Vehicle inspection system, vehicle inspection method, and vehicle inspection program

Abstract

To obtain a vehicle inspection system that can accurately diagnose a vehicle.SOLUTION: A vehicle inspection system 10A comprises: a travel information storage device 11 that stores travel information being information which includes a travel history of a vehicle 1A driven by a driving apparatus; a sound collection device 4 that collects vibrations due to the driving apparatus when a vehicle passes through a railway track; a sound storage device 3 that stores vibration information which is information of vibrations; and a vehicle inspection device 2 that determines whether or not abnormality is occurring in the driving apparatus on the basis of the travel information and the vibration information in an axis position unit of the vehicle.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a vehicle inspection system, a vehicle inspection method, and a vehicle inspection program for inspecting a vehicle. [Background technology]

[0002] It is desirable for railway vehicles to run in a normal condition. For this reason, development of vehicle inspection devices that can determine whether a vehicle is in a normal condition is underway. One type of vehicle inspection device is a device that determines whether there are any abnormalities based on the sound of the vehicle running.

[0003] The anomaly detection device described in Patent Document 1 calculates an acoustic score of sounds picked up by a microphone installed beside the tracks using a normal sound model, and determines whether or not an abnormality exists based on the acoustic score and a threshold value. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-73366 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology of Patent Document 1 only takes into account the train's formation information when determining whether or not there is an abnormality, and does not take into account the vehicle's running history, which means that the vehicle cannot be accurately diagnosed.

[0006] The present disclosure has been made in view of the above, and aims to provide a vehicle inspection system that can accurately diagnose a vehicle. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems and achieve the object, the vehicle inspection system of the present disclosure is driven by a driving machine. trainThe vehicle inspection system of the present disclosure further comprises a driving information storage device that stores driving information including the driving history of the vehicle. Trains located on the ground caused by the driving equipment when passing through the track sound of Outside the train Collect sound The vehicle inspection system of the present disclosure includes a collection device. sound This is information about sound Remember information sound A storage device; A vibration measuring device is installed on a train to measure vibrations other than the sound of the driving equipment on the train, In axle position units, travel information and sound and a vehicle inspection device that determines whether or not an abnormality has occurred in the driving equipment based on the information. The vehicle inspection device performs an analysis of vibration features based on vibration, performs an analysis of sound features based on sound information, and corrects the analysis method of sound features so that the analysis results of the vibration features and the analysis results of the sound features coincide with each other. [Effects of the Invention]

[0008] The vehicle inspection system according to the present disclosure has the effect of being able to accurately diagnose a vehicle. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing a configuration of a vehicle inspection system according to a first embodiment. [Figure 2] FIG. 1 is a diagram for explaining the configuration of a vehicle to be inspected by a vehicle inspection system according to a first embodiment; [Figure 3] FIG. 1 is a diagram for explaining the configuration of a bogie of a vehicle to be inspected by the vehicle inspection system according to the first embodiment; [Figure 4] FIG. 1 is a diagram showing an example of sound information acquired by the vehicle inspection device according to the first embodiment; [Figure 5] FIG. 1 is a diagram for explaining feature quantities calculated by the vehicle inspection device according to the first embodiment. [Figure 6] FIG. 1 is a diagram for explaining feature amounts stored in a vehicle inspection device according to a first embodiment. [Figure 7] FIG. 10 is a diagram for explaining a transition of a feature amount calculated by the vehicle inspection device according to the first embodiment. [Figure 8] 1 is a flowchart showing a processing procedure of a process executed in the vehicle inspection system according to the first embodiment; [Figure 9]FIG. 10 is a diagram showing a configuration of a vehicle inspection system according to a second embodiment. [Figure 10] FIG. 10 is a diagram showing a configuration of a vehicle inspection system according to a third embodiment. [Figure 11] FIG. 10 is a diagram showing a configuration of a vehicle inspection system according to a fourth embodiment. [Figure 12] 10 is a flowchart showing a processing procedure of a process executed in a vehicle inspection system according to a fourth embodiment. [Figure 13] FIG. 10 is a diagram for explaining a correction process of an analysis method executed by a vehicle inspection device according to a fourth embodiment. [Figure 14] FIG. 10 is a diagram showing a configuration of a vehicle inspection system according to a fifth embodiment. [Figure 15] FIG. 20 is a diagram showing a first example of arrangement of a sound collecting device arranged in a vehicle inspection system according to a sixth embodiment; [Figure 16] FIG. 20 is a diagram showing a second example of arrangement of a sound collecting device arranged in a vehicle inspection system according to a sixth embodiment. [Figure 17] FIG. 20 is a diagram showing a third example of the arrangement of the sound collecting device arranged in the vehicle inspection system according to the sixth embodiment. [Figure 18] FIG. 13 is a diagram showing the configuration of a learning device included in a vehicle inspection system according to a seventh embodiment. [Figure 19] FIG. 13 is a diagram for explaining a neural network used by a learning device according to a seventh embodiment. [Figure 20] 13 is a flowchart showing a procedure of a learning process executed by a learning device according to a seventh embodiment. [Figure 21] FIG. 13 is a diagram showing a configuration of an inference device according to a seventh embodiment. [Figure 22] 13 is a flowchart showing the procedure of an inference process executed by an inference device according to a seventh embodiment. [Figure 23] FIG. 10 is a diagram showing an example of the configuration of a processing circuit provided in the vehicle inspection device according to the first to seventh embodiments when the processing circuit is realized by a processor and a memory. [Figure 24] FIG. 10 is a diagram showing an example of a processing circuit when the processing circuit provided in the vehicle inspection device according to the first to seventh embodiments is configured with dedicated hardware. DETAILED DESCRIPTION OF THE INVENTION

[0010] A vehicle inspection system, a vehicle inspection method, and a vehicle inspection program according to embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0011] Embodiment 1 1 is a diagram showing the configuration of a vehicle inspection system according to a first embodiment. The vehicle inspection system 10A is a system that diagnoses the driving equipment of a vehicle 1A and determines whether or not there is an abnormality. The vehicle inspection system 10A includes a driving information storage device 11, a communication device 12, a sound collection device 4, a sound storage device 3, and a vehicle inspection device 2.

[0012] The running information storage device 11 and the communication device 12 are disposed, for example, in a vehicle 1A driven by a drive mechanism, and the sound collection device 4, the sound storage device 3, and the vehicle inspection device 2 are disposed near the track 9 on which the vehicle 1A passes. When the vehicle inspection system 10A inspects multiple vehicles 1A, the running information storage device 11 and the communication device 12 are disposed for each vehicle 1A. The sound storage device 3 and the vehicle inspection device 2 may be disposed in any location. At least one of the running information storage device 11 and the communication device 12 may be disposed in a vehicle (car) other than the vehicle 1A. For example, if the vehicle 1A to be inspected is car number 2, at least one of the running information storage device 11 and the communication device 12 may be disposed in another car, such as the lead car (car number 1). In this way, at least one of the running information storage device 11 and the communication device 12 may be disposed in any location in the train to which the vehicle 1A is coupled. The running information storage device 11 may also have a communication function. In this case, the communication device 12 does not need to be disposed in the vehicle 1A.

[0013] The travel information storage device 11 stores information (hereinafter referred to as travel information) including the travel history of the vehicle 1A in which the travel information storage device 11 is installed. That is, in the vehicle inspection system 10A, the travel information storage device 11 stores the travel information for each vehicle 1A.

[0014] The travel information items (hereinafter sometimes referred to as travel history items) stored in the travel information storage device 11 include at least one of information on the route traveled by the vehicle 1A, information on the distance traveled by the vehicle 1A, information on the speed at which the vehicle 1A traveled, information on the notch at which the vehicle 1A traveled, and information on the motor current to the motor that drives the axle. The motor current at which the vehicle 1A traveled is the current of the motor that drives the vehicle 1A. The travel information storage device 11 sends the travel information to the communication device 12.

[0015] The communication device 12 sends the traveling information sent from the traveling information storage device 11 to the vehicle inspection device 2. That is, the communication device 12 transmits the traveling information to the ground.

[0016] The sound collecting device 4, which is an example of a vibration collecting device, collects sounds caused by the driving equipment when the vehicle 1A passes near the sound collecting device 4. In other words, the sound collecting device 4 is a sound collecting device that acquires passing sounds of the driving equipment (e.g., a motor) that drives the vehicle 1A. The sound collecting device 4 is placed near the track 9.

[0017] The vibration collecting device may be a device that collects sound (air vibrations) like the sound collecting device 4, or it may be a device that collects vibrations other than sound (vibrations propagating via tracks 9, etc.).

[0018] When the vibration collecting device is a sound collecting device 4 that collects sound, the vibration collecting device has a microphone, etc. When the vibration collecting device is a device that collects vibrations of a railway line 9, etc., the vibration collecting device has an acceleration sensor (vibration sensor), etc.

[0019] In the following, a case will be described in which the vibration collecting device is the sound collecting device 4, but in the vehicle inspection system 10A, the vibration collecting device may be a device other than the sound collecting device 4. That is, in the following, a case will be described in which the collected vibration is sound, but the collected vibration may be something other than sound.

[0020] The sound collection device 4 sends collected sound information (hereinafter referred to as sound information) to the sound storage device 3. The sound storage device 3 stores the sound information. The vehicle inspection device 2 reads out the sound information from the sound storage device 3 and receives driving information from the communication device 12.

[0021] The vehicle inspection device 2 is a computer that inspects the drive equipment of the vehicle 1A based on the sound information and driving information. The vehicle inspection device 2 is an analysis device that receives the driving information from on board the vehicle 1A and analyzes the sound information.

[0022] The vehicle inspection device 2 correlates and analyzes the traveling information and sound information for each axle position (each axle of the vehicle 1A), and detects abnormalities in the driving equipment of the vehicle 1A based on changes in the sound over time. Note that the driving equipment in the first embodiment is not limited to a motor. The driving equipment in the first embodiment may also be gears, couplings, etc. Furthermore, the driving equipment in the first embodiment may also be equipment that is disposed under the floor and is used for the traveling of the vehicle 1A, such as a bogie, wheels, or inverter. In other words, any driving equipment that generates sound due to the driving of the vehicle 1A may be the subject of determination for the presence or absence of an abnormality.

