阅读说明：本技术 一种轨道车辆轴箱轴承便携式道旁声学检测系统及方法 (Portable roadside acoustic detection system and method for axle box bearing of railway vehicle ) 是由 胡定玉 童震霆 师蔚 廖爱华 顾佶智 于 2021-07-21 设计创作，主要内容包括：本发明涉及一种轨道车辆轴箱轴承便携式道旁声学检测系统及方法,该监测系统包括红外线传感器装置、传声器阵列装置、机器视觉速度检测装置和数据采集处理装置,红外线传感器装置包括两对分别设置在轨道两侧的红外传感器；传声器阵列装置包括一对设置在轨道两侧的折叠式传声器阵列,轨道每侧的折叠式传声器阵列位于两个红外传感器之间；机器视觉速度检测装置包括设置在轨道一侧的折叠式传声器阵列上的工业相机,用以在车速检测触发后拍摄车辆图像；数据采集处理装置分别与红外线传感器装置、传声器阵列装置和机器视觉速度检测装置连接,与现有技术相比,本发明具有方便携带、提高了声信号的信噪比以及减小轴承故障引起的安全隐患等优点。(The invention relates to a portable roadside acoustic detection system and a method for a bearing of an axle box of a rail vehicle, wherein the monitoring system comprises an infrared sensor device, a microphone array device, a machine vision speed detection device and a data acquisition and processing device, wherein the infrared sensor device comprises two pairs of infrared sensors which are respectively arranged on two sides of a rail; the microphone array device comprises a pair of folding microphone arrays arranged on two sides of the track, and the folding microphone array on each side of the track is positioned between the two infrared sensors; the machine vision speed detection device comprises an industrial camera arranged on a folding microphone array on one side of the track and used for shooting a vehicle image after the vehicle speed detection is triggered; compared with the prior art, the invention has the advantages of convenient carrying, improved signal-to-noise ratio of acoustic signals, reduced potential safety hazard caused by bearing faults and the like.)

1. A portable wayside acoustic detection system for a rail vehicle axle box bearing, the monitoring system comprising:

infrared sensor device: the system comprises two pairs of infrared sensors which are respectively arranged on two sides of a track and used as triggers for detecting the speed of a railway vehicle when the railway vehicle passes in two directions;

microphone array device: the foldable microphone arrays are arranged on two sides of the track, and the foldable microphone arrays on each side of the track are positioned between two infrared sensors and used for collecting acoustic signals of the axle box bearing;

machine vision speed detection device: the system comprises an industrial camera arranged on a folding microphone array on one side of a track, and is used for shooting a vehicle image after vehicle speed detection is triggered so as to acquire vehicle speed information;

the data acquisition and processing device comprises: and the infrared sensor device, the microphone array device and the machine vision speed detection device are respectively connected for acquiring data and carrying out fault detection.

2. A rail vehicle axlebox bearing portable wayside acoustic detection system according to claim 1, it is characterized in that the folding microphone array comprises a microphone array bracket made of aluminum alloy material and a plurality of microphones which are parallelly and equidistantly arranged on the microphone array bracket and are used for collecting bearing full-period sound signals, the middle part of the folding microphone bracket is provided with a rotating shaft for folding and unfolding, the lower part of the rotating shaft is fixed on a telescopic tripod through a fastening screw, the telescopic tripod is provided with a positioning hole, an infrared laser pen is vertically inserted into the positioning hole, and a telescopic tripod is fixed when the light spot is projected on the upper edge of the steel rail, so as to adjust the height of the folded microphone array, the folded microphone array is adjusted to a horizontal position by a level gauge, and the center of the folded microphone array is at the same height from the plane of the track and the center of the axlebox bearing.

3. The portable roadside acoustic detection system of a rail vehicle axlebox bearing according to claim 1, wherein the data acquisition and processing device comprises a machine vision speed measurement module, an acoustic signal acquisition module, a bearing fault diagnosis module and a data processor;

the machine vision speed detection module is connected with the data processor and used for transmitting the vehicle image shot by the industrial camera to the data processor after analog-to-digital conversion and then measuring the speed of the rail vehicle;

the acoustic signal acquisition module is connected with the data processor and used for sending the acoustic signals acquired by the microphone array to the data processor for fault characteristic frequency extraction after analog-to-digital conversion;

the bearing fault diagnosis module is connected with the acoustic signal acquisition module and used for judging whether the axle box bearing of the rail vehicle has a fault or not according to the fault characteristic frequency;

the data processor is connected with the server through 5G communication equipment and used for uploading data.

