Atmospheric turbulence identification and quantification method based on airborne data

The invention discloses an atmospheric turbulence identification and quantification method based on airborne data. The method comprises the following steps: acquiring airborne QAR data, and dividing the airborne QAR data into a training set and a test set after processing; performing interpolation processing on the EDR data; a random forest model is constructed, and then the obtained QAR data and the corresponding interpolated EDR data are used for training; and outputting a turbulence detection result by the trained random forest model. The method has good model accuracy, and achieves a good effect in data detection independent of training data. The process cost is low and the system is scientific. According to the method, the turbulence anomaly information is directly obtained from the airborne QAR data, the correlation between the airborne QAR data and the turbulence anomaly information is autonomously learned by a computer, the turbulence anomaly information can be accurately obtained without manually set