Incremental Support Vector Machine Learning Method for Aircraft Event Recognition

Event indentification of hard landing is the hot spot in civil aviation safety research. In this paper, a incremental model to indentify aircraft land status of civil aircraft is presented to support fault diagnosis and structure maintenance. In previous reserach, traditional artificial neural network is used as a classifier for event detection from certain landing parameters. This paper develop a further recognition model by introducing support vector method, also an incremental algorithm is proposed to solve the problem of on line sample array, and sensitivity and specificity are employed to show the model performance comparing to existing model. Finally, advantage of this method is analysed, and the aspects of each model are given.