4. The portable roadside acoustic detection system of the journal bearing of the rail vehicle as claimed in claim 1, wherein the infrared sensor device comprises a first infrared sensor and a third infrared sensor fixed on one side of the track through a bracket, and a second infrared sensor and a fourth infrared sensor fixed on the other side of the track, the first infrared sensor and the third infrared sensor are respectively connected with the first data processor, and the second infrared sensor and the fourth infrared sensor are respectively connected with the second data processor, so that the rail vehicle on the same track can trigger normal start-up operation when passing through from two directions, and the running direction of the vehicle can be judged, thereby realizing bidirectional vehicle receiving.

5. The portable wayside acoustic detection system of a rail vehicle axlebox bearing according to claim 1, further comprising two outdoor power sources for powering the entire system.

6. Method for detecting a portable wayside acoustic detection system for a railway vehicle axlebox bearing according to any one of the claims 1 to 5, comprising the following steps:

step S1: when a rail vehicle passes through one group of infrared sensors, the infrared sensors send a pulse to the machine vision speed detection device after the detected distance data change, and an industrial camera is triggered to start working;

step S2: the industrial camera positions the axle box for shooting, calculates the vehicle speed through a machine vision speed detection module, and transmits the vehicle speed information to a server through 5G communication equipment;

step S3: the microphone array collects acoustic signals of a vehicle axle box bearing, the collected signals are transmitted to the data collection processing device to be processed and are sent to the server through 5G communication equipment, when an axle box bearing cover can be extracted in the shooting range of the industrial camera, the axle box bearing stays in a picture of the industrial camera for a period of time, and the data collection processing device intercepts the acoustic signal data in the period of time and transmits the acoustic signal data to the fault diagnosis module;

step S4: and the fault diagnosis module receives the intercepted axle box bearing acoustic signal data and then carries out fault judgment on the bearing to form a fault message and fuse the vehicle speed information and transmit the fault message to the server through the 5G communication equipment.

7. The portable roadside acoustic detection method of the rail vehicle axle box bearing according to claim 6, wherein the speed detection process of the machine vision speed detection module in the step S2 is specifically as follows:

step S201, when a rail vehicle passes through a group of infrared sensors, triggering a system to start working by an infrared sensor device, and continuously shooting a plurality of pictures by an industrial camera;

step S202, an image acquisition card receives an analog image signal, digitizes the analog image signal through analog-to-digital conversion, or directly receives digital video data digitized by a camera;

step S203, transmitting the digital image to a data processor to process and identify the image, and obtaining measurement data;

and S204, extracting the axle box contour of each picture by using an edge detection algorithm, identifying the center of the contour, establishing a coordinate system to position the center of the axle box contour, calculating the vehicle speed according to the center coordinates of the contour of each picture, displaying speed information data through a display screen on a data processor, and uploading the speed information to a server.

8. The portable roadside acoustic detection method of the rail vehicle axle box bearing according to claim 6, wherein the fault diagnosis module in the step S4 performs a fault diagnosis process specifically as follows:

step S401, intercepting axle box bearing acoustic signal data;

step S402, a delay summation beam forming algorithm is adopted to carry out weighted summation and filtering on the collected output signals of each microphone, and finally, the acoustic signals in the expected direction are output to form a beam;

step S403, enveloping the signals after the wave beam is formed, and judging whether the bearing fault characteristic frequency occurs or not, so as to judge whether the axle box bearing has a fault or not;

and S404, if the fault exists, forming a fault message, and fusing vehicle speed information and transmitting the vehicle speed information to a server.

9. The portable wayside acoustic detection method for rail vehicle axle box bearings according to claim 8, wherein the delay-and-sum beam forming algorithm in step S402 forms a beam from the far-field plane wave transmitted from the axle box bearing, adds a delay before each microphone receives the signal, aligns the acoustic signals from the axle box bearing direction before the summation, suppresses the interference signals in the non-target direction, and enhances the acoustic signals in the target direction to acquire more accurate rail vehicle axle box bearing acoustic signals.

