Research On Application Of Smartphone Monitoring Data In Urban Rail Transit

The track conditions affect the vehicle safety and ride comfort directly.Real-time and accurate detection of the track conditions can provide reliable data support for the track maintenance department to develop a scientific and economic maintenance plan,it is an important measure to maintain good track condition,ensure vehicle safety and ride comfort.Currently,the monitoring,evaluation and maintenance of the track conditions are mainly based on the data of the track inspection vehicle,which can accurately reflect the track conditions.The monitoring period is generally two times a month,and the track irregularity that threatens the vehicle safety and ride comfort cannot be found in time.Moreover,the manufacturing cost and maintenance cost of the track inspection vehicle are relatively high.At present,smart phones are widely used in the field of intelligent transportation as a new type of monitoring equipment,they have many kinds of sensors and strong computing ability.This paper proposes to use smart phones as auxiliary monitoring devices for track conditions in urban rail transit.The vibration acceleration of vehicle body is monitored by smart phone acceleration sensor.The vehicle operating quality is reflected by the ISO-2631 comfort index,UIC513 comfort index,Sperling stability index.The main conclusions are as follows:(1)Affected by the complexity of the test environment,the mobility of the measurement terminal,and the operation errors of the passengers,the vibration response of the car body detected by the smart phone has strong error interference,and there is an angular deviation between the mobile phone coordinate system and the car body coordinate system.In this paper,a systematic data cleaning process is proposed:data resampling,outlier identification and processing,wavelet denoising,and the phone posture modification matrix is derived.(2)Three problems in the application of smart phones in urban rail transit are solved:the field test of smart phones and precision sensors shows that smart phone sensor performance meets the needs of on-site multiplication;the test of two smart phones of the same type is carried out,and the correlation coefficients of mobile phone detection data of two different connection modes are greater than 0.8248,indicating that smart phones can be put on the floor surface of the car directly to detect the vibration response of the car body;in order to avoid affecting the passenger's experience of taking the subway,it is recommended that the smart phone should be placed on the floor surface of the cab in the first carriage,and 1m away from the center.(3)The smartphone detection data of Tianfu Square to Chunxi Road on Chengdu Metro Line 2 are analyzed.The number of decomposition layers n=8 is determined according to the sampling frequency of smartphone at 100Hz and the frequency range of 0.5-80Hz which affects human comfort.The random noise of the detected data is filtered by wavelet analysis.Then,the angular deviation between the smartphone coordinate system and the vehicle body coordinate system is corrected,so that the smart phone monitoring data can truly reflect the vibration acceleration of the vehicle body.According to the evaluation index of vehicle operating quality,the calculated amplitude of ISO-2631 comfort index is less than 0.315m/s~2,the calculated amplitude of UIC513 comfort index is less than 1m/s~2,and the Sperling stability index is less than 2.5 m/s~2.All evaluation indexes are less than the comfort grade I,which indicates that the vehicle operating quality and local track condition in this area are good.(4)According to the original parameters of the subway vehicles,the first,second and third order bending natural vibration frequency of the vehicle body are 3.56 Hz,9.87 Hz and19.34 Hz respectively.The time-frequency analysis of mobile phone detection data shows that the vertical acceleration vibration frequency of the vehicle mainly distributs at 1~2.9Hz,which is less than the first-order bending frequency of the vehicle body of 3.56Hz,indicating that there are sensitive wavelengths that affect the quality of the vehicle's operation under the current operation mode of the subway system.(5)The deviation between the initial mileage obtained by the quadratic integral of the longitudinal acceleration of the vehicle body and the actual running mileage of the vehicle is corrected by Kalman filter,so that the mileage error of the vehicle can be controlled within1m,and the mileage position of the vehicle swaying point can be quickly determined,and reduce the labor intensity of staff in the public works department.