Stochastic Dynamic Economic Dispatch for Grids with Significant Wind Using Mixed Gaussian Distribution

Generation scheduling is becoming a challenge in power grids with high penetration of renewable energy sources due to their stochastic nature. In this paper, an efficient stochastic multi-period dynamic economic dispatch (DED) model is presented. It allocates optimally generation levels among the online thermal generators in a way that maximizes the utilization of wind resources. In order to accommodate wind uncertainty, the conditional probability distribution function of the wind power output given the forecast level is used. Mixed Gaussian (MG) distribution is utilized for wind uncertainty characterization as it greatly enhances computational speed and accuracy. The statistical analysis shows the advantages of MG function over other distributions presented in the literature. Simulation results of a system with thermal and wind power plants show the merits of the proposed MG-based stochastic DED methodology.