10. The portable roadside acoustic detection method of the rail vehicle axle box bearing according to claim 6, wherein the data acquisition and processing process of the data acquisition and processing device in the step S3 is specifically as follows:

step S301, when no train passes, the system is in a standby state, and only signals of four infrared sensors are subjected to real-time data acquisition and processing in the standby state;

step S302, when a rail vehicle is about to pass through, data collected by one group of infrared sensors are changed, so that the direction of the coming vehicle can be determined, and the whole system is triggered to start data collection;

step S303, the industrial camera starts to continuously shoot a plurality of pictures, and the microphone array device collects acoustic signals;

step S304, an image acquisition card receives an analog image signal, digitalizes the analog image signal through analog-to-digital conversion, an acoustic signal acquisition module converts the acquired acoustic signal into a digital signal through A/D (analog-to-digital) conversion, and then sends the digital signal to a data processor of a data acquisition and processing device for data processing and analysis, and the data processor processes the acquired acoustic signal data and obtains vehicle instantaneous speed data and then sends the vehicle instantaneous speed data to a server through 5G communication equipment;

and S305, when the industrial camera shoots the axle box bearing cover, and the circle center of the bearing cover is extracted, the axle box bearing acoustic signal data is intercepted, when the bearing cover leaves the shooting range of the industrial camera, the data interception is stopped, and when the rail vehicle passes through another group of infrared sensors, the intercepted data is transmitted to a fault diagnosis module.

Technical Field

The invention relates to the technical field of railway vehicle bearing roadside acoustic fault diagnosis, in particular to a portable roadside acoustic detection system and method for a railway vehicle axle box bearing.

Background

The axle box bearing is an important component of a locomotive running gear and is one of the parts with the highest fault probability. In an extreme operating environment, the condition of a hot shaft, a burning shaft, a shaft cutting and the like can be caused by sudden damage of the axle box bearing, so that a major traffic accident is caused, the life safety of passengers is threatened, and major economic loss is caused. Therefore, the use safety of the axle box bearing is improved, certain detection measures are taken, the axle box bearing fault is eliminated in the bud state, and the requirement of driving safety is met.

At present, the monitoring of a bearing of a shaft box of a domestic subway vehicle is mostly realized by monitoring the shaft temperature, and the commonly used method is to paste temperature test paper on a bearing cover and realize the measurement of the bearing temperature through an infrared sensor arranged on the outer side of a track, so that the fault can be detected only when the bearing generates heat. And the bearing is monitored by using vibration detection equipment and detected by a vibration signal, but the equipment cost is high when a vibration monitoring device is additionally arranged on each axle box. And the roadside acoustic monitoring system can detect an impact signal generated by the bearing in the motion process at the early stage of the fault of the bearing, and judge the early fault. At present, the railway vehicle axle box bearing is widely used for Acoustic monitoring at home, namely a railway rolling bearing early fault rail side Acoustic diagnosis System (tracking initial Detection System). The detection mode of the TADS technology is that a magnetic sensor is arranged inside a track to carry out vehicle speed measurement and acoustic sensor triggering, and an acoustic microphone array is placed in a protective box beside the track and is a fixed monitoring facility.

The method comprises the steps that six microphones are arranged on one side of a TADS technology track, each microphone collects partial signals of a bearing in one period, then signals of each microphone are cut and calibrated to obtain full-period signals of the bearing, in order to obtain high-quality sound source signals, the microphones are close to the side of the track, the distance between the microphones is large, the number of the microphones adopted by the TADS technology is small, the distance between the microphones is large, and when a single microphone fails, an error result can be analyzed.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a portable roadside acoustic detection system and method for a bearing of an axle box of a railway vehicle.

The purpose of the invention can be realized by the following technical scheme:

a rail vehicle axle box bearing portable wayside acoustic detection system, the monitoring system comprising:

infrared sensor device: the system comprises two pairs of infrared sensors which are respectively arranged on two sides of a track and used as triggers for detecting the speed of a railway vehicle when the railway vehicle passes in two directions;

microphone array device: the foldable microphone arrays are arranged on two sides of the track, and the foldable microphone arrays on each side of the track are positioned between two infrared sensors and used for collecting acoustic signals of the axle box bearing;

machine vision speed detection device: the system comprises an industrial camera arranged on a folding microphone array on one side of a track, and is used for shooting a vehicle image after vehicle speed detection is triggered so as to acquire vehicle speed information;

the data acquisition and processing device comprises: and the infrared sensor device, the microphone array device and the machine vision speed detection device are respectively connected for acquiring data and carrying out fault detection.

