A reinforcement learning-based optimal scheduling method is proposed for the uncertainty of renewable energy and load in integrated energy systems. Firstly, the basic principles of the reinforcement learning method are explained; then, a reinforcement learning-based optimal scheduling model for integrated energy systems is proposed, and the state space, action space and reward function in it are designed; then, a model solving process based on the asynchronous dominance strategy gradient algorithm is designed, finally, the simulation validation shows that the proposed method can self-adapt to source and load uncertainties and achieve The proposed method is shown to be adaptive to source and load uncertainties and achieves similar optimization results to those of traditional mathematical planning methods.