[0023] The vehicle inspection device 2 has a communication unit 21, an input unit 22, an analysis unit 23, and an output unit 24. The communication unit 21 communicates with the communication device 12. The communication unit 21 transmits a request for driving information to the communication device 12 and receives the driving information from the communication device 12. The communication unit 21 sends the received driving information to the analysis unit 23. The input unit 22 reads out sound information from the sound storage device 3 and sends it to the analysis unit 23.

[0024] The analysis unit 23 analyzes the driving equipment of the vehicle 1A based on the change in sound information (aging) and the driving information, and determines whether or not an abnormality has occurred in the driving equipment. The output unit 24 outputs the analysis results by the analysis unit 23 to an external device. The output unit 24 may transmit the analysis results to a remotely located device such as a server, or to a display device. When the display device receives the analysis results, it displays the analysis results.

[0025] Here, an example of a vehicle 1A to be inspected will be described. FIG. 2 is a diagram for explaining the configuration of a vehicle inspected by the vehicle inspection system according to the first embodiment. The vehicle 1A has a car body 31 and bogies 32. The car body 31 is placed on, for example, two bogies 32. The bogies 32 have a plurality of wheels, and each wheel runs on a track 9.

[0026] Fig. 3 is a diagram for explaining the configuration of a bogie of a vehicle inspected by the vehicle inspection system according to the first embodiment. Fig. 3 shows a top view of bogie 32. Bogie 32 has axles AX and BX, wheels 36A, 36B, 37A, and 37B, motors 33A and 33B, drive units 35A and 35B, joints 34A and 34B, a plurality of bearings 38, and a plurality of gears 39. In Fig. 3, the locations where bearings 38 are located are indicated by common hatching. Also, the locations where gears 39 are located are indicated by common hatching.

[0027] The axles AX and BX are arranged so as to be parallel to each other. In the bogie 32, the axles AX and BX are arranged so that the extension direction of the axles AX and BX is perpendicular to the direction of travel of the bogie 32.

[0028] The wheels 36A and 36B rotate on one of the two tracks 9, and the wheels 37A and 37B rotate on the other of the two tracks 9. For example, if the wheels 36A and 37A of the bogie 32 are the front wheels in the direction of travel, the wheels 36B and 37B are the rear wheels in the direction of travel.

[0029] The wheels 36A and 37A are connected to an axle AX, and the wheels 36B and 37B are connected to an axle BX. The motor 33A is connected to a drive unit 35A via a joint 34A, and the motor 33B is connected to a drive unit 35B via a joint 34B. FIG. 3 shows a case in which two bearings 38 are arranged in each of the motors 33A and 33B, and two gears 39 are arranged in each of the joints 34A and 34B. FIG. 3 also shows a case in which two gears 39 and four bearings 38 are arranged in each of the drive units 35A and 35B. Note that the numbers of bearings 38 and gears 39 arranged on the bogie 32 are not limited to those described in FIG. 3.

[0030] In the bogie 32, the wheels 36A and 37A, the motor 33A, the drive unit 35A, and the joint 34A constitute the drive equipment on the axle AX side. The drive equipment on the axle AX side includes a bearing (not shown) of the axle AX, the motor 33A, the drive unit 35A, or a gear 39 and a bearing 38 arranged in the joint 34A.

[0031] In addition, the wheels 36B and 37B, the motor 33B, the drive unit 35B, and the joint 34B are driving equipment on the axle BX side of the bogie 32. The driving equipment on the axle BX side includes a bearing (not shown) of the axle BX, the motor 33B, the drive unit 35B, or a gear 39 and a bearing 38 arranged in the joint 34B.

[0032] Vehicle inspection system 10A inspects the drive equipment of vehicle 1A based on sounds generated from bogie 32. For example, abnormal noise may be generated from gears 39 or bearings 38 due to aging deterioration of vehicle 1A. Vehicle inspection device 2 inspects the drive equipment of vehicle 1A based on sounds generated from gears 39 or bearings 38 and travel information including the travel history of vehicle 1A.

[0033] On the bogie 32, the axles AX and BX are arranged at a specific distance along the direction of travel of the bogie 32. Therefore, the timing at which the wheels 36A, 37A connected to the axle AX pass a specific position on the track 9 is different from the timing at which the wheels 36B, 37B connected to the axle BX pass this specific position. The vehicle inspection device 2 utilizes this timing difference to inspect the drive equipment of the vehicle 1A for each axle.

[0034] Furthermore, since each driving device arranged on the bogie 32 has a different type, model number, etc., the frequency of the sound emitted when an abnormality occurs varies for each driving device. Therefore, the vehicle inspection device 2 can identify which driving device has an abnormality based on the frequency of the sound.

[0035] Next, the inspection process by the vehicle inspection device 2 will be described in detail. Fig. 4 is a diagram showing an example of sound information acquired by the vehicle inspection device according to the first embodiment. Sound information 40 includes waveforms of sounds when the wheels 36B, 37B are rotating at various rotational speeds (number of rotations). Fig. 4 shows a case where sound information 40 includes sound information 41 to 44.

[0036] In each of sound information 41 to 44, the horizontal axis represents time and the vertical axis represents sound pressure (Pascal). Sound information 41 shows the waveform of a sound when the rotation speed of wheels 36B, 37B is 120 rpm (rotations per minute), and sound information 42 shows the waveform of a sound when the rotation speed of wheels 36B, 37B is 360 rpm. Sound information 43 shows the waveform of a sound when the rotation speed of wheels 36B, 37B is 600 rpm, and sound information 44 shows the waveform of a sound when the rotation speed of wheels 36B, 37B is 1200 rpm.

[0037] For example, Fig. 4 shows a case where sound information 42 includes impact sounds F1 and F2 that occur when a scratch occurs in bearing 38. Fig. 4 also shows a case where sound information 43 includes impact sounds F3 to F6 that occur when a scratch occurs in bearing 38.

[0038] The vehicle inspection device 2 calculates sound feature values ​​from the sound information 40 and determines whether the vehicle 1A is in an abnormal state based on the sound feature values. An example of a feature value is a peak value of suddenly high sound pressure that appears periodically (hereinafter, may be referred to as a sound pressure peak value). The period in which the feature value appears is the time it takes for the motor rotation shafts of the motors 33A and 33B to make one rotation.

[0039] FIG. 5 is a diagram for explaining feature quantities calculated by the vehicle inspection device according to the first embodiment. Here, an example of a process in which the analysis unit 23 of the vehicle inspection device 2 extracts feature quantities will be explained. Note that the process of extracting feature quantities by the analysis unit 23 is not limited to the process explained in FIG. 5. The analysis unit 23 may extract feature quantities by any process. The analysis unit 23 reads out sound information 51 from the sound storage device 3. The sound information 51 is, for example, any of the sound information 41 to 44.

[0040] The horizontal axis of the first to fourth graphs in Fig. 5 is time, and the vertical axis is sound pressure. The horizontal axis of the fifth graph in Fig. 5 is frequency, and the vertical axis is sound pressure.

[0041] The analysis unit 23 applies a BPF (Band Pass Filter) to the sound information 51 to generate sound information 52 by extracting a specific range of frequencies from the frequencies of the sound information 51. The analysis unit 23 calculates the absolute value (ABS: ABSolute value) of the frequency of the sound information 52 to generate sound information 53 from the sound information 52.

[0042] The analysis unit 23 generates sound information 54 by performing envelope (ENV) processing on the waveform of the sound information 53. The analysis unit 23 generates sound information 55 by performing fast Fourier transform (FFT (Fast Fourier Transform) analysis) on the ENV waveform included in the sound information 54. In other words, the analysis unit 23 generates sound information 55 from the sound information 54 by performing frequency analysis on the ENV waveform included in the sound information 54.

[0043] The analysis unit 23 extracts, as a feature, a sound pressure peak value at a characteristic frequency (for example, a frequency greater than a specific value) from the sound pressure components included in the sound information 55. FIG. 5 shows a case where the analysis unit 23 extracts a feature P1 from the sound information 55.

[0044] In the vehicle inspection system 10A, the sound collecting device 4 collects sound information 51 every time the vehicle 1A passes through a location where the sound collecting device 4 is installed. Then, the analysis unit 23 of the vehicle inspection device 2 extracts a feature amount from each piece of collected sound information 51. This feature amount changes with the deterioration of the driving equipment provided in the vehicle 1A. Therefore, the analysis unit 23 inspects the driving equipment of the vehicle 1A based on the change in the extracted feature amount.

[0045] Fig. 6 is a diagram for explaining feature quantities stored by the vehicle inspection device according to the first embodiment. Fig. 6 shows the configuration of feature quantity information 60, which is information on feature quantities stored by the analysis unit 23. In the feature quantity information 60, the date and time at which sound information 51 was collected and the feature quantity at this date and time are associated with each axle. The analysis unit 23 stores, for each axle, the feature quantity extracted for each date and time at which sound information 51 was collected.

[0046] 6 shows a case where the analysis unit 23 calculates the feature amounts of the axles 61A to 66A and 61B to 66B and stores them in the feature amount information 60. The axles 61A to 66A correspond to the axles AX, and the axles 61B to 66B correspond to the axles BX.

[0047] For example, the feature value of the axle 62A on May 1, 2021 is 20. FIG. 6 shows a case where the analysis unit 23 extracts "49" from the sound information 55 as the feature value P1, and registers the extracted "49" as the feature value of the axle 62A on December 1, 2021.

[0048] For example, the analysis unit 23 determines that an axle whose feature amount exceeds a threshold (hereinafter referred to as the deterioration detection threshold) is more likely to be an axle in an abnormal state than a specific value. That is, the analysis unit 23 provisionally determines that an axle whose feature amount included in the sound information 51 exceeds the deterioration detection threshold is an axle whose possibility of failure within a specific period is more than a specific value.

