System and method for detecting defects in a railway track.

Abstract

System and method for detecting defects in a railway track. The invention mainly relates to a system (1) for detecting defects in a railway track (2) configured to: Analyze each recorded data item and determine a difference between said data item and a determined reference data item. Classify the analyzed data item as discontinuity data item if the determined difference is greater than a determined threshold; Select at least one contextual information item associated with said discontinuity data item; Establish a correlation index between the contextual information item and the discontinuity data item, and Qualify the discontinuity data item as a railway track defect (2) if the correlation index is less than a second determined threshold. Figure to be published with the abstract: Fig. 2

Description

Title of the invention: System and method for detecting defects in a railway track. Technical field

[0001] The invention falls within the railway field and relates more specifically to the monitoring of faults in railway tracks.

[0002] The invention relates more particularly to a system and a method for qualifying data relating to the rails of a railway track with the aim of detecting defects. The invention also relates to the qualification of data relating to the rolling elements of railway rolling stock. PRIOR ART AND DISADVANTAGES OF PRIOR ART

[0003] The integrity of railway tracks is an essential parameter for rail traffic. Indeed, the number of track defects and especially their amplitude must be as low as possible to allow trains to run while ensuring passenger comfort and avoiding premature damage to equipment.

[0004] Correlatively, the integrity of the rolling parts of trains and other railway rolling stock is also an essential parameter for safety and comfort of circulation.

[0005] It is known to directly measure the profile of a railway track using a measuring device on board a railway vehicle, the measuring device generally being an optical system of the laser beam rangefinder type. The profile of each rail of the track is thus obtained, and consequently any geometric defects and their position on the track are detected.

[0006] However, it is necessary to use a railway machine dedicated to carrying out these measurements, because the installation of the optical measuring system requires prior calibration tests to be carried out which are dependent on the railway machine. In addition, the railway machine dedicated to the measurement must necessarily regularly travel all the tracks, which causes a disruption to the commercial circulation of trains. Finally, this measuring device does not always allow the acquired data to be correctly qualified, which ultimately does not allow a real railway defect to be differentiated from a simple discontinuity in said railway track. OBJECTIVE OF THE INVENTION

[0007] The invention thus aims to propose a system for detecting defects in a railway track which overcomes the drawbacks of the prior art. Statement of the invention

[0008] To this end, the invention relates to a system for detecting track defects. railway comprising two opposite rails, suitable for being installed on railway rolling stock and comprising at least data acquisition means configured to be arranged opposite the two rails of the railway track, and computer means for managing said acquisition means configured to control the acquisition and recording of data representative of at least one characteristic of the railway track and contextual information at least representative of an infrastructure element of the railway track other than the rails, which computer management means are further configured to: • Analyze each recorded data and determine a difference between said data and a determined reference data; • Classify the analyzed data as discontinuity data if the determined difference is greater than a determined threshold; • Select at least one piece of contextual information associated with said discontinuity data; • Establish a correlation index between the contextual information and the discontinuity data, and • Qualify the discontinuity data as a railway fault if the correlation index is lower than a second determined threshold.

[0009] The system of the invention may also include the following optional characteristics considered in isolation or according to all possible technical combinations: - Data acquisition means include image acquisition means. - The image acquisition means comprise at least two linear or matrix cameras for each rail of the railway track configured to be arranged opposite the rails of the railway track. - The image acquisition means comprise at least two cameras sensitive to the ultraviolet spectrum of light, respectively configured to be arranged opposite the rails of the railway track. - The data acquisition means include means of stabilizing the image acquisition means. - The acquisition means comprise at least one microphone intended to be installed on a wheel of the railway rolling stock, the image acquisition means being intended to acquire the infrastructure data and the microphone being intended to acquire the contextual information, and the management means comprise means for synchronizing the data with the contextual information. - The acquisition means comprise a plurality of accelerometers provided to be installed on the wheels of railway rolling stock and configured to detect defects in the wheels of said railway stock. - The system includes at least one temperature probe intended to be installed on at least one axle of the railway rolling stock.

