The study on hard landing prediction model with optimized parameter SVM method

Flight safety is an extremely important area of research. Hard landing, as one of seriously dangerous events during the flight landing phase, is an important issue to be taken into consideration in terms of flight safety, but has not been fully researched and the advance warning about it is almost blank despite there have been some researches on the landing safety problems. The prediction model of hard landing is proposed using support vector machine (SVM) method in this paper. Firstly, the influence factors of landing safety were explored and the relevant flight data were collected. Subsequently, flight data were preprocessed and analyzed on the basis of correlation analysis and factor analysis. Therefore, the method of slicing the flight data by flight height was put forward to expanse the sample size for prediction and the SVM model was established to train the flight data sample. According to the results, the optimal SVM model was adopted and demonstrated effectively.