Application Of Economic Model Predictive Control In Wind Power Generation System

With the development of industrial processes and the intensification of urbanization,more and more electricity is being generated from renewable sources.Wind energy has become an important source worldwide in order to product electrical energy.With the increasing installed capacity of wind power,the economical and safe operation of wind turbines,i.e.,mitigating fatigue damage to reduce maintenance costs,and extending wind turbine life are particularly important.In this paper,the variable-speed constant-frequency doubly-fed wind turbine is taken as the research model,and the subsystems model of the aerodynamic system,transmission system,tower,pitch system and generator are established.Considering the components that are easily damage of the wind turbine,the nonlinear model of wind power system is finally established.More focus on economic factors of wind turbines should be considered:maximizing the electric power generated from the wind,mitigating fatigue damage to reduce maintenance costs,i.e.,reducing the fatigue damage and wear of the pitch actuator,mitigating drive train transient load caused by torque fluctuation,and reducing the oscillations in shaft torsion to prevent the turbine components damage from occurring.A multi-objective and multi-constraints optimization problem is established,and an economic model predictive control(EMPC)strategy based on nonlinear model of wind power generation system is proposed.By adjusting the pitch angle and generator torque,the dynamic economic performance of the system is improved,and the economical and safe operation of the wind turbine under full working conditions is realized.In the process of wind power generation,wind speed changes,tower shadow effect,wind shear and other factors will cause frequent fluctuations in wind turbine output power,which poses greater challenges for safe and stable operation of large-scale wind power grid connection.For coping with this wind speed uncertainty/disturbance,a robust economic model predictive control(REMPC)should be constituted for wind turbine system application,taking this wind speed error as disturbances.For this purpose,the nominal wind turbine system model,which is not affected by the disturbances,is introduced to solve the original optimization problem,and its solution defines the nominal trajectory.A linear feedback control law and A robust control invariant set are designed through a LMI problem.Then a linear feedback controller is designed for the real wind turbine system to keep the real trajectory within the invariant sets around the nominal trajectory.Realize the robustness of wind turbine control,improve the grid-connected capacity of wind turbines,and reduce the impact on the grid.The robust feasibility of the algorithm is verified by theoretical proof.Finally,the effectiveness of the proposed REMPC control strategies is verified by the simulations based on a 5WM wind turbine system.The simulation results indicate that the proposed REMPC strategy is rather suitable for enhancing the economy of the wind turbine system,and the control objectives are successfully achieved in different conditions.Moreover,the proposed REMPC scheme can guarantee a robust performance under the extreme situation in which the wind speed changes dramatically.