[0010] The invention also relates to a method for detecting defects in a railway track comprising two opposite rails, implemented by a detection system installed on railway rolling stock running on said railway track, which system comprises at least means for acquiring data and contextual information configured to be arranged opposite the two rails of the railway track, and computer means for managing said data acquisition means, which method comprises at least the successive steps of: • Acquisition and recording of data representative of at least one characteristic of the railway track rails and contextual information; • Analysis of each recorded data, and determination of a difference between said data and a determined reference data; • If the determined deviation is greater than a determined threshold, classification of the associated data as discontinuity data; • Selection of at least one contextual information associated with said analyzed data; • Establishment of a correlation index between the contextual information and the associated discontinuity data, and • Qualification of the discontinuity data as a railway defect if the correlation index is lower than a second determined threshold.

[0011] The detection method may also include the following optional characteristics considered in isolation or according to all possible technical combinations: - The means of managing the qualification system comprise synchronization means, and the method comprises, prior to the analysis of each data item, a synchronization step between said data and the contextual information. - The synchronization means comprise geolocation means, the detection system comprises communication means connected to the management means and to a wide area network, the method comprises a step of acquiring position coordinates concomitant with the data acquisition step and a step of associating each data item with each position coordinate acquired at the same time, and the step of qualifying the discontinuity data item comprises the following sub-steps: • Recovery of correlation indices established by at least two other detection systems installed on two other railway rolling stock, said correlation indices being established from discontinuity data sharing the same position coordinates; • Calculation of a final correlation index from the three correlation indices from the three detection systems; • Final qualification of the discontinuity as a railway defect if the final correlation index is greater than a third determined threshold.

[0012] The invention also relates to railway rolling stock comprising a fault detection system as described above. PRESENTATION OF THE FIGURES

[0013] Other characteristics and advantages of the invention will emerge clearly from the description given below, for information purposes only and in no way limiting, with reference to the appended figures, among which:

[0014] [Fig.l] [Fig.l] represents a schematic cross-sectional view of a part of the qualification system of the invention;

[0015] [Fig.2] [Fig.2] represents a schematic side view of the quality system application of the invention installed on railway rolling stock. DETAILED DESCRIPTION OF THE INVENTION

[0016] It is first of all specified that in the figures, the same references designate the same elements regardless of the figure in which they appear and regardless of the form of representation of these elements. Similarly, if elements are not specifically referenced in one of the figures, their references can be easily found by referring to another figure.

[0017] It is also specified that the figures essentially represent an embodiment of the subject of the invention but that there may be other embodiments which meet the definition of the invention.

[0018] The present invention relates to a method for detecting defects in the rails 3 of the railway tracks 2. This method is implemented by a simple and inexpensive detection system 1, which is installed on a commercial train 4. Advantageously, this detection system 1 also makes it possible to detect defects on the wheels 12 of said railway rolling stock 4 on which it is installed.

[0019] The detection system 1 of the invention is intended to be installed on railway rolling stock 4 of the commercial train type intended for the transport of passengers or freight. In other words, it is not necessary to use railway rolling stock specifically dedicated to the installation of the detection system 1, but any commercial train can be used. Not only does this avoid the occupation of railway track 2 by trains dedicated solely to measurements for the maintenance of tracks 2, but in addition the measurements can be carried out on a large number of tracks 2, and the frequency of measurements on each railway track 2 is simply limited by the number of commercial trains 4 running on railway track 2.

[0020] In the remainder of the description, the term train will be used to define the railway rolling stock 4, but it goes without saying that the detection system 1 of the invention can be installed on any type of railway rolling stock 4.

[0021] The train 4 on which the detection system 1 of the invention is installed is put into circulation on a railway track 2 comprising two opposite rails 3 and of which one wishes to know the topography and determine the structural defects of the railway track 2 - that is to say the breaks in the rails 3, the surface defects of the rails 3, the breaks of fishplates (not shown) which ensure the junction between two consecutive rails 3, the missing nuts (not shown). The train 4, which extends along a longitudinal axis, comprises at least one body 14 and at least two axles, each axle comprising two axle boxes arranged at its respective ends. Typically, the train 4 comprises several bodies 14 and in fact several axles, and each axle comprises two wheels 12.

[0022] According to the invention, the detection system 1 comprises data acquisition means 5 configured to acquire: • data representative of at least one characteristic of rails 3 of track 2 - which will be called qualification data in the remainder of the description, and • contextual information representative of a static infrastructure element of railway track 2 other than rails 3, or representative of an element of train 4 interacting with rail 3.

