According to the non-stationary characteristics of wind power, a novel wind power combined multi-step prediction model based on differential auto regressive moving average model in variational mode decomposition and time series analysis is proposed. Firstly, the wind power sequence is subjected to variational mode decomposition to reduce the non-stationary characteristics of the wind power sequence. Secondly, the ARIMA model is established for each component. Finally, the component prediction results are superimposed to obtain the final wind power prediction value. The experimental results show that the proposed combined prediction model has higher prediction accuracy.