An Optimized Yaw Strategy Based On Wind Condition Prediction

Yaw system is an important part of wind turbines(WTs).Yaw strategy directly determines the utilization efficiency of wind turbine,and also affects the failure rate and service life of wind turbine.The yaw strategy of fixed yaw error Angle,fixed yaw delay time and fixed yaw Angle is generally adopted in active WTs.Under the frequent change of wind speed and direction,the phenomenon of "over-yaw" and "under-yaw"often appears in wind turbine,which reduces the utilization efficiency of wind energy and shortens the service life of yaw parts.To solve these problems,an optimal yaw strategy based on wind condition prediction is proposed in this paper.Firstly,the wind performance of the wind turbine under the existing yaw strategy is analyzed and the improvement target is proposed.Based on the SCADA historical data of WTs,the yaw times,yaw error Angle,yaw error duration and other indicators were counted.The performance of wind turbine in service is analyzed,and the problems of "over-yaw" and "under-yaw" that need to be solved under the existing yaw strategy are pointed out.A method for optimizing yaw strategy based on wind farm historical wind data is proposed.Secondly,the historical data of the wind farm are analyzed to obtain the optimal yaw error Angle.The data of historical wind speed and change Angle of wind direction in each month are statistical,and several curves of the corresponding relationship between them in different months are obtained,namely,wind state characteristic curves.On this basis,K-means algorithm is used to cluster the wind regime characteristic curves of each month,and a group of wind regime characteristic curves which can represent the wind regime characteristic of each month are obtained.The group of wind characteristic curves were used as the yaw error Angle parameters of the optimized yaw strategy of wind turbine.Thirdly,the optimal yaw delay parameter is determined with the objective of maximizing the comprehensive economic benefit.A comprehensive economic benefit of life cycle based on the sum of power generation,fatigue life loss of yaw bearing and power loss of yaw motor operation is proposed.Taking the wind characteristic curve as the allowable yaw error Angle parameter,the yaw results under different yaw time delay parameters are simulated.The optimal yaw delay parameters were determined by comparing the comprehensive economic benefits of life cycle.Finally,the wind regime prediction is applied to the optimized yaw strategy to predict the yaw error Angle parameters and respond to it.The CEEMDAN-LSTM prediction model is used to predict the future wind speed and direction.Based on the prediction results,the optimal yaw error Angle parameters are determined and whether yaw is started is judged.By using this strategy,the yaw policy parameters can be changed in real time under the wind conditions,and the wind turbine’s performance to the wind at the next moment can be evaluated in advance and reacted in time to reduce the occurrence probability of "over-yaw" and "under-yaw" problems.The example verification shows that the yaw optimization control strategy model can improve the accuracy of wind turbine,increase the power generation,reduce the loss of yaw power and life of wind turbine,and prolong the service life of yaw parts.