Work Fatigue Monitoring Of Rail Transit Train Drivers Based On PPG Technology

The number of operating cities and lines of urban rail transit is increasing rapidly.With the continuous improvement of the network,the subway passenger flow is increasing year by year,and the demand for talents is increasing.As a key link in the operation and production of urban rail transit,electric train drivers bear more responsibilities than in the past.The fatigue of their work cannot be ignored.By analyzing existing methods of driver fatigue detection,it is found that the accuracy of fatigue detection based on physiological signals and body reaction information is high after decades of experiments.However,it cannot be applied to daily work.Because the process of signal acquisition is complicated,the acquisition equipment is required to be high,and the promotion cost is high,which has a great influence on the normal driving of the subject during the experiment.At the same time,the physiological signals mentioned above are complex,so it is usually impossible to quickly identify driver fatigue by analyzing the signals after the experiment.In order to realize the goal of real-time monitoring of driving fatigue,meet the market demand,consider the cost and achieve economic benefits,this study designs a portable optoelectronic plethysmography device model suitable for train drivers,involving the module composition and operation process of the monitoring system.Firstly,based on the analysis of the daily operation and fault disposal of the train drivers of urban rail transit,this paper teases out a competency model suitable for train drivers.A total of 100 train drivers were selected and their psychological quality was evaluated by using VTS.The results showed that they were in poor condition,especially under pressure.Therefore,timely detection of driving fatigue is the basis for ensuring safety in operation and production.Then,the mechanism of pulse wave generation,pulse baud point and its physiological significance involved in the photoelectric pulse plethysmography technology were elaborated in detail.The correlation between pulse and heart was emphatically explained,laying a theoretical foundation for the subsequent analysis of heart rate variability based on pulse signals.In experiment 1,we selected the simulated driver of Shanghai metro line 3.ECG was used to verify the reliability of the data of the device in four groups,including 2 subjects and 2 tasks.The analysis results showed that the consistency test results of RR interval under different tasks were high(p <0.05).Experiment 2 designed and conducted positive line simulated driving experiment.The above-mentioned equipment was used to collect the PPG signal of the subject driver,and the pupil area of the subject was obtained by Dikablis eyeglasses eye movement instrument.PPG data were analyzed every 5 minutes.The time domain and frequency domain indexes of the subjects showed a certain tendency.The nonlinear indexes reflected the pressure under different lengths and short-term conditions.In order to maintain consistency with PPG data,the average pupil area of the two eyes of the subjects was counted as a group every 5 minutes and k-means clustering algorithm was used to divide the fatigue level of the drivers of the subjects.Finally,based on the classification of eye movement fatigue,the fatigue recognition model of electric train driver is constructed by using the support vector machine theory.The time domain,frequency domain,and nonlinear indexes were selected,including 8 sets of indexes and 576 sets of data.Randomly select 80% of the sample input trainers for training,and then identify the remaining 20% of the sample input classifier.The results show that the accuracy of the test machine under the gaussian kernel function is the highest,reaching 86.36%.The work fatigue monitoring system of rail transit train drivers based on PPG technology proposed in this paper provides theoretical and technical support for the realization of fatigue monitoring of rail transit train drivers.If the number of samples is increased and the fatigue identification norm is established,mass production and use can be considered.