At present, new energy substitution on the one hand helps to realize the carbon neutral goal, on the other hand, the uncertainty of wind and solar output brings great challenges to the stable operation of China's new power system. Meanwhile the coupling relationship between electric power and thermal energy becomes closer. Based on this, a data-driven distributional robust optimization scheduling model for integrated electric and thermal energy systems is established. Firstly, on the basis of considering the uncertainty of new energy output, the initial scenario is obtained by using Latin hypercubic sampling based on the original data, and the elbow method is utilized to determine the number of scenario clusters for the K-means algorithm, so as to obtain the typical scenario. Secondly, a two-stage distribution robust optimization scheduling model is constructed on the basis of the historical data of wind and light outputs, and a comprehensive norm-1 and norm-∞ constraints on the uncertainty probability distribution confidence set is integrated; finally, the CCG algorithm is used for the solving.