Time Series Modeling Of Flight Data And Identification Of Flight Action Subjects

The objective and efficient quality evaluation of flight training can provide important basis for improving the safety and quality of flight training.The evaluation of flight training quality is usually based on the evaluation of the completion quality of flight training subjects.Therefore,the automatic and accurate identification of the corresponding training subjects from the recorded flight data is an important basis for flight parameter-driven flight training quality evaluation.At present,the research on the identification of flight movements and subjects in flight parameter data is mainly focused on the identification of flight movements.This paper aims to design an automatic recognition algorithm of flight training action subject based on time series modeling representation based on the analysis of flight parameters.Based on the pre-processing of flight data,the time series of flight data was modeled based on the piecewise linear representation method,and the maneuvering flight segments were accurately divided by setting reasonable parameter thresholds.Secondly,through searching authoritative data and consulting expert experience,the flight movements are classified in detail,the main parameters that can reflect the flight movements are determined and the change threshold of parameters is reasonably set,and the flight movements rule base is constructed.Thirdly,on the basis of flight action recognition,a flight training subject description method based on flight action sequence features and a flight training subject recognition algorithm based on finite state automata are designed.Finally,experiments are carried out to verify the practicability and effectiveness of the algorithm used in this paper to realize automatic identification of flight action subjects.The experimental results show that the proposed method and the designed algorithm can accurately identify the flight movements and subjects from the flight parameter sequence,and the recognition rate is high,and the method has good practicability and accuracy.However,the relevant parameter thresholds are derived from relevant flight standards and expert experience,which is somewhat subjective and still has room for further optimization.