[0049] Fig. 7 is a diagram for explaining the transition of the feature amount calculated by the vehicle inspection device according to the first embodiment. Fig. 7 shows the transition of the feature amount of the sound extracted for a specific axle. The horizontal axis of the graph shown in Fig. 7 represents the year and month (date and time) when the sound information 51 was collected, and the vertical axis represents the feature amount of the sound.

[0050] The sound pressure peak value, which is a characteristic quantity of the sound, increases as the traveling distance of the vehicle 1A increases. The analysis unit 23 provisionally determines that an axle whose characteristic quantity of the sound exceeds the deterioration detection threshold is an axle whose likelihood of failure within a specific period is equal to or greater than a specific value.

[0051] In this case, the analysis unit 23 determines whether the axle is in an abnormal state based on the travel information of the vehicle 1A and the sound information 51 (sound feature amount). An axle is in an abnormal state when the possibility of the axle failing within a specific period is equal to or greater than a specific value.

[0052] For example, as described above, the travel history item of the travel information may include information about the route on which the vehicle 1A has traveled. Among the routes on which the vehicle 1A travels, there are routes that place a heavy burden on the vehicle and routes that place a light burden on the vehicle. In other words, the damage that the route causes to the vehicle 1A varies from route to route. Routes that cause little damage to the vehicle 1A include, for example, routes on which new tracks 9 are installed, routes that have not been repaired for a long time, routes on which the vehicle has traveled infrequently, etc.

[0053] The driving history items may also include the distance traveled by vehicle 1A. The shorter the distance traveled by vehicle 1A, the more likely the axle is in good condition. The driving history items may also include the history of the speed at which vehicle 1A traveled. The slower the speed at which vehicle 1A traveled, the more likely the axle is in good condition. The driving history items may also include the history of the notches at which vehicle 1A traveled. The less frequently sudden notch changes were made to vehicle 1A, the more likely the axle is in good condition. The driving history items may also include the history of the motor current to the motor that drives the axle. The smaller the motor current, the more likely the axle is in good condition.

[0054] For example, if the sound feature exceeds the deterioration detection threshold when the travel distance of vehicle 1A is shorter than a specific value, it is highly likely that the sound feature exceeds the deterioration detection threshold due to an external disturbance. Therefore, if the analysis unit 23 determines that the travel history item is unlikely to cause an abnormality in the axle, it determines that the axle is normal even if the sound feature exceeds the deterioration detection threshold. For example, the analysis unit 23 compares the travel history item with the threshold for this travel history item for each travel history item. For example, if any one of the travel history items exceeds its threshold, the analysis unit 23 determines that the axle is abnormal if the sound feature exceeds the deterioration detection threshold.

[0055] The sound feature quantity often does not suddenly increase. For this reason, the analysis unit 23 may determine that the axle is normal if the sound feature quantity suddenly exceeds the deterioration detection threshold. That is, the analysis unit 23 may determine that the axle is normal, for example, if the sound feature quantity suddenly increases or decreases by a specific rate or more, resulting in the sound feature quantity exceeding the deterioration detection threshold. In other words, the analysis unit 23 may determine that the axle is normal if the rate of increase or decrease of the sound feature quantity exceeds the deterioration detection threshold while exceeding a specific value.

[0056] Furthermore, the analysis unit 23 may assign a score to the entire driving history item after weighting each driving history item. The score is higher the more likely the driving history item is to cause an abnormality in the axle. For example, the score is higher the longer the driving distance. If the score exceeds a score threshold and the sound feature value exceeds the threshold, the analysis unit 23 determines that the axle is abnormal.

[0057] Furthermore, the analysis unit 23 may assign a score to the entire driving history items and sound feature quantities after weighting each driving history item and sound feature quantity. The score increases as the axle becomes abnormal. For example, the score increases as the sound feature quantity increases. The analysis unit 23 determines that the axle is abnormal when the score exceeds a score threshold.

[0058] Next, a description will be given of a procedure of processing executed in the vehicle inspection system 10 A. Fig. 8 is a flowchart showing the procedure of processing executed in the vehicle inspection system according to the first embodiment.

[0059] In the vehicle inspection system 10A, the travel information storage device 11 starts storing travel information of the vehicle 1A (step S10). The travel information storage device 11 acquires travel information from a detection device that detects travel history items such as travel distance and travel speed, and stores the travel information. The travel information storage device 11 transmits the travel information to the communication device 12. The communication device 12 transmits the travel information sent from the travel information storage device 11 to the vehicle inspection device 2 (step S20).

[0060] The sound collecting device 4 collects the sound of the vehicle 1A when the vehicle 1A passes near the sound collecting device 4 (step S30). The sound collecting device 4 transmits the sound information 51 to the sound storage device 3. The sound storage device 3 stores the sound information 51 (step S40).

[0061] The vehicle inspection device 2 reads out the sound information 51 from the sound storage device 3 and receives the driving information from the communication device 12 (step S50). The vehicle inspection device 2 determines whether or not the vehicle 1A is abnormal based on the sound information 51 and the driving information (step S60). The vehicle inspection device 2 outputs the determination result as to whether or not the vehicle 1A is abnormal.

[0062] It should be noted that either the processing of steps S10 and S20 or the processing of steps S30 and S40 may be executed first. Furthermore, in step S50, the vehicle inspection device 2 may execute either the processing of reading out sound information 51 from the sound storage device 3 or the processing of receiving driving information from the communication device 12 first. Furthermore, the vehicle inspection device 2 may read out sound information 51 from the sound storage device 3 before the processing of steps S10 and S20. Furthermore, the vehicle inspection device 2 may receive driving information from the communication device 12 before the processing of steps S30 and S40.

[0063] The vehicle inspection system 10A inspects the driving equipment of the vehicle 1A using condition-based maintenance (CBM). In condition-based maintenance, the timing of inspection is determined according to the condition of the driving equipment. Condition-based maintenance can inspect the driving equipment more efficiently than time-based maintenance (TBM).

[0064] In this way, the vehicle inspection system 10A can inspect the driving machinery without installing sensors on the driving machinery, making it possible to inspect the driving machinery at low cost. Furthermore, the vehicle inspection device 2 inspects the driving machinery of the vehicle 1A based on the running information and sound information 51, improving the accuracy of detecting abnormalities in the driving machinery.

[0065] As described above, in the first embodiment, the sound collecting device (vibration collecting device) 4 installed on the ground collects vibration information caused by the driving equipment when the vehicle 1A passes. In addition, the vehicle inspection device 2 determines whether or not an abnormality has occurred in the driving equipment for each axle position of the vehicle 1A based on the driving information including the driving history of the vehicle 1A and the vibration information. This enables the vehicle inspection system 10A to accurately diagnose the vehicle 1A. Therefore, the user can repair or replace the driving equipment of the vehicle 1A before the vehicle 1A breaks down.

[0066] Embodiment 2 Next, a second embodiment will be described with reference to Fig. 9. A vehicle inspection device 2 according to the second embodiment determines at least one of an analysis parameter and a deterioration detection threshold value to be used when analyzing sound information 51, according to various pieces of information about the driving machinery (hereinafter referred to as driving machinery information).

[0067] Fig. 9 is a diagram showing the configuration of a vehicle inspection system according to embodiment 2. Among the components in Fig. 9, components that achieve the same functions as those in vehicle inspection system 10A according to embodiment 1 shown in Fig. 1 are assigned the same reference numerals, and redundant explanations will be omitted.

[0068] Compared to the vehicle inspection system 10A of the first embodiment, the vehicle inspection system 10B of the second embodiment includes a vehicle 1B instead of the vehicle 1A. In addition to the components included in the vehicle 1A, the vehicle 1B includes an equipment information storage device 13. The equipment information storage device 13 is arranged in the vehicle 1B together with the running information storage device 11 and the communication device 12. The equipment information storage device 13 may be arranged in a vehicle (car) other than the vehicle 1B. The equipment information storage device 13 may also be arranged in a location (for example, a server) different from the train to which the vehicle 1B is coupled. The running information storage device 11 and the equipment information storage device 13 may each have a communication function. In this case, the communication device 12 does not need to be arranged in the vehicle 1B.

[0069] The equipment information storage device 13 stores driving equipment information of the driving equipment mounted on the vehicle 1B. The driving equipment information, which is the specifications of the driving equipment, includes the model numbers of the motors 33A and 33B, the model numbers of the bearings, and the characteristics of the driving equipment (e.g., gear ratio). The driving equipment information in the equipment information storage device 13 is read by the communication device 12. The communication device 12 of the vehicle 1B sends the driving equipment information and the running information to the vehicle inspection device 2. Note that the communication device 12 may send the driving equipment information and the running information at different times.

[0070] The vehicle inspection device 2 of the second embodiment receives driving equipment information and driving information from the communication device 12. The analysis unit 23 of the vehicle inspection device 2 stores a plurality of analysis parameters and a plurality of deterioration detection thresholds in advance.

[0071] The analysis unit 23 individually determines the analysis parameters and the deterioration detection threshold of the sound information 51 according to the moving machine information of the moving machine to be inspected. That is, the analysis unit 23 sets the analysis parameters corresponding to the moving machine from among a plurality of analysis parameters, and sets the deterioration detection threshold corresponding to the moving machine from among a plurality of deterioration detection thresholds.

[0072] The analysis parameters are signal processing parameters used in the process of analyzing the moving machine. For example, when a bearing is damaged, a sound of a specific frequency is generated. The frequency at which this sound is generated varies depending on the diameter of the bearing and the number of rolling elements. In other words, the frequency of the sound generated when a bearing is damaged differs for each moving machine. The analysis parameters set by the analysis unit 23 may be the sound generation frequency itself, or may be signal processing parameters (such as the cutoff frequency of a high-pass filter) used in the process of calculating the sound level for this generation frequency.

[0073] An example of the deterioration detection threshold set by the analysis unit 23 is a threshold for determining whether or not the level of the sound of the generated frequency is abnormal. Note that the analysis unit 23 may analyze the abnormality of the moving machine by any method.

[0074] The analysis unit 23 of the vehicle inspection device 2 correlates and analyzes the driving information and sound information 51 for each axle position. In this case, the analysis unit 23 inspects the driving equipment of the vehicle 1B using the set analysis parameters and deterioration detection threshold. As a result, the analysis unit 23 detects an abnormality in the driving equipment of the vehicle 1B based on the aging of the sound.

