An Intelligent Approach to Flight Data Processing Basedon a Fuzzy Kohonen Neural Network
Zhang Zong-lin · Systems engineering and electronics · 2002
In order to make good use of the data in the flight data recorder and to evaluate the pilot's wingmanship, it is necessary to divide the data into some pieces by action. This paper is concerned with fuzzy Kohonon neural network, a dividing model for flight data, whose design is based on the presence of standard flight actions. First the degree of dependence vector obtained by applying the experts' experience is input as fuzzy vector, and then the pattern of data frame is classified automatically through this fuzzy Kohonen. Finally, the flight data classified according to time is divided into two segments-the data of standard flight and the data of nonstandard flight, which will be used as the basis of automatic flight data dentification. Proved by a large number of actual flight data, this processing method is accurate and effective.