Wind Speed Calibration Based On Field Data And Correction Of Wind Turbine Power Curve

As the wind is random and unstable,this directly leads to the wind turbine output volatility,coupled with the wind wheel after the anemometer wind deviation,to further exacerbate the difficulty of wind turbine control and handling of volatility,so the need for After the cabin anemometer calibration.In order to establish the calibrated model of windmill anemometer,during the operation of wind turbine,the wind speed of the wind wheel is measured by using lidar to get the actual wind speed and the wind speed on the wind wheel,The corresponding data are time-correlated,and the mapping model of the two is obtained by the correlation algorithm,namely the calibration model.The wind speed calibration model is applied to the actual operating power curve of the wind turbine,and the correlation analysis of the annual power generation is carried out.The main contents are as follows:(1)Based on the lidar data,80% of all the data were used to establish the linear model of the calibrated cup anemometer,the BP neural network model and the BP neural network model optimized by the genetic algorithm.The remaining 20% of the data was used as the validation of the model.The results show that the BP neural network model based on genetic algorithm optimization is more suitable for wind speed calibration model,and the RMSE of three calibration models is 0.5372,0.2323 and 0.1646,respectively.(2)Wind turbine generator 's actual operating power curve.Since the wind speed of the cup anemometer is different from the front wind speed of the wind wheel,the actual power curve of the unit is deviated according to the wind speed.Since the lidar is testing the front wind speed of the wind wheel,the actual wind power curve of the unit is obtained according to the wind speed of the lidar.The power curve measured by wind cup anemometer is corrected by wind speed correction model,and the corrected power curve is obtained.The ann ual power generation of the unit is estimated.The estimated deviation is 1.22%,while the unmodified model is 9.87%,which indicates that wind speed correction model can be used to revise the power curve of wind turbine.