[0075] In this way, the analysis unit 23 sets the analysis parameters and the deterioration detection threshold corresponding to the moving machine information and inspects the moving machine. That is, the analysis unit 23 changes the sound analysis method according to the moving machine information.

[0076] As described above, according to the second embodiment, the vehicle inspection system 10B sets analysis parameters and deterioration detection thresholds corresponding to the driving equipment information and inspects the driving equipment, thereby making it possible to improve the accuracy of abnormality detection.

[0077] Embodiment 3 Next, a third embodiment will be described with reference to Fig. 10. The vehicle inspection system of the third embodiment detects the passage of a vehicle, and inspects the driving equipment of the vehicle by associating sound information (passing sound information to be described later) collected at the time the vehicle passes with traveling information.

[0078] Fig. 10 is a diagram showing the configuration of a vehicle inspection system according to embodiment 3. Among the components in Fig. 10, components that achieve the same functions as those in vehicle inspection system 10A according to embodiment 1 shown in Fig. 1 are assigned the same reference numerals, and redundant explanations will be omitted.

[0079] Vehicle inspection system 10C of embodiment 3 includes a wheel passing detection device 5 in addition to the components included in vehicle inspection system 10A of embodiment 1. Wheel passing detection device 5 is placed near track 9 and near sound collecting device 4. The positions where wheel passing detection device 5 is placed and the positions where sound collecting device 4 is placed are placed within a specific distance.

[0080] The wheel passing detection device 5 detects that the wheels 36B, 37B have passed near the sound collecting device 4. The wheel passing detection device 5 generates information indicating the timing at which the wheels 36B, 37B have passed, and transmits this information to the vehicle inspection device 2 as wheel passing information.

[0081] In the third embodiment, the analysis unit 23 of the vehicle inspection device 2 extracts passing sound information, which is sound information at the time when the wheels 36B, 37B passed, from the sound information 51 based on the wheel passing information. That is, the analysis unit 23 extracts the passing sound information at the time when the wheels 36B, 37B passed from the sound information 51. The analysis unit 23 associates the extracted passing sound information with the traveling information and inspects the driving equipment.

[0082] In this way, the analysis unit 23 correlates and analyzes the passing sound information collected when the wheels 36B, 37B pass over the sound collecting device 4 for each axle position, and detects abnormalities in the drive equipment from changes in the sound over time. By matching the passing sound information to be analyzed with the timing when the wheels 36B, 37B pass, the analysis unit 23 can analyze the drive equipment while reducing noise (for example, sounds generated by drive equipment at other axles). In other words, the analysis unit 23 can analyze the sounds generated by drive equipment at each axle position after excluding sounds generated by drive equipment at adjacent axles. The inspection method of the third embodiment using the wheel passing detection device 5 may also be applied to the vehicle inspection system 10B of the second embodiment.

[0083] As described above, according to the third embodiment, the vehicle inspection system 10C synchronizes the analyzed passing sound information with the timing when the wheels 36B and 37B pass by, so that the sound generated by the drive equipment at each axle position can be analyzed while excluding the sound generated by the drive equipment at other axles. Therefore, the vehicle inspection system 10C can improve the accuracy of anomaly detection.

[0084] Embodiment 4 Next, a fourth embodiment will be described with reference to Figures 11 to 13. The vehicle inspection system of the fourth embodiment measures vibrations of the driving machinery using a vibration measuring device arranged near the driving machinery, and corrects the analysis method based on sound information 51 so that the analysis results based on the vibration measurement results and the analysis results based on sound information 51 coincide with each other.

[0085] Fig. 11 is a diagram showing the configuration of a vehicle inspection system according to embodiment 4. Among the components in Fig. 11, components that achieve the same functions as those in vehicle inspection system 10A according to embodiment 1 shown in Fig. 1 are assigned the same reference numerals, and redundant explanations will be omitted.

[0086] Compared to vehicle inspection system 10A of embodiment 1, vehicle inspection system 10D of embodiment 4 includes vehicle 1D instead of vehicle 1A. In addition to the components included in vehicle 1A, vehicle 1D includes a vibration measuring device 14. Vibration measuring device 14 is arranged on vehicle 1D together with driving information storage device 11 and communication device 12. Vibration measuring device 14 is arranged near the driving equipment on vehicle 1D.

[0087] In the vehicle inspection system 10D, even if the vehicle 1D has a plurality of bogies 32, it is sufficient that a vibration measuring device 14 is provided for one bogie 32. Note that if the vehicle 1D has a plurality of bogies 32, a vibration measuring device 14 may be provided for each bogie 32.

[0088] The vibration measuring device 14 is equipped with a vibration sensor or an acceleration sensor, and measures the vibration of the driving equipment using the vibration sensor or the acceleration sensor. The vibration measuring device 14 sends the vibration measurement results to the driving information storage device 11. The vibration measuring device 14 may also have a communication function. In this case, the vibration measuring device 14 may transmit the vibration measurement results to the vehicle inspection device 2.

[0089] The driving information storage device 11 stores the vibration measurement results. The vibration measured by the vibration measuring device 14 may be a device that collects sound (air vibration) or a device that collects vibration other than sound. In the fourth embodiment, a case will be described in which the vibration measuring device 14 measures vibration propagated via a member connected to a driving device.

[0090] The communication device 12 sends the vibration measurement results and the driving information to the vehicle inspection device 2. Note that the communication device 12 may transmit the vibration measurement results and the driving information at different times.

[0091] The main cause of noise from driving equipment is the vibration of the driving equipment itself, so there is a very high correlation between vibration and sound, and in principle it is possible to detect abnormalities in bearings, etc. by analyzing sound alone. However, if sound emitted from sources other than the driving equipment is included in the sound measurement, accurate inspection becomes difficult.

[0092] Therefore, in the fourth embodiment, a vibration measuring device 14 arranged on the vehicle 1D measures the vibration of the driving equipment, and the vehicle inspection device 2 uses the measurement results to inspect the driving equipment. When the vibration measuring device 14 is arranged near the driving equipment, vibrations emitted from devices other than the driving equipment rarely dominate the vibration measurement results, so the accuracy of vibration-based fault determination is high. To reduce the installation cost of the vibration measuring device 14, in the fourth embodiment, the vibration measuring device 14 is arranged on only some of the driving equipment in the vehicle inspection system 10D. In the vehicle inspection system 10D, for example, the vibration measuring device 14 is arranged on the vehicle 1D, which is one of multiple vehicles included in a train.

[0093] The analysis unit 23 of the vehicle inspection device 2 analyzes the vibrations using the vibration measurement results. That is, the analysis unit 23 analyzes whether or not an abnormality has occurred in the driving equipment based on the vibration measurement results.

[0094] Furthermore, as described in the first embodiment, the analysis unit 23 analyzes whether or not an abnormality has occurred in the moving machine based on the sound information 51 collected by the sound collecting device 4. The analysis unit 23 compares the analysis based on the vibration measurement results with the analysis based on the sound information 51. Hereinafter, in the fourth embodiment, the measurement results of the vibration measured by the vibration measuring device 14 may be referred to as on-board measured vibration, and the measurement results of the vibration (sound information 51) collected by the sound collecting device 4 may be referred to as ground measured vibration.

[0095] The analysis unit 23 corrects the analysis method based on the sound information 51 so that the on-board measured vibration has a value corresponding to the ground measured vibration. That is, the analysis unit 23 corrects the analysis method based on the ground measured vibration so that the vibration feature amount calculated based on the on-board measured vibration and the vibration feature amount calculated based on the ground measured vibration have corresponding values. The analysis unit 23 corrects, for example, at least one of the analysis parameters, thresholds (such as a deterioration detection threshold), and noise removal method used when performing an analysis based on the sound information 51.

[0096] The analysis unit 23 uses the modified analysis method to determine whether or not there is an abnormality in the vehicle 1D and another vehicle different from the vehicle 1D. This enables the analysis unit 23 to perform accurate abnormality determination even for driving equipment at an axle position where the vibration measuring device 14 is not installed. Note that the analysis unit 23 may apply the modified analysis method to trains other than the train including the vehicle 1D.

[0097] Next, the processing procedure executed in the vehicle inspection system 10D will be described. Fig. 12 is a flowchart showing the processing procedure executed in the vehicle inspection system according to the fourth embodiment. Note that the description of the same processing as that explained in Fig. 8 of the first embodiment will be omitted.

[0098] The processes executed by the vehicle inspection system 10D include a process ST1 executed for the moving machine whose vibration is to be measured, and a process ST2 executed for all moving machines, including moving machines other than the measuring target.

[0099] Process ST1 includes steps S110 to S150, and process ST2 includes steps S170 and S180. Hereinafter, the moving machine whose vibration is being measured will be referred to as the target moving machine, and other moving machines whose vibration is not being measured will be referred to as the different moving machines. Process ST2, which is executed after process ST1, is executed for both the different moving machine and the target moving machine, but the following will describe the case where process ST2 is executed for the different moving machine.

[0100] In the vehicle inspection system 10D, the vibration measuring device 14 measures vibrations on the vehicle (step S110). That is, the vibration measuring device 14 arranged near the drive equipment measures vibrations caused by the target drive equipment on the vehicle 1D. The on-vehicle measured vibrations measured by the vibration measuring device 14 are sent to the vehicle inspection device 2 via the driving information storage device 11 and the communication device 12.

[0101] The analysis unit 23 of the vehicle inspection device 2 analyzes the vibrations based on the on-board measured vibrations (step S120). The analysis unit 23 analyzes the vibrations, for example, by processing similar to the processing described in FIG. 5. In this case, the analysis unit 23 calculates vibration feature amounts based on the on-board measured vibrations. The analysis unit 23 determines whether or not an abnormality has occurred in the target driving equipment based on the vibration feature amounts.

