A calibration method for pre fit data screening is proposed to address the fitting errors caused by fluctuations in the atmospheric environment during the observation period,and a Langley calibration is carried out for the radiometer using the observation data of the solar radiometer at the Lijiang Plateau from March to May 2021.By excluding the data with excessive fluctuation during the observation period,the reliability of the data participating in the calibration is improved,so as to improve the accuracy and stability of the calibration results.The results show that the Langley fitting linearity after data screening is significantly improved,the dispersion of calibration values is greatly reduced,with an average reduction of 70.4% in the distribution interval and 65.4% in the standard deviation of the calibration values of each channel.For the obtained multi day calibration results,the standard deviation of the calibration values of each band is reduced by 28.7%～67.6% by using the fitting average method and the relative error between the results and the factory calibration values of the instrument is less than 3%,which proves that the method is feasible and effective.