| With the rapid development of the civil aviation transportation industry,the International Civil Aviation Organization proposed a "Multi-crew Pilot License" in 2006,referred to as MPL.The system provides a new direction for pilot training.The use of simulators has increased significantly,accounting for about two-thirds of the total training time,shortening the link time between pilots and airlines,accelerating the growth of pilots,and ensuring civil aviation operations safety.At present,in the training of MPL simulator,the flying quality is mainly evaluated by "satisfaction/dissatisfaction",which has some problems such as strong subjectivity and rough evaluation standards.In this paper,the take-off stage of MPL simulator training is taken as the research content.Based on the research foundation of using QAR data used to guide pilot simulator training,the parameter evaluation interval of the take-off stage is quantitatively analyzed and divided more accurately,and the evaluation system is constructed to assist and supplement the flight quality evaluation of the current simulator.First of all,the screening and processing of QAR take-off data.The key parameters of normal QAR take-off of an airline company were screened out by Matlab.Due to the lack of data,collection frequency,multi-dimension and other reasons,the collected data were preprocessed by correction,filtering and dimensionless processing.Secondly,establish an evaluation index system.According to the original related variables,the common factors of the influencing variables were found out to explain the original data.Factor analysis method was used to establish the flight quality evaluation index system of MPL simulator training in the take-off stage from the take-off speed control,pitch control,roll control and response control.Then,the evaluation standard interval of MPL simulator training take-off phase is constructed.Based on the characteristics of small,medium and large take-off parameters of QAR monitoring standard value and the reference value of aircraft manual,the modified quartile model is an important statistical feature to describe the trend of data set.Combined with the statistical characteristics of data bias to lower quantile(Q1),median quantile(Q2)and upper quantile(Q3),the evaluation interval of key parameters in take-off phase is refined.Through the optimized evaluation criteria,each key take-off parameter is subdivided into four categories: excellent,good,medium and poor.The refined scoring table makes the scoring difference small,objective and more accurate.It can be better used as a fine auxiliary and supplement to the current flight quality evaluation of MPL simulator,improve the evaluation standard of simulator takeoff training,and ensure the pilot’s safe operation ability.Finally,based on the evaluation standard interval table,the evaluation system of MPL simulator training take-off stage is developed by using Java programming language to assist the evaluation.The system includes the assessment of the take-off stage of the simulator,the instructor’s guidance and the query of the pilot’s performance.It can improve the objectivity and impartiality of teacher evaluation,so that pilots can correct their operational skills more accurately according to the evaluation results and opinions.To better explore the standardization and normalization of the assessment during the take-off training stage of the simulator,and promote the consistency of training and teaching of MPL simulator. |