[0102] Furthermore, the sound collecting device 4 measures sound on the ground (step S130). That is, the sound collecting device 4 measures sound caused by the target driving equipment on the ground. Specifically, the sound collecting device 4 collects sound of the vehicle 1D when the vehicle 1D passes near the sound collecting device 4. Sound information 51, which is ground-measured vibration measured by the sound collecting device 4, is sent to the vehicle inspection device 2 via the sound storage device 3.

[0103] The analysis unit 23 of the vehicle inspection device 2 analyzes the sound based on the ground-measured vibrations for the target driving equipment (step S140). That is, the analysis unit 23 analyzes the sound information 51 for the target driving equipment. The analysis unit 23 analyzes the sound, for example, by processing similar to the processing described in FIG. 5. In this case, the analysis unit 23 calculates the feature amount of the sound based on the ground-measured vibrations. The analysis unit 23 determines whether or not an abnormality has occurred in the target driving equipment based on the feature amount of the sound and driving information including the driving history of the vehicle 1D.

[0104] When it is determined that an abnormality has occurred in the target movable equipment, the analysis unit 23 determines whether the analysis result based on the on-board measured vibration (first analysis) and the analysis result based on the ground measured vibration (second analysis) match (step S150). The analysis results here are vibration feature quantities and sound feature quantities. That is, the analysis unit 23 determines whether the vibration feature quantities calculated based on the on-board measured vibration correspond to the sound feature quantities calculated based on the ground measured vibration. Note that even when the analysis unit 23 determines that the target movable equipment is normal based on the analysis result based on the ground measured vibration, it may determine whether the vibration feature quantities calculated based on the on-board measured vibration correspond to the sound feature quantities calculated based on the ground measured vibration.

[0105] When the analysis unit 23 determines that the analysis result based on the on-board measured vibration does not match the analysis result based on the ground measured vibration (step S150, No), it corrects the sound analysis method (step S160). That is, when the value of the vibration feature does not correspond to the value of the sound feature, the analysis unit 23 corrects the sound analysis method so that the value of the vibration feature corresponds to the value of the sound feature.

[0106] After this, the vehicle inspection system 10D analyzes the sound of the other driving equipment. In this case, the sound collecting device 4 measures the sound on the ground (step S170). That is, the sound collecting device 4 measures the sound caused by the other driving equipment on the ground. Specifically, the sound collecting device 4 collects the sound of a vehicle different from the vehicle 1D (hereinafter referred to as the other vehicle) when the other vehicle passes near the sound collecting device 4. Sound information 51, which is the ground measurement vibration measured by the sound collecting device 4, is sent to the vehicle inspection device 2 via the sound storage device 3.

[0107] If the analysis unit 23 determines that the analysis results based on the on-board measured vibrations match the analysis results based on the ground measured vibrations (step S150, Yes), it executes the processing of step S170 without executing the processing of step S160.

[0108] The analysis unit 23 of the vehicle inspection device 2 analyzes the sound based on the ground-measured vibrations for the separate driving equipment (step S180). That is, the analysis unit 23 analyzes the sound information 51 of the separate driving equipment. The analysis unit 23 analyzes the sound, for example, by processing similar to the processing described in FIG. 5. In this case, the analysis unit 23 analyzes the sound by the modified analysis method. The analysis unit 23 calculates the sound feature amount based on the ground-measured vibrations. The analysis unit 23 determines whether an abnormality has occurred in the target driving equipment based on the sound feature amount and driving information including the driving history of the separate vehicle.

[0109] After correcting the sound analysis method, analysis unit 23 may also analyze the sound of vehicle 1D using the corrected analysis method. Analysis unit 23 outputs the analysis results obtained using the corrected analysis method to an external device such as a display device.

[0110] It is noted that the order in which the processes of steps S110 and S120 and the processes of steps S130 and S140 are executed may be any. Furthermore, the vehicle inspection device 2 may execute the process of step S170 before the processes of steps S110 and S120.

[0111] In vehicle inspection system 10D, the processes of steps S170 and S180 may be executed multiple times. Also, in vehicle inspection system 10D, the processes of steps S110 to S160 may be executed multiple times. That is, the sound analysis method may be modified multiple times.

[0112] Furthermore, in the vehicle inspection system 10D, after the sound analysis method has been modified a specific number of times, the vibration measuring device 14 may be removed from the vehicle 1D. The removed vibration measuring device 14 may be installed in another vehicle. In this case, the processes of steps S110 to S180 may be performed on the other vehicle on which the vibration measuring device 14 is installed. In this way, the vibration measuring device 14 can be reused.

[0113] 13 is a diagram for explaining the correction process of the analysis method executed by the vehicle inspection device according to the fourth embodiment. If the vibration feature amount calculated based on the on-board measured vibration does not have a value corresponding to the feature amount calculated based on the ground measured vibration, the analysis unit 23 determines that the analysis results of the on-board measured vibration and the ground measured vibration do not match (s1).

[0114] In this case, the analysis unit 23 adjusts the sound analysis method (s2). The analysis unit 23 adjusts the sound analysis conditions, for example, by changing the range of frequencies to be analyzed. The analysis unit 23 may also adjust the analysis conditions, for example, by changing (modifying) the deterioration detection threshold used to determine whether or not the moving machine is abnormal.

[0115] Analysis unit 23 inspects each car of the train to which car 1D is coupled under the adjusted analysis conditions. Fig. 13 shows a case where analysis unit 23 changes the range of sound frequencies used when inspecting cars 1 to N (N is a natural number).

[0116] In this way, the vehicle inspection device 2 compares the result of determining whether or not there is an abnormality in the target driving equipment from which vibration data is acquired, based on the vibration data alone, with the result of determining based on the driving history and sound data. If there is a discrepancy between the vibration analysis result and the sound analysis result, the vehicle inspection device 2 corrects the sound analysis method (adjusts the analysis conditions). This improves the accuracy of detecting abnormalities in the driving equipment. In other words, the vehicle inspection device 2 can improve the accuracy of determining whether or not there is an abnormality in the driving equipment by applying the sound analysis method corrected by the vibration analysis to the analysis of sounds at other axle positions.

[0117] The inspection method of the fourth embodiment using the vibration measuring device 14 may be applied to the vehicle inspection systems 10B and 10C of the second and third embodiments.

[0118] Thus, according to embodiment 4, the vehicle inspection system 10D modifies the analysis method based on sound information 51 so that the analysis results based on vibration measurement results match the analysis results based on sound information 51, thereby making it possible to improve the accuracy of abnormality detection with a simple configuration.

[0119] Embodiment 5 Next, a fifth embodiment will be described with reference to Fig. 14. In the fifth embodiment, a server collects analysis results from a plurality of vehicle inspection devices and aggregates the analysis results.

[0120] Fig. 14 is a diagram showing the configuration of a vehicle inspection system according to embodiment 5. Among the components in Fig. 14, components that achieve the same functions as those in vehicle inspection system 10A according to embodiment 1 shown in Fig. 1 are assigned the same reference numerals, and redundant explanations will be omitted.

[0121] The vehicle inspection system 10E of the fifth embodiment includes a plurality of vehicle inspection systems and a server 8. In FIG. 14, the vehicle inspection system 10E includes vehicle inspection systems 10A1 to 10A n (n is a natural number of 2 or more) and a server 8. nis a system similar to vehicle inspection system 10A described in embodiment 1. Vehicle inspection system 10E may also include vehicle inspection systems 10B to 10D described in embodiments 2 to 4.

[0122] Vehicle Inspection System 10A1~10A n Each vehicle inspection device 2 transmits the determination result of whether or not there is an abnormality for each axle based on the traveling information and sound information 51 to the server 8. As a result, the server 8 transmits the determination result of whether or not there is an abnormality for each axle based on the traveling information and sound information 51 to the server 8. n It is possible to aggregate the abnormality determination results sent from

[0123] Server 8 can determine whether or not there is an abnormality for each axle based on multiple determination results for vehicle 1A. For example, even if one vehicle inspection system determines that a specific axle is abnormal, if multiple other vehicle inspection systems determine that this axle is normal, server 8 will determine that this axle is normal.

[0124] In addition, vehicle inspection systems 10A1 to 10A n Each vehicle inspection device 2 may calculate sound features for each axle based on the traveling information and sound information 51, and transmit the calculated features to the server 8. In this case, the server 8 performs abnormality determination of the driving equipment for each axle based on the sound features transmitted from each vehicle inspection device 2 and the deterioration detection threshold.

[0125] A vehicle that repeatedly enters and exits different inspection yards passes through the same sound collecting device 4 less often, and the number of times that sound features can be calculated is also reduced. n Since the vehicle inspection system 10E can perform abnormality determination based on the feature quantities of a large number of sounds acquired from the vehicle inspection systems 10A1 to 10A, the accuracy of the abnormality determination can be improved. n By calculating feature quantities using multiple vehicle inspection systems such as those described above, the accuracy of determining abnormalities in driving equipment can be improved.

[0126] Furthermore, the vehicle inspection system 10E enables early detection of abnormalities, for example, even if the vehicle 1A is stored at a different base for each specific period, because the inspection results can be shared on the server 8, allowing for early collection of features.

[0127] Also, vehicle inspection systems 10A1 to 10A n Alternatively, the vehicle inspection systems 10A1 to 10A2 may transmit the sound information 51 and the travel information for each axle to the server 8. In this case, the server 8 performs abnormality determination of the driving equipment for each axle based on the sound information 51 and the travel information. n may not be equipped with the vehicle inspection device 2.

[0128] Furthermore, the vehicle inspection system 10E may have a plurality of vehicle inspection devices 2, and may have a single vehicle 1A. The server 8 may store the driving equipment information described in the second embodiment. In this case, the vehicle inspection device 2 reads out the driving equipment information from the server 8 when analyzing the sound information 51.

[0129] In this way, in vehicle inspection system 10E, server 8 correlates and analyzes sound information 51 with driving information including driving history for each axle position, and detects abnormalities in the driving equipment from changes in the sound over time. Furthermore, vehicle inspection system 10E aggregates abnormality determination results on server 8, allowing each vehicle inspection device 2 to share the analysis results of other vehicle inspection devices 2. Server 8 monitors trends in the acquired feature amounts and comprehensively determines whether the driving equipment is normal or abnormal.

