Research On Influence Of Wind Turbine Wake And Wind Farm Power Improvement

Wind power generation is one of the important ways to solve the fossil energy crisis.The centralized construction and development of wind farms are conducive to reducing costs and saving land resources.But concentrated construction also leads to aerodynamic coupling between wind turbines,namely the wake effect.Due to the wake effect,the traditional maximum power tracking control strategy cannot maximize the wind farm production capacity.Consider the relationship between the wake effect and the output power of the wind turbine.The output power of some wind turbines can be limited.Reduce the wake effect on the rear wind turbine.Thereby increasing the production capacity of wind farms.Thesis research focuses on wake models,wind turbine control strategies and power distribution in wind farms.The main research contents are as follows:Firstly,the characteristics of the three wake models are analyzed.The theoretical knowledge of artificial neural network is introduced.The BP neural network was used to fit and correct the wake model.The BP neural network wake model,the existing wake model and the data of a wind farm are compared and analyzed.The results show that the wake radius and radial wind speed of the BP neural network wake model are closer to the actual measured data.Secondly,a power limiting control model of wind turbines is established based on the deep Q network.The input of the deep Q network is the reference output power and its own operating state,and the output is the pitch angle,yaw and rotor speed.By setting the reference output power for the wind turbine,the adjustment range of its pitch angle and rotor speed can be increased.Adding yaw can make the control more flexible.In wind farm control,the adjustment of yaw can change the influence of wake effect on the rear wind turbines.A deep Q-network power-limiting control model is established in Matlab/Simulink.The simulation results show that the deep Q network power limiting control model can effectively control the output power of the wind farm to the reference output power.Finally,in view of the wind farm energy loss caused by the wake effect,the dual-delay depth deterministic strategy is used to distribute the power of the wind farm,and it is combined with the BP neural network wake model and the wind turbine power limit control strategy to build a Wind farm power distribution optimization model.The simulation results show that,compared with the traditional maximum power tracking control strategy,the control strategy in this paper reduces the influence of the wake effect on the rear wind turbines and improves the overall output power of the wind farm.