The foldable microphone array comprises a microphone array support made of aluminum alloy materials and a plurality of microphones which are parallelly and equidistantly arranged on the microphone array support and used for collecting bearing complete-period sound signals, a rotating shaft used for opening, closing, folding and retracting is arranged in the middle of the foldable microphone support, the lower portion of the rotating shaft is fixed on a telescopic tripod through a fastening screw, a positioning hole is formed in the telescopic tripod, an infrared laser pen is vertically inserted into the positioning hole, the telescopic tripod is fixed when a shot light spot is located on the upper edge of a steel rail, the height of the foldable microphone array is adjusted, the foldable microphone array is adjusted to be in a horizontal position through a level, and the height of the center of the foldable microphone array from a track plane to the center of an axle box bearing is equal.

The data acquisition and processing device comprises a machine vision speed measurement module, an acoustic signal acquisition module, a bearing fault diagnosis module and a data processor;

the machine vision speed detection module is connected with the data processor and used for transmitting the vehicle image shot by the industrial camera to the data processor after analog-to-digital conversion and then measuring the speed of the rail vehicle;

the acoustic signal acquisition module is connected with the data processor and used for sending the acoustic signals acquired by the microphone array to the data processor for fault characteristic frequency extraction after analog-to-digital conversion;

the bearing fault diagnosis module is connected with the acoustic signal acquisition module and used for judging whether the axle box bearing of the rail vehicle has a fault or not according to the fault characteristic frequency;

the data processor is connected with the server through 5G communication equipment and used for uploading data.

The infrared sensor device comprises a first infrared sensor and a third infrared sensor which are fixed on one side of the track through a support, and a second infrared sensor and a fourth infrared sensor which are fixed on the other side of the track, wherein the first infrared sensor and the third infrared sensor are respectively connected with a first data processor, and the second infrared sensor and the fourth infrared sensor are respectively connected with a second data processor, so that the track vehicle on the same track can be normally started through homoenergetic triggering from two directions, the running direction of the vehicle is judged, and the two-way vehicle receiving is realized.

The system also includes two outdoor power supplies for powering the entire system.

The invention also provides a detection method of the portable roadside acoustic detection system for the bearing of the axle box of the rail vehicle, which comprises the following steps:

step S1: when a rail vehicle passes through one group of infrared sensors, the infrared sensors send a pulse to the machine vision speed detection device after the detected distance data change, and an industrial camera is triggered to start working;

step S2: the industrial camera positions the axle box for shooting, calculates the vehicle speed through a machine vision speed detection module, and transmits the vehicle speed information to a server through 5G communication equipment;

step S3: the microphone array collects acoustic signals of a vehicle axle box bearing, the collected signals are transmitted to the data collection processing device to be processed and are sent to the server through 5G communication equipment, when an axle box bearing cover can be extracted in the shooting range of the industrial camera, the axle box bearing stays in a picture of the industrial camera for a period of time, and the data collection processing device intercepts the acoustic signal data in the period of time and transmits the acoustic signal data to the fault diagnosis module;

step S4: and the fault diagnosis module receives the intercepted axle box bearing acoustic signal data and then carries out fault judgment on the bearing to form a fault message and fuse the vehicle speed information and transmit the fault message to the server through the 5G communication equipment.

The process of the machine vision speed detection module performing speed detection in the step S2 specifically includes:

step S201, when a rail vehicle passes through a group of infrared sensors, triggering a system to start working by an infrared sensor device, and continuously shooting a plurality of pictures by an industrial camera;

step S202, an image acquisition card receives an analog image signal, digitizes the analog image signal through analog-to-digital conversion, or directly receives digital video data digitized by a camera;

step S203, transmitting the digital image to a data processor to process and identify the image, and obtaining measurement data;

and S204, extracting the axle box contour of each picture by using an edge detection algorithm, identifying the center of the contour, establishing a coordinate system to position the center of the axle box contour, calculating the vehicle speed according to the center coordinates of the contour of each picture, displaying speed information data through a display screen on a data processor, and uploading the speed information to a server.