[0130] Vehicle Inspection System 10A1~10A n In this case, when transmitting the feature amount to the server 8, it is possible to reduce the amount of data to be transmitted compared to when transmitting the sound information 51 and the driving information to the server 8.

[0131] As described above, according to the fifth embodiment, inspection results are shared by the server 8, and thus feature amounts can be collected in a short period of time, making it possible to detect abnormalities in the vehicle 1A at an early stage.

[0132] Embodiment 6 Next, a sixth embodiment will be described with reference to Figures 15 to 17. In the sixth embodiment, a plurality of sound collecting devices are placed at positions where vehicles pass, and sound is collected by the plurality of sound collecting devices. In the sixth embodiment, a case will be described in which a plurality of sound collecting devices 4 are placed in the vehicle inspection system 10A, but a plurality of sound collecting devices 4 may also be placed in the vehicle inspection systems 10B to 10E.

[0133] Fig. 15 is a diagram showing a first arrangement example of sound collecting devices arranged in a vehicle inspection system according to a sixth embodiment. In the sixth embodiment, a plurality of sound collecting devices 4 are arranged at one sound collection location. Fig. 15 shows a case where a plurality of sound collecting devices 4 are arranged in a direction parallel to the extension direction of the sleepers 7. That is, the plurality of sound collecting devices 4 are arranged in a line in a direction perpendicular to the track 9. Note that a plurality of sound collecting devices 4 arranged in a line may be arranged in the extension direction of the sleepers 7.

[0134] FIG. 16 is a diagram showing a second arrangement example of sound collecting devices arranged in a vehicle inspection system according to the sixth embodiment. FIG. 16 shows a case where a plurality of sound collecting devices 4 are arranged in a direction perpendicular to the extension direction of the sleepers 7. That is, the plurality of sound collecting devices 4 are arranged in a direction parallel to the tracks 9 (the traveling direction of the vehicle 1A). In the second arrangement example shown in FIG. 16, the plurality of sound collecting devices 4 are arranged linearly in an area sandwiched between two tracks 9. Note that a plurality of sound collecting devices 4 arranged linearly in a direction parallel to the tracks 9 may also be arranged in an area sandwiched between two tracks 9.

[0135] 17 is a diagram showing a third arrangement example of sound collecting devices arranged in a vehicle inspection system according to the sixth embodiment. FIG. 17 shows a case where a plurality of sound collecting devices 4 are arranged in a direction perpendicular to the extension direction of the sleepers 7. That is, the plurality of sound collecting devices 4 are arranged in a direction parallel to the tracks 9. In the third arrangement example shown in FIG. 17, the plurality of sound collecting devices 4 are arranged in two straight lines in the outer areas of the two tracks 9. Note that three or more sound collecting devices 4 arranged in a straight line in a direction parallel to the tracks 9 may be arranged in the outer areas of the two tracks 9.

[0136] 15 to 17 may be combined. For example, sound collecting devices 4 may be placed at all of the placement positions shown in Figures 15 to 17. The number of sound collecting devices 4 placed in the vehicle inspection system 10A may be any number equal to or greater than two.

[0137] Thus, according to embodiment 6, by arranging multiple sound collecting devices 4, even if one sound collecting device 4 collects noise, the vehicle inspection device 2 can inspect the vehicle 1A using sound information 51 collected by the other sound collecting devices 4, thereby improving inspection accuracy.

[0138] Furthermore, since the vehicle 1A can be inspected using the sound information 51 collected by the multiple sound collecting devices 4, the vehicle inspection device 2 can easily identify which drive equipment is in an abnormal state based on the relative positions of the axles and drive equipment. Therefore, the vehicle inspection device 2 can improve the accuracy of identifying abnormalities in the drive equipment.

[0139] Embodiment 7 Next, a seventh embodiment will be described with reference to Figures 18 to 22. In the seventh embodiment, a vehicle inspection system learns how to determine whether or not an abnormality exists, and infers the determination result based on the learning result. In the seventh embodiment, a case will be described in which a learning device and an inference device are applied to vehicle inspection system 10A, but a learning device and an inference device may also be applied to vehicle inspection systems 10B to 10E.

[0140] 18 is a diagram showing the configuration of a learning device provided in the vehicle inspection system according to the seventh embodiment. The learning device 70 may be located anywhere. The learning device 70 may be located within the vehicle inspection device 2 or the server 8, or may be connected to the vehicle inspection device 2 or the server 8.

[0141] The learning device 70 has a data acquisition unit 71, which is a first data acquisition unit, and a model generation unit 72. The data acquisition unit 71 acquires vehicle data DA1 and inspection results DA2 from outside the learning device 70. The vehicle data DA1 includes at least one of sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A. Below, a case will be described in which the vehicle data DA1 includes all of the sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the vehicle 1A. Furthermore, the inspection results DA2 are data of a determination result indicating whether the driving equipment of the vehicle 1A is abnormal or normal.

[0142] The running conditions include the gradient of the track 9 when the vehicle 1A was running, the degree of deterioration of the track 9, etc. The inspection history includes the presence or absence of abnormalities in the periodic inspection (inspection results) and data on vibrations when abnormal or normal.

[0143] The data acquisition unit 71 acquires driving conditions, driving history, etc. from the communication device 12, the sound collection device 4, the sound storage device 3, the vehicle inspection device 2, or the server 8. The data acquisition unit 71 also acquires vibration data, etc. from the sound collection device 4, the sound storage device 3, the vehicle inspection device 2, or the server 8. The data acquisition unit 71 also acquires inspection history from an inspection device that performed an inspection of the vehicle 1A, etc. The data acquisition unit 71 also acquires inspection results DA2 from the vehicle inspection device 2 or the server 8. The data acquisition unit 71 sends the acquired vehicle data DA1 and inspection results DA2 to the model generation unit 72.

[0144] The model generation unit 72 learns the inspection result DA2 corresponding to the vehicle data DA1 based on the learning data created based on the combination of the vehicle data DA1 and the inspection result DA2 sent from the data acquisition unit 71. In other words, the model generation unit 72 learns the inspection result DA2 when the state of the vehicle 1A is the vehicle data DA1 based on the learning data created based on the combination of the vehicle data DA1 and the inspection result DA2. That is, the model generation unit 72 generates a trained model 73 that infers the inspection result DA2 from the state of the vehicle 1A, such as the sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A. Here, the learning data is data in which the vehicle data DA1 and the inspection result DA2 are associated with each other. The model generation unit 72 outputs the generated trained model 73. The trained model storage unit 75 stores the trained model 73 output from the model generation unit 72.

[0145] The model generation unit 72 can use known algorithms such as supervised learning, unsupervised learning, reinforcement learning, etc. As an example, a case where a neural network is applied to the learning algorithm used by the model generation unit 72 will be described.

[0146] The model generation unit 72 learns the inspection result DA2 corresponding to the vehicle data DA1, for example, by so-called supervised learning in accordance with a neural network model. Here, supervised learning refers to a technique in which data sets (learning data) of inputs and results (labels) are provided to the learning device 70, and the learning device 70 learns the features contained in the learning data and infers the results from the inputs.

[0147] A neural network consists of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer or two or more layers.

[0148] Fig. 19 is a diagram illustrating a neural network used by the learning device according to the seventh embodiment. For example, in a three-layer neural network as shown in Fig. 19, when a plurality of input data are input to the input layer (X1 to X3), the values ​​are multiplied by weights W1 (w11 to w16) and input to the intermediate layer (Y1 to Y2). The results are then further multiplied by weights W2 (w21 to w26) and output from the output layer (Z1 to Z3). This output result varies depending on the values ​​of weights W1 and W2.

[0149] 18 learns the inspection result DA2 corresponding to the vehicle data DA1 by so-called supervised learning in accordance with learning data created based on a combination of the vehicle data DA1 and the inspection result DA2 acquired by the data acquisition unit 71. In other words, the neural network used by the learning device 70 learns the inspection result DA2 corresponding to the vehicle data DA1 by so-called supervised learning in accordance with the vehicle data DA1 and the inspection result DA2 created based on a combination of the first input and the second input (correct answer) acquired by the data acquisition unit 71.

[0150] That is, the neural network learns by adjusting the weights W1 and W2 so that the result output from the output layer after receiving the vehicle data DA1 as the first input approaches the second input (correct answer).

[0151] In this way, the neural network learns by inputting sound information 51 of vehicle 1A, driving conditions of vehicle 1A, driving history of vehicle 1A, and inspection history of vehicle 1A into the input layer and adjusting weights W1 and W2 so that the result output from the output layer approaches inspection result DA2. By learning the correspondence between vehicle data DA1 and inspection result DA2, the neural network generates trained model 73 that can output appropriate inspection result DA2 when vehicle data DA1 is input. In this way, learning device 70 learns trained model 73 that can output inspection result DA2, which is the correct answer, when vehicle data DA1 is input.

[0152] By performing the above-described learning, the model generation unit 72 generates and outputs a trained model 73. The trained model 73 output from the model generation unit 72 is stored in a trained model storage unit 75.

[0153] Next, a processing procedure of the process in which the learning device 70 learns the trained model 73 will be described with reference to Fig. 20. Fig. 20 is a flowchart showing the processing procedure of the learning process executed by the learning device according to the seventh embodiment.

[0154] The data acquisition unit 71 acquires learning data to be used for learning (step S210). Specifically, the data acquisition unit 71 acquires vehicle data DA1 including sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A, and inspection results DA2.

[0155] The data acquisition unit 71 may acquire the sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A simultaneously or at different times. The inspection results DA2, the sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, the inspection history of the vehicle 1A, etc. may be input to the data acquisition unit 71 in an associated manner. The data acquisition unit 71 sends the vehicle data DA1 and the inspection results DA2 to the model generation unit 72.

[0156] The model generation unit 72 executes a learning process using the vehicle data DA1 and the inspection results DA2 (step S220). Specifically, the model generation unit 72 learns the inspection results DA2 corresponding to the vehicle data DA1 by so-called supervised learning in accordance with learning data created based on a combination of the vehicle data DA1, which includes the sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A acquired by the data acquisition unit 71, and generates a learned model 73.

