Abstract: In order to deal with the risk brought by the uncertainty of wind power to the safe operation of power grid, the interval forecasting method has received extensive attention in recent years. However, the existing researches mainly focus on the forecasting method for the single wind farm, and negligibly focus on the methods for regional wind farms. In view of the above problem, the dynamic R-vine Copula model is established in this paper, and the ultra-short-term interval forecasting method for regional wind farms is proposed. Firstly, the framework of the ultra-short-term interval forecasting method is described in detail. Secondly, the method of establishing the joint probability distribution of the forecast power and total forecast error for multiple wind farms based on the R-vine Copula model is briefly introduced. Then, the dynamic R-vine Copula model is established in three steps as follows: the dynamic marginal distribution model is established based on the ARIMA-GARCH model; the DCC and Patton models are introduced to establish the dynamic Pair Copula model; on this basis, the calculation method for the topology structure of the dynamic R-vine Copula is proposed. Finally, the case study is carried out based on the data from 9 wind farms in Xinjiang Autonomous Region (Northeast of China) for one year. The simulation results verify the effectiveness of the proposed model, and show that the forecasting results of the proposed model have good reliability, sharpness and skill score indexes. The research can provide a reference for the topic of the prediction of ultra-short-term interval forecasting for regional wind farms.