[0023] The qualification data are those containing the structural information of the rails 3 and the connecting elements of the consecutive rails 3 (braids, fishplates). In other words, if the rail 3 or the connecting element analyzed include structural defects, the qualification data include indications of the presence of these structural defects. The qualification data may advantageously relate to the train 4 on which the detection system 1 is installed - in particular the wheels 12 of the train 4 via acceleration data experienced by the wheels 12.

[0024] The contextual information includes the elements associated with the railway track 2 - other than the rails 3 - and with the train 4 which are nevertheless in interaction with the rails 2. This acquired contextual information can come from the structural elements of the railway track 3 - the fishplates connecting two rails 3 together, the nuts connecting the fishplates to the rails 3, the sleepers 15, the ballast -, or even from the interaction between the running train 4 and rails 3 of track 2 - for example the sound of wheels 12 on rails 3, or the temperature of the axles.

[0025] The detection system 1 also comprises computer means for managing the data acquisition means 5. These management means typically comprise a central unit 6 comprising at least one processor, a graphics card or FPGA (Field-Programmable Gate Array) processor and data storage means in which one or more data processing algorithms are installed. The management means also comprise means for time-stamping the data and contextual information, a geographic positioning device of the GPS (Global Positioning System) type and wireless communication means.

[0026] With reference to [Fig.l], the data acquisition means 5 comprise image acquisition means 7 connected to the management means and mounted on a frame 16 connected to the train 4.

[0027] This frame 16 comprises a transverse central rod 17 (i.e. which extends perpendicularly to the longitudinal axis of the train 4 or of the railway track 2) pivotally connected to the body in question 14 of the train 4 along a transverse pivot axis, and two V-shaped frame portions 18 secured to the ends of the central rod 17, the frame portions 18 being arranged opposite the respective rails 3 of the railway track 2.

[0028] The image acquisition means 7 comprise four linear or matrix cameras 8, distributed two by two on each V-shaped frame portion 18. More precisely, for each V-shaped frame portion 18, two linear or matrix cameras 8 are mounted at the end parts of the branches 19 of the V-shaped portion considered 18. Advantageously, the data acquisition means 5 comprise lighting devices (not shown) associated with the cameras 8 to improve the quality of the images acquired by the latter.

[0029] In addition, each linear or matrix camera 8 is mounted free to translate in a direction parallel to the branch 19 on which it is mounted. Advantageously, each linear or matrix camera 8 is mounted on a motorized translation rail secured to the branch in question 19 of the V-shaped frame portion 18 and controlled by the central unit 6 of the management means. In the same way, the central rod 17 is mounted on the body in question 14 of the train 4 via a rotary motorized device controlled by the central unit 6. In this way, the central unit 6 can orient each linear or matrix camera 8 precisely and individually, so that each of them explores the desired part of the rail. The linear or matrix cameras 8 are designed to detect surface defects on the rails 3 and breaks in the rails 3.

[0030] The image acquisition means 7 also comprise two cameras sensitive in the ultraviolet spectrum of light 9 - which will be called UV cameras 9 in the remainder of the description - which are mounted in transverse translation (i.e. perpendicular to the longitudinal axis of the train 4 or of the railway track 2) at the base 20 of each V-shaped frame portion 18 and are arranged opposite the respective rails 3. Like the linear or matrix cameras 8, each UV camera 9 is mounted on a motorized translation rail controlled by the central unit 6. The central unit 6 can thus orient each UV camera 9 precisely and individually, so that the latter are perfectly opposite the respective rails 3. The UV cameras 9 make it possible to detect hydrocarbon stains on the surface of the rails 3.

[0031] With reference to [Fig. 2], the detection system 1 of the invention is connected to the body considered 14 of the train 4 by means of securing means 21, which securing means 21 comprise for example two attachments 22 cooperating with each other and respectively secured to the detection system 1 and to the body considered 14 of the train 4.

[0032] The detection system 1 also comprises stabilization means 10 arranged between the frame 16 and the attachment considered 22 of the securing means 21. These stabilization means 10 comprise at least one damper 23 which make it possible to maintain the constant distance between the image acquisition means 7 and the rails 3, whether or not the train 4 is moving on the railway track 2.

