Mechanism Analysis Of Wind Field Characteristics And Research On Output Power Optimization Of Wind Turbines

As a clean and renewable energy with no any pollution to the environment,the proportion of wind energy in primary energy is increasing day by day.How to utilize wind energy resource more efficiently and maximize power generation greatly has become a hot topic in wind power research.However,due to the strong randomness of wind and the increasing installed capacity of wind turbines,the control difficulty of wind turbines is much harder than it used to be.Nowadays,the difficulty in the development of wind power generation is that the optimization research of wind turbine performance depends on the optimization of multi-dimensional control parameters,but the improvement of traditional control methods is difficult to achieve the desired goal.Based on the large data of wind farm operation,the output power prediction model,power optimization and power evaluation of wind turbines are studied in this paper.The main contents include:Through the factors that influence the wind energy resource capture characteristics of wind turbines,including the basic parameters of wind resources,topography,array distribution of generating units and wake,the wind field characteristics are analyzed.Based on large wind field data,BP neural network,support vector machine an d long short-term memory network power models of wind turbines are established.Through verification,analysis and comparison,the results show that the long short-term memory network power prediction model can predict the output power of wind turbines more accurately.A power optimization model based on the combination of wind direction prediction of support vector machine optimized by genetic algorithm and yaw strategy by wind speed subsection control is proposed.The wind direction prediction model is validated under various wind directions.The results of the wind direction prediction errors are less than 5.It shows that the model can predict the wind direction better,and the prediction performance is better than that of the n eural network.Based on the large data of wind turbine operation,the yaw error angle of wind turbine is analyzed,and the yaw control strategy of wind speed subsection is used to optimize the output power of wind turbines.The verification results show that the output power of the unit has been improved to a certain extent.Based on the large data of a wind farm and the output power characteristics of wind turbines,an optimal evaluation model for the output power of wind turbines is designed,and the optimization of the output power of wind turbines is evaluated and analyzed.The calculation shows that the power increase ratio of the wind turbines is 3.36%.It shows that the optimization method can effectively improve the absorption efficiency of wind energy and the output power of the wind turbines.