The Kriging model based scenario generation and its application in the energy internet with high wind penetration

In this paper, a scenario generation method based on Kriging model is proposed to describe the wind uncertainty in the Energy Internet. It uses the “forecast bin” method to generate a large number of scenario samples. Kriging model is used to estimate the operational cost of the corresponding scenario. Then, an importance sampling method based on the Kriging model is used to select the scenario samples. This method considered the wind power uncertainty and the influence of uncertainty on the objective function. Finally, an Energy Internet with ten units is used to verify the effectiveness of the proposed approach. Compared with the probability analysis method, it can see that the proposed method has better economy, stability and reliability on the problem of Energy Internet.