[0033] Optionally, the data acquisition means 5 may comprise means for detecting defects in the rail 3, such as for example eddy current detectors (not shown) integral with the armature 16 and connected to the management means.

[0034] The data acquisition means 5 further comprise microphones 11 connected to the management means, ideally four microphones 11 arranged two by two on the wheels 12 of an axle. For each wheel 12, one of the microphones 11 is oriented towards the rails 2 to pick up the sounds caused by the movement of the wheel 12 on the rail in question 3 - and in particular the noise emitted by a wheel 12 passing over a joint between two rails 3 - and the other microphone 11 is oriented so as to pick up the aerodynamic noises caused by the movement of the train 4. The measurement of the aerodynamic noises makes it possible to subtract these noises from the recordings made by the microphones oriented towards the tracks and thus isolate the noises caused by the defects.

[0035] The detection system 1 also comprises means for measuring the accelerations experienced by the wheels of the train 4 running on the railway track 3. Ideally, the wheels 12 of two axles are each equipped with an accelerometer 13, so that the detection system 1 comprises four accelerometers connected to the management means. These accelerometers 13 make it possible to identify defects in the rolling parts of the train, which include the axles and wheels 12.

[0036] The detection system 1 comprises at least one temperature probe (not shown) connected to the management means and installed on a box of an axle of the train 4. This probe measures the temperature of the axle box to identify a possible seizure of said axle box, in particular if the measured temperature is higher than a determined value.

[0037] Finally, the detection system 1 comprises energy supply means, for example batteries.

[0038] According to the invention, a method for detecting defects in a railway track 2 implemented by the detection system 1 will now be described.

[0039] The train 4 being in circulation on the railway track 2, the management means control the acquisition means 5 so that the latter continuously acquire a data flow and a contextual information flow. More precisely, the linear or matrix cameras 8 and the UV cameras 9 acquire characteristic images of the surface of the rails, the accelerometers 13 acquire acceleration data of the wheels of the train, the microphones 11 continuously record the sounds emitted by the wheels 12 circulating on the rails 3, and the temperature of the axle box(es) is continuously recorded by the temperature probe(s).

[0040] In parallel with their acquisition, each railway track data 2 and each contextual information is associated with a geolocation acquired by the positioning means, and where appropriate are timestamped by the timestamping means. Thus, the acquired data and contextual information are synchronized during a second step of the method.

[0041] During a third step, the acquired data are analyzed by the central unit 6 of the management computer means in order to determine a difference between each data item and a determined reference data item. For example, the central unit 6 determines whether the amplitude of a discontinuity at the surface of the rail in question 3 is greater than a threshold amplitude determined and recorded in the storage means of the central unit 6.

[0042] The analysis of these data and the associated deviations is carried out by the algorithms recorded in the storage means of the central unit 6. In addition, the contextual information is also classified by these algorithms. Advantageously, these analyses and comparisons are carried out by convolutional neural networks.

[0043] According to the invention, several convolutional neural networks are used, in particular: • A first neural network analyzes the surface data coming from the image acquisition means, to detect defective attachments, surface defects, broken fishplates or rails 3, missing nuts 3 and by extension other defects present on the railway infrastructure 2 (for example: defective sleepers, bottoms, ballast defects); • A second neural network to analyze part of the contextual information, in particular the presence of fishplates, crosspieces 15 or braids (elements which provide connections between metal structures to ensure their equipotentiality), and • A third neural network to analyze another part of the contextual information, in particular the sounds acquired by the microphones 11.

[0044] Other data from the accelerometers and contextual information from the temperature probes are recorded and associated with position coordinates, and where appropriate time-stamped.

[0045] During a fourth step, if for one of the data considered the central unit 6 detects that the determined deviation is greater than the threshold deviation, then said central unit 6 classifies this data as discontinuity data. This means that this discontinuity data is likely to be associated with a railway track defect 2 - in particular a break in rail 3 or fishplate or a surface defect in rail 3.

