithm, the energy scheduling optimization under uncertain conditions was achieved. Results show that the DDPG algorithm can effectively solve the problem of multi-energy flow coordinated scheduling, achieving supply-demand balance and efficient utilization of electricity, heat, cooling, and hydrogen under different load demands and energy supply conditions. Moreover, after training, the system exhibits excellent scheduling performance and adaptability when confronted with diverse uncertain scenarios, enabling flexible responses to energy fluctuations.