The process of performing fault diagnosis by the fault diagnosis module in step S4 is specifically as follows:

step S401, intercepting axle box bearing acoustic signal data;

step S402, a delay summation beam forming algorithm is adopted to carry out weighted summation and filtering on the collected output signals of each microphone, and finally, the acoustic signals in the expected direction are output to form a beam;

step S403, enveloping the signals after the wave beam is formed, and judging whether the bearing fault characteristic frequency occurs or not, so as to judge whether the axle box bearing has a fault or not;

and S404, if the fault exists, forming a fault message, and fusing vehicle speed information and transmitting the vehicle speed information to a server.

The delay-sum beam forming algorithm in step S402 forms a beam from the far-field plane wave transmitted from the axle box bearing, and adds a delay before each microphone receives a signal, so that the acoustic signals from the axle box bearing direction are aligned before the summation, the interference signals in the non-target direction are suppressed, and the acoustic signals in the target direction are enhanced, so as to acquire more accurate acoustic signals of the axle box bearing of the rail vehicle.

The data acquisition and processing process of the data acquisition and processing device in the step S3 specifically includes:

step S301, when no train passes, the system is in a standby state, and only signals of four infrared sensors are subjected to real-time data acquisition and processing in the standby state;

step S302, when a rail vehicle is about to pass through, data collected by one group of infrared sensors are changed, so that the direction of the coming vehicle can be determined, and the whole system is triggered to start data collection;

step S303, the industrial camera starts to continuously shoot a plurality of pictures, and the microphone array device collects acoustic signals;

step S304, an image acquisition card receives an analog image signal, digitalizes the analog image signal through analog-to-digital conversion, an acoustic signal acquisition module converts the acquired acoustic signal into a digital signal through A/D (analog-to-digital) conversion, and then sends the digital signal to a data processor of a data acquisition and processing device for data processing and analysis, and the data processor processes the acquired acoustic signal data and obtains vehicle instantaneous speed data and then sends the vehicle instantaneous speed data to a server through 5G communication equipment;

and S305, when the industrial camera shoots the axle box bearing cover, and the circle center of the bearing cover is extracted, the axle box bearing acoustic signal data is intercepted, when the bearing cover leaves the shooting range of the industrial camera, the data interception is stopped, and when the rail vehicle passes through another group of infrared sensors, the intercepted data is transmitted to a fault diagnosis module.

Compared with the prior art, the invention has the following advantages:

the microphone array used by the invention has better directivity than a single microphone, and the main beam is generated by the microphone array in the expected direction, so that the acoustic signal of the axle box bearing of the rail vehicle can be more accurately collected, the noise interference is reduced by utilizing the beam forming technology, and the signal-to-noise ratio of the acoustic signal is improved.

The invention can detect the bearing fault in the running process of the rail vehicle, and can inform an overhaul department when the bearing fault occurs to process the fault in time, thereby reducing the potential safety hazard caused by the fault and having important significance for the safe running of railways in China.

Thirdly, the portable bearing roadside acoustic detection device designed by the invention is movable, the whole system is convenient to carry, and all equipment can be stored in a 32-inch draw-bar box; the triggering and speed measuring module does not need to invade a vehicle clearance, any equipment does not need to be installed inside the track, the track to be detected can be changed according to actual conditions, one set of detection equipment shared by a plurality of subway lines is realized, and the detection cost is reduced.

Drawings

Fig. 1 is a schematic structural view of a folded microphone array;

FIG. 2 is a schematic diagram of a delay-and-sum beamforming spatial filter;

FIG. 3 is a schematic diagram of a machine vision speed detection module;

FIG. 4 is a schematic diagram of the structure of a roadside acoustic monitoring module;

FIG. 5 is a schematic view of a wayside monitoring system layout;

FIG. 6 is a flow chart of a data acquisition and storage module;

FIG. 7 is a fault diagnosis module flow diagram;

FIG. 8 is a plane wave model diagram.

The notation in the figure is:

1. industrial camera, 2, first microphone array, 21, microphone array holder, 31, first microphone, 32, second microphone, 33, third microphone, 34, fourth microphone, 35, fifth microphone, 36, sixth microphone, 37, seventh microphone, 38, eighth microphone, 39, ninth microphone, 310, tenth microphone, 4, positioning hole, 5, telescopic tripod, 6, ground, 7, second microphone array, 81, first infrared sensor, 82, second infrared sensor, 83, third infrared sensor, 84, fourth infrared sensor, 91, first data processor, 92, second data processor.

Detailed Description

The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.