[0157] After generating the trained model 73, the model generation unit 72 outputs the trained model 73 to the trained model storage unit 75 (step S230). The trained model storage unit 75 stores the trained model 73 generated by the model generation unit 72.

[0158] 21 is a diagram showing the configuration of an inference device according to the seventh embodiment. The inference device 80 may be located anywhere. The inference device 80 may be located within the vehicle inspection device 2 or the server 8, or may be connected to the vehicle inspection device 2 or the server 8. The inference device 80 has a data acquisition unit 81, which is a second data acquisition unit, and an inference unit 82.

[0159] The data acquisition unit 81 acquires vehicle data DB1 from outside the inference device 80. The vehicle data DB1 is the same information as the vehicle data DA1. That is, the vehicle data DB1 includes sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A. The vehicle data DA1 is data used during learning, and the vehicle data DB1 is data used during inference. Both the vehicle data DA1 and DB1 are information about the vehicle 1A. In this way, the vehicle inspection system 10A of the seventh embodiment learns the trained model 73 based on the vehicle data DA1, which is information about the vehicle 1A, and then infers the inspection result (the inspection result DB2, described later) by applying the vehicle data DB1, which is information about the vehicle 1A, to the trained model 73. As a result, after training the trained model 73, the vehicle inspection system 10A of the seventh embodiment can infer the inspection result DB2 using the trained model 73.

[0160] The data acquisition unit 81 acquires the vehicle data DB1 in the same manner as the data acquisition unit 71. The data acquisition unit 81 sends the acquired vehicle data DB1 to the inference unit .

[0161] The inference unit 82 receives the vehicle data DB1 sent from the data acquisition unit 81. The inference unit 82 also reads out the learned model 73 from the learned model storage unit 75. The inference unit 82 uses the learned model 73 to infer the inspection result DB2 corresponding to the vehicle data DB1. That is, by inputting the vehicle data DB1 including the sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A acquired by the data acquisition unit 81 into the learned model 73, the inference unit 82 can output the appropriate inspection result DB2 inferred from the vehicle data DB1 including the sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A.

[0162] Furthermore, in the seventh embodiment, the case has been described in which the inference device 80 outputs an appropriate test result DB2 using the trained model 73 trained by the model generation unit 72 of the learning device 70, but the inference device 80 may acquire the trained model 73 from another learning device. In this case, the inference device 80 outputs an appropriate test result DB2 based on the trained model 73 acquired from the other learning device, etc.

[0163] Next, a processing procedure of the inference device 80 inferring the test result DB2 using the trained model 73 will be described with reference to Fig. 22. Fig. 22 is a flowchart showing the processing procedure of the inference processing executed by the inference device according to the seventh embodiment.

[0164] The data acquisition unit 81 acquires inference data to be used for inferring the inspection result DB2 (step S310). Specifically, the data acquisition unit 81 acquires vehicle data DB1 including sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A. The data acquisition unit 81 sends the vehicle data DB1 to the inference unit 82. The inference unit 82 acquires the vehicle data DB1 from the data acquisition unit 81, and acquires the learned model 73 from the learned model storage unit 75.

[0165] The inference unit 82 inputs vehicle data DB1 including sound information 51 of the vehicle 1A, the driving conditions of the vehicle 1A, the driving history of the vehicle 1A, and the inspection history of the vehicle 1A into the trained model 73 (step S320), and obtains appropriate inspection results DB2.

[0166] The inference unit 82 outputs data inferred using the trained model 73 and the vehicle data DB1 (step S330). Specifically, the inference unit 82 outputs the appropriate inspection result DB2 obtained by the trained model 73 to an external device such as a display device.

[0167] The display device displays the inspection results DB2 corresponding to the vehicle data DB1, thereby allowing the user to refer to the inspection results DB2 corresponding to the vehicle data DB1.

[0168] In the seventh embodiment, an example has been described in which the model generation unit 72 uses a supervised learning algorithm as a learning algorithm, but the learning algorithm used by the model generation unit 72 is not limited to a supervised learning algorithm. The model generation unit 72 can also apply a reinforcement learning algorithm, an unsupervised learning algorithm, a semi-supervised learning algorithm, etc., in addition to a supervised learning algorithm.

[0169] The model generation unit 72 may also learn an appropriate inspection result DA2 corresponding to the vehicle data DA1 in accordance with learning data created by multiple vehicle inspection systems 10A. The model generation unit 72 may also perform learning using learning data acquired from multiple vehicle inspection systems 10A used in the same area, or may perform learning using learning data acquired from multiple vehicle inspection systems 10A operating independently in different areas.

[0170] In addition, vehicle inspection systems 10A that collect learning data may be added or removed from the targets at any time. Furthermore, a learning device 70 that has learned appropriate inspection results DA2 corresponding to vehicle data DA1 for a certain vehicle inspection system 10A may be applied to another vehicle inspection system 10A, and the appropriate inspection results DA2 corresponding to vehicle data DA1 for the other vehicle inspection system 10A may be re-learned and updated.

[0171] Deep learning, which learns to extract features themselves, can also be used as the learning algorithm of the model generation unit 72. The model generation unit 72 may also perform machine learning according to other known methods, such as genetic programming, functional logic programming, or support vector machines.

[0172] The learning device 70 may learn a deterioration detection threshold corresponding to the vehicle data DA1. In this case, the vehicle data DA1 and the deterioration detection threshold are input to the learning device 70. The learning device 70 learns the deterioration detection threshold corresponding to the vehicle data DA1 based on learning data created based on a combination of the vehicle data DA1 and the deterioration detection threshold. The inference device 80 also uses the trained model 73 to infer a deterioration detection threshold corresponding to the vehicle data DB1.

[0173] Furthermore, the above-described learning device 70, inference device 80, and trained model storage unit 75 may be applied to the vehicle inspection systems 10B to 10E.

[0174] As described above, in the seventh embodiment, the learning device 70 generates the trained model 73, and the inference device 80 infers the inspection result DB2 from the vehicle data DB1 using the trained model 73. This enables the vehicle inspection system 10A to easily provide the user with the inspection result DB2 corresponding to the vehicle data DB1.

[0175] Here, we will explain the hardware configuration of the vehicle inspection device 2. The vehicle inspection device 2 is realized by a processing circuit. The processing circuit may be a processor and memory that executes a program stored in a memory, or may be dedicated hardware.

[0176] FIG. 23 is a diagram illustrating an example of the configuration of a processing circuit provided in the vehicle inspection apparatus according to the first to seventh embodiments, when the processing circuit is realized by a processor and a memory. The processing circuit 90 illustrated in FIG. 23 includes a processor 91 and a memory 92. When the processing circuit 90 is configured with the processor 91 and the memory 92, each function of the processing circuit 90 is realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a vehicle inspection program and stored in the memory 92. In the processing circuit 90, each function is realized by the processor 91 reading and executing the vehicle inspection program stored in the memory 92. That is, the processing circuit 90 includes the memory 92 for storing the vehicle inspection program that results in the processing of the vehicle inspection apparatus 2 being executed. This vehicle inspection program can also be said to be a program that causes the vehicle inspection apparatus 2 to execute each function realized by the processing circuit 90. This vehicle inspection program may be provided by a storage medium on which the vehicle inspection program is stored, or by other means such as a communication medium.

[0177] The vehicle inspection program can also be said to be a program that causes the vehicle inspection device 2 to execute the processes of steps S50 and S60 in Fig. 8. In other words, the vehicle inspection program can also be said to be a program that causes the vehicle inspection device 2 to execute the steps of reading out sound information 51 and receiving driving information, and determining whether or not there is an abnormality in vehicle 1A based on sound information 51 and the driving information.

[0178] Here, the processor 91 is, for example, a CPU (Central Processing Unit), a processing device, an arithmetic device, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor), etc. Furthermore, the memory 92 is, for example, a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable ROM), or an EEPROM (registered trademark) (Electrically EPROM), a magnetic disk, a flexible disk, an optical disk, a compact disk, a minidisk, or a DVD (Digital Versatile Disc).

[0179] FIG. 24 is a diagram showing an example of a processing circuit provided in the vehicle inspection device according to the first to seventh embodiments, configured with dedicated hardware. The processing circuit 93 shown in FIG. 6 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof. The processing circuit 93 may be partially realized with dedicated hardware and partially realized with software or firmware. In this way, the processing circuit 93 can realize each of the above-mentioned functions by dedicated hardware, software, firmware, or a combination thereof.

[0180] The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies, or different embodiments may be combined with each other. It is also possible to omit or modify parts of the configurations as long as they do not deviate from the gist of the invention.

[0181] Various aspects of the present disclosure are summarized below as appendices.

