Processing And Analysis Of Flight Training Data

In recent years,the rapid development of China's civil aviation industry has led to the rapid rise of China's civil aviation education and training.The demand of pilots is increasing,and the training tasks are increasing.At the same time,the application of data mining technology in the field of civil aviation safety analysis,anomaly detection,fault detection,state prediction and other work is also increasing.Data mining technology can find the ability of hidden flight data behind some events that are difficult to be found by conventional data analysis.The application of data mining method in flight data increases the theory and method of machine learning on the basis of traditional flight data analysis method.Based on big data and knowledge discovery,it integrates knowledge discovery,data mining and mathematical modeling.The application of data mining algorithm in flight data not only retains the advantages of traditional methods,but also can play the characteristics of data mining methods.The purpose of this thesis is to use the related algorithms of data mining to carry out data mining related work on flight training data,which is of great significance to flight safety and risk aversion,and plays a very important and positive role in flight training management.It mainly involves the use of support vector machine model to predict the engine speed parameters in the flight training data,and the results are compared with the original data to test the practicability and effectiveness of the method.Two mainstream correlation algorithms,Apriori and FP growth,are used to analyze the relevant parameters of misoperation events in flight training data,and the results are compared to explore the correlation between events and data,and to test the practical effectiveness of the method.By using the methods of statistical analysis and data analysis,the flight data are analyzed and sorted out to provide decision-making information for safety management and training management personnel.