[0046] As the acquisition means 5 are not able to differentiate between a real defect and a normal structural discontinuity - for example a joint between two rails 3 will be detected as discontinuity data even though it is not a railway track defect 2 - the method implements a fifth step which will make it possible to qualify this discontinuity data according to a determined confidence index. Typically, this confidence index varies between 70% and 90% depending on the sensor having acquired the data.

[0047] To carry out this qualification of the discontinuity data, the central unit 6 comprises additional algorithms making it possible to carry out the fusion of data coming from the different convolutional neural networks, the temperature sensors and the accelerometers.

[0048] As an example, two fusion models, one using the Bayesian method of Markov random fields and the other the probabilistic belief theory of Dempster Shafer, are recorded in the storage means of the central unit 6.

[0049] The data and contextual information having been previously synchronized, the fusion model will select one or more contextual information associated with the discontinuity data, then will establish a correlation index between said discontinuity data and the associated contextual information(s). Obtaining this correlation index depends on the type of fusion model used. For example, the Bayesian Markov random field method will establish a correlation link between the discontinuity data and the associated contextual information, whereas the method based on Dempster Shafer's probabilistic belief theory will establish a probability between 0 (no correlation) and 1 (maximum correlation).

[0050] In particular, if a break in continuity at the level of a rail 3 is detected, the fusion model will search in the acquired image for the presence of a fishplate or a braid. If a fishplate or a braid is detected, then the discontinuity data will be qualified as normal data, in particular a joint between two rails 3 - that is to say not a defect - with a determined confidence index representative of the correlation index determined by the fusion model.

[0051] If no fishplate or braid is detected, the fusion model will analyze the sound produced by the movement of the wheels 12 on the rail 3. If no particular sound is detected, the fusion model will conclude that there is a rail joint - therefore a normal situation - with a second determined confidence index representative of the correlation index determined by the fusion model. If a particular sound is detected, the data will be qualified as faulty with this same confidence index.

[0052] Furthermore, the contextual information from the UV cameras 9 also makes it possible to qualify a discontinuity data item presenting a surface defect: if the UV camera 9 returns contextual information characteristic of a hydrocarbon stain, the discontinuity data item is qualified as such by the fusion model with a third determined confidence index representative of the correlation index determined by the fusion model. Conversely, if the UV camera 9 does not return any particular information, the discontinuity data item is qualified as a rail surface defect 3 with this same confidence index.

[0053] Finally, in the absence of discontinuity data associated with the rails 3, the detection system 1 also makes it possible to identify wheel defects 12: if the accelerometer(s) detect a periodic acceleration, the central unit 6 will conclude that there is a defect in the rolling members.

[0054] The fusion model makes it possible to refine this diagnosis with information from the temperature probe: if the temperature of the axle in question exceeds a determined value, for example 100 degrees Celsius, then the central unit concludes that the wheels are seized.

[0055] The detection system 1 also comprises, recorded in the storage means of the central unit, at least one algorithm implementing a decision fusion model to qualify the discontinuity data with an increased confidence index, making it possible to avoid having to resort to the intervention of human operators.

[0056] The decision fusion model may implement one or more of the following methods: • Principal Component Analysis (PCA); • Latent Structure Projections (LSP); • Discriminant Analysis (DA), Qualitative Trend Analysis (QTA); • k nearest neighbors (kNN); • Decision tree; • Belief theory; • The bond Graph, and • Fuzzy logic.

[0057] The above list is of course not exhaustive.

[0058] To improve the confidence index of a qualification of discontinuity data, the method implements the following sub-steps:

[0059] The central unit 6 will firstly recover the discontinuity data having the same position coordinates and obtained by at least two other detection systems 1 installed on two other respective trains 4, and their respective confidence indices.

[0060] An overall confidence index will thus be calculated by the central unit 6 via the decision fusion model. This confidence index is all the higher as the number of discontinuity data with the same coordinates and coming from other trains is high. In this case, by using the discontinuity data coming from three detection systems 1 installed on three different trains 4, the overall confidence index obtained to qualify the discontinuity data is comparable to the confidence index associated with a human operator.

[0061] Thus, the system and the detection method of the invention make it possible to determine in a precise, reliable and reproducible manner the structural defects of the railway track 3 and, where appropriate, of the rolling elements of the train 4 without requiring a complex and costly installation, without requiring specific railway rolling stock and without requiring the intervention of human operators in the processing of the acquired data.