[0182] (Appendix 1) a driving information storage device that stores driving information including a driving history of a vehicle driven by a driving machine; a vibration collecting device that collects vibrations caused by the driving equipment when the vehicle passes through a railroad track; a vibration storage device that stores vibration information that is information about the vibration; a vehicle inspection device that determines whether or not an abnormality has occurred in the driving equipment based on the running information and the vibration information for each axle position of the vehicle; A vehicle inspection system comprising: (Appendix 2) the vehicle inspection device determines whether or not an abnormality has occurred in the driving equipment based on the secular change in the vibration included in the vibration information. 1. A vehicle inspection system as described in Appendix 1. (Appendix 3) the vehicle inspection device calculates a feature amount of the vibration from the vibration information, and determines whether or not an abnormality has occurred in the driving equipment based on the feature amount. 1. A vehicle inspection system as described in Appendix 1 or 2. (Appendix 4) The driving information includes at least one of information on the route on which the vehicle has traveled, information on the distance traveled by the vehicle, information on the speed at which the vehicle has traveled, information on the notch at which the vehicle has traveled, and information on the motor current to the motor that drives the axle of the vehicle. 10. A vehicle inspection system according to any one of claims 1 to 3. (Appendix 5) further comprising a device information storage device that stores driving device information, which is information about the driving device; the vehicle inspection device determines data to be used when analyzing the vibration information based on the driving equipment information, and analyzes the vibration information using the determined data. 10. A vehicle inspection system according to any one of claims 1 to 4. (Appendix 6) The data used when analyzing the vibration information includes at least one of an analysis parameter used when analyzing the vibration information and a deterioration detection threshold which is a threshold to be compared with the vibration feature amount. 1. A vehicle inspection system as described in Appendix 5. (Appendix 7) a wheel passage detection device for detecting when a wheel of the vehicle passes over the track; the vehicle inspection device determines whether or not an abnormality has occurred in the driving equipment for each axle position of the vehicle based on the travel information and vibration information collected at the timing when the vehicle passes; 10. A vehicle inspection system according to any one of claims 1 to 6. (Appendix 8) Further provided is a vibration measuring device that measures vibration of the driving device, the vehicle inspection device performs a first analysis based on the vibration measured by the vibration measuring device, performs a second analysis based on the vibration information, and corrects an analysis method of the second analysis so that an analysis result of the first analysis coincides with an analysis result of the second analysis; 8. A vehicle inspection system according to any one of claims 1 to 7. (Appendix 9) the vehicle inspection device modifies at least one of an analysis parameter, a threshold value, and a noise removal method used when analyzing the vibration information as an analysis method for the second analysis; 10. A vehicle inspection system as described in Appendix 8. (Appendix 10) Further comprising a server that collects and analyzes data on the vehicle inspections from the plurality of vehicle inspection devices. 10. A vehicle inspection system according to any one of claims 1 to 9. (Appendix 11) the vehicle inspection data is a feature amount of the vibration calculated from the vibration information, or a determination result of whether or not an abnormality has occurred in the driving machine determined based on the feature amount; 11. A vehicle inspection system as described in Appendix 10. (Appendix 12) The vibration collecting device is composed of a plurality of the vehicle inspection device determines whether or not an abnormality has occurred in the driving machine based on the vibration information collected by the plurality of vibration collecting devices. 12. A vehicle inspection system according to any one of claims 1 to 11. (Appendix 13) a learning device that learns the determination result of whether or not an abnormality has occurred in the moving machine; The learning device includes a first data acquisition unit that acquires learning data including vehicle data, which is data including at least one of the vibration information, the vehicle's driving conditions, the driving information, and the vehicle's inspection history, and the determination result associated with the vehicle data; a model generation unit that generates a trained model for inferring the determination result from the vehicle data using the training data; Equipped with 13. A vehicle inspection system according to any one of claims 1 to 12. (Appendix 14) further comprising an inference device for inferring the determination result, The inference device a second data acquisition unit that acquires the vehicle data; an inference unit that infers the determination result from the vehicle data acquired by the second data acquisition unit using the trained model; Equipped with 14. A vehicle inspection system as described in Appendix 13. (Appendix 15) The vibration is a sound generated by the driving device or a vibration propagated from the driving device via a line. 15. A vehicle inspection system according to any one of claims 1 to 14. (Appendix 16) The driving information storage device is disposed in the vehicle, The vibration collection device is located on the ground. 16. A vehicle inspection system according to any one of claims 1 to 15. (Appendix 17) a first storage step in which the travel information storage device stores travel information including a travel history of a vehicle driven by a drive machine; a collecting step in which a vibration collecting device collects vibrations caused by the driving equipment when the vehicle passes through a railroad track; a second storage step in which the vibration storage device stores vibration information, which is information about the vibration; a determination step in which the vehicle inspection device determines whether or not an abnormality has occurred in the driving equipment based on the running information and the vibration information for each axle position of the vehicle; A vehicle inspection method including: (Appendix 18) a receiving step of receiving driving information including a driving history of a vehicle driven by a driving machine; an analysis step of analyzing vibration information that is information about vibrations caused by the driving equipment when the vehicle passes through a railroad track; a determination step of determining whether or not an abnormality has occurred in the driving equipment based on the running information and the vibration information for each axle position of the vehicle; A vehicle inspection program that causes a computer to execute the above. [Explanation of symbols]

[0183] 1A, 1B, 1D vehicle, 2 vehicle inspection device, 3 sound storage device, 4 sound collection device, 5 wheel passage detection device, 7 sleeper, 8 server, 9 track, 10A, 10A1 to 10A n,10B~10E Vehicle inspection system, 11 Traveling information storage device, 12 Communication device, 13 Equipment information storage device, 14 Vibration measurement device, 21 Communication unit, 22 Input unit, 23 Analysis unit, 24 Output unit, 31 Car body, 32 Bogie, 33A, 33B Motor, 34A, 34B Joint, 35A, 35B Drive unit, 36A, 36B, 37A, 37B Wheel, 38 Bearing, 39 Gear, 40~44, 51~55 Sound information, 60 Feature information, 61A~66A, 61B~66B, AX, BX Axle, 70 Learning device, 71, 81 Data acquisition unit, 72 Model generation unit, 73 Trained model, 75 Trained model memory unit, 80 Inference device, 82 Inference unit, 90 Processing circuit, 91 Processor, 92 Memory, 93 processing circuits, DA1,DB1 vehicle data, DA2,DB2 inspection results, P1 feature.

Claims

1. a running information storage device that stores running information including the running history of a train driven by the driving equipment; a sound collecting device disposed on the ground and configured to collect, outside the train, sounds generated by the driving machinery when the train passes through a track; a sound storage device that stores sound information, which is information about the sound; a vibration measuring device that is disposed on the train and measures vibrations other than the sound of the driving equipment on the train; a vehicle inspection device that determines whether or not an abnormality has occurred in the driving equipment based on the running information and the sound information for each axle position of the train; Equipped with The vehicle inspection system includes a vehicle inspection device that performs an analysis of the vibration features based on the vibration, performs an analysis of the sound features based on the sound information, and corrects the analysis method of the sound features so that the analysis results of the vibration features and the analysis results of the sound features match.

2. the vehicle inspection device determines whether or not an abnormality has occurred in the driving equipment based on the aging change of the sound included in the sound information. The vehicle inspection system of claim 1 .

3. the vehicle inspection device calculates a feature amount of the sound from the sound information, and determines whether or not an abnormality has occurred in the driving equipment based on the feature amount of the sound. The vehicle inspection system of claim 1 .

4. The running information includes at least one of information on the route on which the train has traveled, information on the distance traveled by the train, information on the speed at which the train has traveled, information on the notch at which the train has traveled, and information on the motor current to the motor that drives the axles of the train. The vehicle inspection system of claim 1 .

5. further comprising a device information storage device that stores driving device information, which is information about the driving device; the vehicle inspection device determines data to be used when analyzing the sound information based on the driving equipment information, and analyzes the sound information using the determined data. The vehicle inspection system of claim 1 .

6. The data used when analyzing the sound information includes at least one of an analysis parameter used when analyzing the sound information and a deterioration detection threshold which is a threshold to be compared with the feature quantity of the sound.

6. The vehicle inspection system of claim 5.

7. a wheel passing detection device for detecting when the wheels of the train have passed over the track; the vehicle inspection device determines whether or not an abnormality has occurred in the driving equipment for each axle position of the train based on the running information and sound information collected at the timing when the train passes; The vehicle inspection system of claim 1 .

8. the vehicle inspection device modifies at least one of an analysis parameter, a threshold, and a noise removal method used when analyzing the sound information as an analysis method for analyzing the sound feature quantity; 8. The vehicle inspection system of claim 7.

9. further comprising a server that collects and analyzes the train inspection data from the plurality of vehicle inspection devices; The vehicle inspection system of claim 1 .

10. The train inspection data is the sound feature calculated from the sound information, or a determination result of whether or not an abnormality has occurred in the driving equipment, determined based on the sound feature.

10. The vehicle inspection system of claim 9.

11. The sound collecting device comprises a plurality of the vehicle inspection device determines whether or not an abnormality has occurred in the driving machine based on the sound information collected by the plurality of sound collecting devices. The vehicle inspection system of claim 1 .

12. a learning device that learns the determination result of whether or not an abnormality has occurred in the moving machine; The learning device includes a first data acquisition unit that acquires learning data including vehicle data, which is data including at least one of the sound information, the running conditions of the train, the running information, and the inspection history of the train, and the determination results associated with the vehicle data; a model generation unit that generates a trained model for inferring the determination result from the vehicle data using the training data; Equipped with The vehicle inspection system of claim 1 .

13. further comprising an inference device for inferring the determination result, The inference device a second data acquisition unit that acquires the vehicle data; an inference unit that infers the determination result from the vehicle data acquired by the second data acquisition unit using the trained model; Equipped with 13. The vehicle inspection system of claim 12.

14. The sound is a sound generated by the driving device. The vehicle inspection system of claim 1 .

15. The running information storage device is disposed on the train.

15. A vehicle inspection system according to any one of claims 1 to 14.

16. a first storage step in which the running information storage device stores running information including the running history of a train driven by a driving machine; a collecting step in which a sound collecting device disposed on the ground collects, outside the train, sounds generated by the driving equipment when the train passes through a track; a second storage step in which a sound storage device stores sound information, which is information about the sound; a measuring step in which a vibration measuring device arranged on the train measures vibrations other than the sound of the driving equipment on the train; a determination step in which a vehicle inspection device determines, for each axle position of the train, based on the running information and the sound information, whether or not an abnormality has occurred in the driving equipment; Including, In the determination step, the vehicle inspection device performs an analysis of the vibration features based on the vibration, performs an analysis of the sound features based on the sound information, and corrects the analysis method of the sound features so that the analysis result of the vibration features and the analysis result of the sound features coincide.

17. a receiving step of receiving running information including a running history of a train driven by a driving machine and a measurement result of vibrations other than the sound of the driving machine measured on the train by a vibration measuring device arranged on the train; a receiving step of receiving sound information, the sound information being information on the sound caused by the driving machinery when the train passes through a track, collected outside the train by a sound collecting device arranged on the ground; a determination step of determining, for each axle position of the train, based on the running information and the sound information, whether or not an abnormality has occurred in the driving equipment; on the computer, In the determination step, the vehicle inspection program performs an analysis of the vibration features based on the vibration measurement results, performs an analysis of the sound features based on the sound information, and corrects the analysis method of the sound features so that the analysis results of the vibration features and the analysis results of the sound features match.

Images