Study On Fatigue Characteristics And Flight Quality Evaluation Of Flying Students Based On Physiological Index Measurement

In recent years,China’s civil aviation industry has ushered in a fast-growing economy and society that has ushered in a golden period of development and has achieved fruitful results in all fields.At the same time,the demand for trainee pilot has increased significantly.As the largest flying training school in the world,the Civil Aviation Flying Academy of China has seen a significant increase in the training volume of trainee pilot,which will inevitably lead to a decrease in the relevant flight safety factor.For the purpose of improving the flight safety of flight training,this research will be conducted from the following two aspects.The first part studies the fatigue characteristics of trainee pilot.Design experiments to obtain trainee pilots’ heart rate data as fatigue analysis indicators.In the course of the experiment,determine yawning,flow tears,and decline in response ability and operation ability as evaluation indicators for fatigue characterization;analyze the characteristics of trainee pilots’ heart rate data.In order to overcome the difference in the physical and mental rate of different flight students,the heart rate interval change rate is used as an analysis index,and the ROC curve can be used to determine the role of the threshold value.The ROC curve is plotted combining the rate of change of heart rate interval and fatigue status state quantity.Through the analysis of ROC curve,the critical rate of heart rate interval of fatigue is determined for the trainee pilot,and then the fatigue time point is determined.The use of binary logit regression models can be used to determine the weight of the independent variable,and to select fatigue as the secondary variable,whether to select time,whether it is simulated or true flight(difficult and easy to operate),whether it is a sophomore flight student or a senior flight student(high and low proficiency)and rate of heart rate variability were used as four independent variables.Then a binary Logit regression model has been established,the weights of the four independent variables were obtained,and the degree of influence on fatigue was determined.The rate of change of fixed heart rate interval is 3%.The fatigue characteristics curve of trainee pilot is plotted.The fatigue characteristics of flight students are analyzed and related suggestions are made.The second part establishes a flight quality assessment model.The experiment was designed to obtain the flight parameter data and physiological signal data of the trainee pilot.Five flight parameter data,such as descent rate,yaw,vertical velocity variation,degree of flap and landing speed,and two physiological signal data of heart rate and skin temperature were selected as flight quality assessments index.after data processing,each student needs to be scored by experts during each flight phase,scoring is divided into three levels: excellent,good and bad.A combination of five groups of evaluation indicators and expert scoring in 20 simulation flights was randomly selected as the test set of the support vector machine(SVM).The remaining 15 groups were used as training sets.Because the scores are divided into three categories,SVM multiple classifiers are selected to classify them.The results show that the flight quality assessment model constructed by the support vector machine is accurate and feasible.With the newly established assessment model,flight quality assessments and comparisons are made between the same flight phase before and after the fatigue of the flight student(the first five-edge approach landing phase and the third five-sided approach landing phase)to verify the effect of fatigue on flight quality.