Abstract:In order to build a ground flight safety situation monitoring system, an aircraft status prediction method based on SSA-TSVR was proposed to analyze the feature importance of real flight data, aiming at the problem that real-time monitoring of aircraft status data could not be carried out due to abnormal data transmission during the transmission process of aircraft status data to the ground. The important parameters that are closely related to the aircraft state parameters to be predicted are screened, and the importance relationship between the parameters to be predicted and the flight data is obtained. The twin support vector regression algorithm was used to build a prediction model to predict the missing key flight state parameters. The twin support vector regression model is optimized by using the flying squirrel search algorithm, and the optimal kernel function is selected according to different prediction objects to improve the prediction accuracy of the model. With flight altitude and speed as prediction objects, the prediction model realizes the accurate prediction of aircraft state by using incomplete flight data, which is of great significance for aircraft flight state monitoring.
