Research Of Model Predictive Control Strategy For Offshore Wind Farms Integration

In the trend of energy shortage and the growing environmental pollution,carrying out offshore wind power actively can effectively alleviate the energy crisis,and protect the ecological environment.With the maturity of large offshore wind power technology,its grid-connected control strategy has become the current research hotspot.In this thesis,the model predictive control strategy is applied to the offshore wind power VSC-HVDC grid-connected control to solve the problem of difficulty in setting PI parameters in traditional voltage orientation control,complicated control structure and difficult to carry out multi-objective optimization.Firstly,the topology of offshore wind power grid-connected system is introduced,and the mathematical model of grid-connected inverter in different coordinate system is analyzed in detail.In order to solve the complex decoupling process and cumbersome PI tuning in traditional voltage orientation control,the wind power grid model is proposed to predict the current control strategy.This method combines the discrete mathematical model of the grid-connected inverter,constructs the value function through the current error,and uses the value function as the optimization target to predict the switch state of the grid-connected inverter in the future.In order to avoid the calculation of time delay and realize multi-objective optimization,the delay compensation and weight coefficient are introduced.The system output of all the switching function combinations is calculated by the traversal method,and the optimal switching combination is proposed to trigger the grid-connected inverter.Theoretical analysis shows that the model predictive control principle is simple,and it can reduce the cumbersome parameter design process.Secondly,in order to reduce the delay problem caused by rolling optimization in traditional model predictive control and improve the accuracy of system control,an improved model is proposed to predict power control.Through the complex power to construct value function,using space voltage vector to obtain the best combination of voltage vector,and according to its corresponding optimal switch combination to trigger the grid inverter.Compared with the traditional model predictive power control,the improved method proposed in this thesis can effectively reduce the computational complexity and improve the control precision.Finally,the simulation model of wind power grid-connected system is established in MATLAB/Simulink.The model predictive control and the traditional PI control method are used to control the grid.The simulations of the wind turbine power fluctuation and grid failure are carried out.The proposed model predictive control and its improved method can been correctness and rationality applied to the grid system of VSC-HVDC offshore wind farm.