Claims

Claims

1. System (1) for detecting defects in a railway track (2) comprising two opposite rails (3), suitable for being installed on railway rolling stock (4) and comprising at least data acquisition means (5) configured to be arranged opposite the two rails (3) of the railway track (2), and computer management means (6) of said acquisition means (5) configured to control the acquisition and recording of data representative of at least one characteristic of the railway track (2) and contextual information at least representative of an infrastructure element of the railway track (2) other than the rails (3), which computer management means (6) are further configured to: • Analyze each recorded data item and determine a difference between said data item and a determined reference data item; • Classify the analyzed data item as discontinuity data if the determined difference is greater than a determined threshold;• Select at least one piece of contextual information associated with said discontinuity data; • Establish a correlation index between the contextual information and the discontinuity data, and • Qualify the discontinuity data as a railway fault (2) if the correlation index is lower than a second determined threshold.;

2. System (1) according to the preceding claim, characterized in that the data acquisition means (5) comprise image acquisition means (7).

3. System (1) according to the preceding claim, characterized in that the image acquisition means (7) comprise at least two linear or matrix cameras (8) for each rail (3) of the railway track (2) configured to be arranged opposite the rails (3) of the railway track (2).

4. System (1) according to claim 2 or 3, characterized in that the image acquisition means (7) comprise at least two cameras sensitive in the ultraviolet spectrum of light (9) respectively configured to be arranged opposite the rails (3) of the railway track. (2).

5. System (1) according to any one of claims 2 to 4, characterized in that the data acquisition means (5) comprise stabilization means (10) of the image acquisition means (7).

6. System (1) according to any one of claims 2 to 5, characterized in that the acquisition means (5) comprise at least one microphone (11) intended to be installed on a wheel (12) of the railway rolling stock (4), the image acquisition means (5) being intended to acquire the infrastructure data and the microphone (11) being intended to acquire the contextual information, and in that the management means (6) comprise means for synchronizing the data with the contextual information.

7. System (1) according to any one of the preceding claims, characterized in that the acquisition means (5) comprise a plurality of accelerometers (13) intended to be installed on the wheels (12) of the railway rolling stock (4) and configured to detect defects in the wheels (12) of said railway stock (4).

8. System (1) according to the preceding claim, characterized in that it comprises at least one temperature probe intended to be installed on at least one axle of the railway rolling stock (4).

9. Method for detecting defects in a railway track comprising two opposite rails, implemented by a detection system installed on railway rolling stock running on said railway track, which system comprises at least means for acquiring data and contextual information configured to be arranged opposite the two rails of the railway track, and computer means for managing said data acquisition means, characterized in that it comprises at least the successive steps of: • Acquisition and recording of data representative of at least one characteristic of the rails of the railway track and contextual information; • Analysis of each recorded data item, and determination of a difference between said data item and a determined reference data item; • If the determined difference is greater than a determined threshold, classification of the associated data item as discontinuity data item;• Selection of at least one contextual information associated with; said analyzed data; • Establishment of a correlation index between the contextual information and the associated discontinuity data, and • Qualification of the discontinuity data as a railway defect (2) if the correlation index is lower than a second determined threshold.

10. Method according to the preceding claim, characterized in that the management means (6) of the qualification system (1) comprise synchronization means, and in that the method comprises, prior to the analysis of each data item, a synchronization step between said data and the contextual information.

11. Detection method according to the preceding claim, characterized in that the synchronization means comprise geolocation means, in that the detection system (1) comprises communication means connected to the management means (6) and to a wide area network, in that the method comprises a step of acquiring position coordinates concomitant with the data acquisition step and a step of associating each data item with each position coordinate acquired at the same time, and in that the step of qualifying the discontinuity data item comprises the following sub-steps: • Recovery of the correlation indices established by at least two other detection systems installed on two other railway rolling stock, said correlation indices being established from discontinuity data sharing the same position coordinates;• Calculation of a final correlation index from the three correlation indices from the three detection systems (1), and • Definitive qualification of the discontinuity as a railway fault (2) if the final correlation index is greater than a third determined threshold.;

12. Railway rolling stock (4) comprising a fault detection system (1) according to any one of claims 1 to 8.

Images