Real-time Monitoring and Early Warning Analysis of Urban Railway Operation Based on Multi-parameter Vital Signs of Subway Drivers in Plateau Environment

In order to ensure the personal safety of the drivers and passengers of rail transit in plateau environment, the vital signs and train conditions of the drivers and passengers are taken as the research object, and the dynamic relationship between them is studied and analyzed. In this paper, subway drivers under normal operation conditions are taken as research objects to establish the vital signs monitoring and early warning system. The vital signs data of the subway drivers, such as heart rate (HR), respiratory rate (RR), body temperature (T) and blood oxygen saturation (SPO2) of the subway driver are collected by the head-mounted sensor, and the least mean square adaptive filtering algorithm is used to preprocess the data and eliminate the interference information. Based on the improved BP (Back Propagation) neural network algorithm, a prediction model is established to predict the vital signs of subway drivers in real-time. We use the early warning score evaluation method to measure the risk of subway drivers' vital signs, and then the necessary judgment basis can be provided to dispatchers in the control center. Experiments show that the system developed in this paper can accurately predict the evolution of subway drivers' vital signs, and timely warn the abnormal states. The predicted value of vital signs is consistent with the actual value, and the absolute error of prediction is less than 0.5 which is within the allowable range.