Coordinated Control of Wind Farm Power Prediction Based on PSO-MPC Model

Abstract

For wind farms, traditional control strategies cause a large amount of energy loss when dealing with wake effect, and traditional control strategies require all units to participate in, resulting in unnecessary unit actions. In this paper, a PSO-MPC control strategy is proposed. Based on the MPC (Model Predictive Control) model, the future outputs of wind farms are predicted according to the historical information and future inputs. Based on Jensen's wake model, a dynamic wind farm power model considering wake delay was established, and the output power of wind farm was connected with the axial flow factor controlling each unit. PSO (Particle swarm Optimization) algorithm was used to optimize the performance indexes in the prediction time domain by using predictive control information, and the control values of each fan were obtained. The simulation results show that compared with traditional wind farm control, the proposed control algorithm can effectively suppress the power fluctuation and improve the power tracking accuracy according to the prediction information, and control the speed of each unit in advance through the optimization results, reduce the unnecessary action of the unit, and thus reduce the unit loss.