Granted Patent B2

US 12,612,089 · App. 18/245,252 · Granted Apr 28, 2026

Inventors: Ole Tommy Vorren (Munich, DE); Olav Stetter (Munich, DE)

Assignee: KONUX GMBH

B61L25/021G01M17/10

Abstract

Claims (63)

1 . A method for analyzing properties related to a passing train on a railway network, the method comprising

collecting a first sensor data of railway at a first time via at least one sensor arranged on the railway network, wherein the first sensor data comprises a vibration signal, wherein the at least one vibration signal comprises a signal frequency and a sampling rate, wherein the signal frequency is between 0 and 10,000 Hz and wherein the sampling rate is between 0 and 20 kHz,

processing the first sensor data via at least one processing component to generate at least one processed sensor dataset, wherein the processing comprises extracting at least one Power Spectral Density (PSD) data set from the least one vibration signal and/or extracting at least one Mel spectrogram from the least one vibration signal, and

estimating at least a speed of the passing train based on the at least one Power Spectral Density (PSD) data set and/or the at least one Mel spectrogram.

2 . The method according to claim 1 , further comprising:

mapping the first sensor data to generate a mapped data set, and

curating the at least one estimation of the train speed based on the mapped data.

3 . The method according to claim 1 , wherein the signal frequency is between 50 and 8,000 Hz.

4 . The method according to claim 1 , wherein the signal frequency is between 100 and 5,000 Hz.

5 . The method according to claim 1 , wherein the sampling rate is between 0.1 and 8 KHz.

6 . The method according to claim 1 , wherein the sampling rate is between 1 and 5 kHz.

7 . The method according to claim 1 , wherein the method further comprises bidirectionally connecting the at least one sensor to at least one server, and connecting the at least one server with the at least one processing component.

8 . The method according to claim 7 , further comprising retrieving from the at least one sensor at least one of

the first sensor data, and

at least one sensor ID,

supplying to the at least one server at least one of

the first sensor data, or

the at least one sensor ID,

wherein the at least one sensor ID is related to the at least one sensor.

9 . The method according to claim 1 , wherein the at least one vibration signal comprises at least one of:

at least frequency data;

at least displacement data;

at least velocity data; or

at least acceleration data,

further comprising automatically generating at least one acceleration trace associated with the first sensor data based on the at least one vibration signal.

10 . The method according to claim 9 , further comprising pre-processing the first sensor data via the at least processing component, wherein the step of pre-processing further comprises at least one of:

flagging at least one noisy component of the first sensor data,

removing at least one exponential wakeup,

cutting off the edge of the at least one acceleration trace, or

stretching the at least one first sensor data to a pre-determined size,

representing the at least one first sensor data as a time-frequency spectrogram.

11 . The method according to claim 10 , wherein the health status hypothesis comprises information related to the at least one axle, the method further comprises:

estimating a specific axle of the passing train to generate a first axle finding,

correlating the first axle finding to the unique ID of the at least one axle,

generating at least one final axle finding comprising the health status of the at least one axle,

automatically reporting the unique ID of the at least one axle comprising a flat wheel, and

automatically identifying the unique ID of the at least one axle comprising the flat wheel across the at least 2 sensors.

12 . The method according to claim 9 , further comprising:

dynamically extracting at least one temporal and spectral content to standardize an output to a fixed size, and

automatically converting the at least one acceleration trace to at least one time-frequency spectrogram.

13 . The method according to claim 1 , further comprising generating a health status hypothesis of at least one component of the passing train, wherein the at least one component is at least one wheel of the passing train, and wherein the health status hypothesis of the at least one wheel is used to identify a flat wheel.

14 . The method according to claim 13 , further comprising counting at least one of

axle, and

wagon,

wherein the method comprises assigning a unique ID to each of the at least one axle and at least one wagon, wherein the unique ID is assigned based on a passing order, wherein the passing order comprises a passing sequence of the at least one axle and/or at least one wagon over the at least one sensor.

15 . The method according to claim 1 , further comprising:

collecting at least a second sensor data of railway at at-least a second time via the least one sensor arranged on the railway network,

processing the at least second sensor data via the at least one processing component, and

estimating at least a second value of the at least one property related to the passing train based on the at least second processed sensor data

wherein the method comprises tracking the health status of the at least one component of the passing train across at least 2 sensors.

16 . A system for analyzing properties related to a passing train on a railway network, the system comprising

at least one sensor configured to measure at least one property related to the passing train,

at least one processing component configured to process the at least one property related to the passing train,

at least one analyzing component configured to analyze the at least one property related to the passing train, and

at least one interface configured to access at least one server configured to be bidirectionally connected to the system,

wherein the system is configured to

estimate a value of the at least one property related to the passing train based on the at least one property, and

execute the method according to claim 1 .

17 . The system according to claim 16 , wherein the system is configured to establish a bidirectional communication between the at least one server and the at least one sensor, and wherein the at least one processing component is configured to generate at least one health status hypothesis of at the at least one component of the passing train, wherein the at least one server is configured to execute at least one of

predicting the at least one health status of the at least one component of the passing train,

generating the at least one health status hypothesis of the at the at least one component of the passing train,

generating the at least one suggestion procedure, wherein the at least one suggestion procedure comprises the at least one action to remediate the at least one flat wheel, and

prompting a user to implement the at least one of the at least one suggestion procedure and/or the at least one of the at least one action.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (10)