Train Onboard Real-Time Monitoring System and its Implementation Method

TWI937963BActive Publication Date: 2026-09-01许维伦
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
TW114128013
Authority / Receiving Office
TW · TW
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-07-23
Publication Date
2026-09-01
Estimated Expiration
2045-07-22

Smart Images

  • Figure TWG2TB001908951_001
    Figure TWG2TB001908951_001
  • Figure TWG2TB001908951_002
    Figure TWG2TB001908951_002
  • Figure TWG2TB001908951_003
    Figure TWG2TB001908951_003
Patent Text Reader

Abstract

The present invention relates to a train onboard real-time monitoring system and its execution method. By using a single system device on the train body, the system measures the vibration and noise signals of the components and the train's position information in real time, performs index calculations, and interprets the obtained data to instantly confirm the real-time status of the performance of components such as the track, wheelset, and car body, as well as the train's coordinates, thereby providing long-term monitoring to facilitate system maintenance.
Need to check novelty before this filing date? Find Prior Art

Claims

1. A train-borne real-time monitoring system, comprising: a field-end host, electrically connected to a signal acquisition system; at least two sets of accelerometers attached to the train floor to acquire vibration waves from the carriage floor, electrically connected to the signal acquisition system; at least two sets of floor microphones to acquire noise from inside and outside the carriage, electrically connected to the signal acquisition system; and a Global Positioning System (GPS) signal receiver, electrically connected to the field-end host.

2. The train onboard real-time monitoring system as described in claim 1, wherein the field-end host is connected to a router transmission unit.

3. The train onboard real-time monitoring system as described in claim 2, wherein the router transmission unit further communicates with a remote terminal host.

4. The train onboard real-time monitoring system as described in claim 1, wherein the floor microphone is installed on the carriage floor.

5. A method for implementing a train onboard real-time monitoring system as described in claims 1 to 4, comprising: measuring the amplitude of train floor vibration using two or more sets of accelerometers, and transmitting the obtained signals to a signal acquisition system; receiving audio signals from inside and outside the train using two or more sets of floor microphones, and transmitting the obtained signals to the signal acquisition system; obtaining coordinate information using a satellite positioning signal receiver; transmitting the collected signals to a field host after performing a convergence operation using the signal acquisition system, whereby the field host performs preliminary calculations and records the aforementioned signals; and communicating with a terminal host at a remote control center via a network using a router transmission unit.

6. A method for implementing a train onboard real-time monitoring system as described in claims 1-4, comprising: measuring the amplitude of train floor vibration using two or more sets of accelerometers, and transmitting the obtained signals to a signal acquisition system; receiving audio signals from inside and outside the train using two or more sets of floor microphones, and transmitting the obtained signals to the signal acquisition system; obtaining coordinate information using a satellite positioning signal receiver; transmitting the collected signals to a field-end host after performing a convergence operation using the signal acquisition system, whereby the field-end host performs preliminary calculations and records the aforementioned signal indicators; and transmitting the signal using a router transmission unit. The method involves communicating with a terminal host at a remote control center via the network; further utilizing the vibration of the vehicle floor and satellite positioning signals to perform non-contact steel wheel diameter determination; wherein the terminal host performs a selection operation on the vibration acceleration signal obtained from the accelerometer to obtain a window function, performs a short-time Fourier transform operation to obtain time-frequency domain data, performs an extraction operation to obtain wheel rotation characteristic frequency data, and simultaneously refers to the vehicle speed signal to perform an instantaneous steel wheel diameter estimation operation to obtain long-term steel wheel diameter information, and then performs a rejection operation to remove outlier diameter data to obtain the estimated steel wheel diameter value.

7. The method for executing the train onboard real-time monitoring system as described in claim 6, wherein the window function length is preferably set to 10 seconds.

8. A method for implementing a train-borne real-time monitoring system as described in claims 1-4, comprising: performing amplitude measurement using two or more accelerometers, transmitting the obtained signal to a signal acquisition system; performing audio reception using two or more floor microphones, transmitting the obtained signal to the signal acquisition system; obtaining coordinate information using a satellite positioning signal receiver; transmitting the signal to a field-end host after performing a convergence collection operation using the signal acquisition system, whereby the field-end host performs preliminary signal index calculations and records the signal; communicating with a terminal host at a remote control center via a network using a router transmission unit; and further utilizing sound pressure-induced response... The method for judging track wavy wear includes: using the terminal host to apply the vibration acceleration signal obtained by the accelerometer into a moving window function for calculation; performing a transformation operation using a Fourier transform formula to obtain time-frequency domain data; then converting it into an image matrix by a combination operation; performing singular value decomposition on the matrix data by a decomposition operation; then reconstructing the matrix by a reconstruction operation to retain the main characteristic values ​​in the matrix; performing wavy wear characteristic frequency setting and energy summation within a characteristic frequency range by empirical rules; then normalizing the wavy wear index by an extraction operation; and finally, performing a judgment operation to calculate the wavy wear index status by combining the information obtained from the extraction operation with the train position information.

9. A method for implementing a train-borne real-time monitoring system as described in claims 1-4, comprising: performing amplitude measurement using two or more accelerometers, transmitting the obtained signal to a signal acquisition system; performing audio reception using two or more floor microphones, transmitting the obtained signal to the signal acquisition system; obtaining coordinate information using a satellite positioning signal receiver; performing a convergence collection operation using the signal acquisition system and transmitting the signal to a field host, whereby the field host performs preliminary signal index calculations and records the signal; and transmitting the signal via a network using a router transmission unit. The terminal host set up in the Tongyi remote control center further utilizes the vibration of the carriage floor and the vehicle speed and position to determine the riding comfort. The execution method includes: obtaining acceleration signals using an accelerometer, converting them into a 1 / 3 octave frequency spectrum information through short-time Fourier transform, and then obtaining a human perception weighting function by combining the human perception weighted curve; obtaining a comfort curve by merging X / Y / Z three-axis comfort, and calculating the comfort index value of a specific interval by referring to the vehicle speed and position information within a specific interval; and calculating the proportion of the factors affecting the comfort index by deep learning operation on the above index values.

Citation Information

Patent Citations

  • Railway state monitoring device, bogie for railway vehicle, railway vehicle, and railway brake control device

    CN112498413A

  • Steel rail corrugation acoustic diagnosis system for rail transit

    CN112722010A

  • High-speed railway train-track-environment integrated monitoring method based on acoustic monitoring

    CN114771605A

  • Navigation apparatus and method of generating a view therefor

    TW200949208A

  • Onboard Railway Health Monitoring

    US20220017129A1