Examples

The invention provides a portable roadside acoustic detection system of a bearing of an axle box of a railway vehicle, as shown in figure 1, the total length of a folding microphone array in the embodiment is 1m, the width is 4cm, the thickness is 3cm, in order to obtain higher resolution, 10 microphones are arranged, the spacing of array elements is 10cm, a rotating shaft is arranged at the center of the array, the array can be folded and put into a luggage case along the rotating shaft after the test is finished, the structure of the array is similar to that of a door, the upper half part of the rotating shaft is connected with the left half part and the right half part of the array through a hinge or a hinge and used for mutually folding, the lower half part is provided with threads and used for being connected with a nut for fixing, the array is arranged on a telescopic tripod, aluminum alloy is selected as a material of the array, the weight of the array is reduced as much as possible, a positioning hole of an infrared laser pen is reserved on the telescopic tripod 40cm below the center of the array, the infrared laser pen is vertically inserted into the positioning hole during the test, the height is adjusted to enable the light spot emitted by the laser pen to be just on the upper edge of the steel rail, at the moment, the telescopic tripod is fixed, and the microphone array is adjusted to be in a horizontal position by the level gauge.

Fig. 2 is a schematic diagram of a delay-sum beam forming spatial filter, in which, since the microphone array is far away from the axle box bearing, the acoustic wave transmitted from the axle box bearing is assumed to be a far-field plane wave, and a far-field plane wave model considers only phase differences and ignores amplitude differences for the acoustic source signal received by the microphone, and beam forming needs to be implemented by the microphone array, if all the output signals of the microphone are summed as an array output result, the maximum amplitude output will be obtained when the signal is perpendicular to the array surface, because the signals arrive at the same time, and the signals between all the channels are enhanced by each other; and when the signals come from the direction which is not vertical to the array surface, the amplitude of the output signals is smaller because of the time difference of the signals received by the microphones. The basic idea of delay-and-sum is to use delay-and-sum to design a spatial filter, called a beamformer, that achieves the ability to steer the beam without physically moving the array. By controlling the arrays, the beam main lobe direction can be controlled by adding a delay before each microphone receives the signal, so that the acoustic signals from the axlebox bearing direction are aligned before summing.

As shown in fig. 3, when the vehicle passes the infrared sensor, the infrared sensor sends a pulse to the machine vision speed detection module through the data processor, triggering the industrial camera to start working. After the vehicle completely passes through the other infrared sensor, the camera stops scanning, and a next train waits for passing and then starts to acquire and output a new image. The image acquisition part receives analog image signals, digitizes the analog image signals through A/D, or directly receives digital video data digitized by a camera. The image acquisition part transmits the digital image to a data processor, and the data processor processes, analyzes and identifies the image to obtain a measurement result or a logic control value. The processing results control the actions, positioning, etc. of other devices. And extracting the outline of the axle box bearing cover in the image by using an edge detection algorithm, positioning the circle center of the bearing cover, measuring the speed of the train by moving the circle center position of the bearing cover, and uploading the speed information to a server.

As shown in fig. 4, the roadside acoustic monitoring system of the present invention includes a data processor, a data acquisition module, a fault diagnosis module, two sets of linear array microphone arrays, and an infrared sensor. The data processor is responsible for integrating and processing data and sending the message to the server; the system is characterized in that a touch screen is arranged on a data acquisition unit, after the distance between an infrared sensor and the center of a microphone array is measured and the distance between the infrared sensor and the array is short, the two data need to be input from the touch screen, four infrared sensors are used for receiving vehicles in two directions, the rail vehicles can pass through from left to right and can also pass through from right to left on the same track, the system can be started normally under the two conditions, and the running direction of the vehicles is judged. The two groups of microphone arrays are respectively arranged on two sides of the rail in an aligned mode, and the two rows of wheel bearings are simultaneously detected, so that the collection of the sound signals of the bearing of the axle box is completed. The machine vision speed detection module is responsible for taking pictures of axle boxes of the vehicle, extracting the contour of the bearing cover by using an edge detection algorithm, and measuring the speed of the vehicle through the position change of the circle center.

As shown in fig. 5, linear microphone arrays are vertically installed on both sides of the roadside, and one set of microphone array is provided on each side, and are assembled in such a manner as to satisfy the directivity requirements. The linear microphone array is arranged on the support and is arranged at a position which is about 3000mm away from the inner side of the steel rail, and the microphones on the array need to be numbered, so that the purpose of connecting and debugging with other equipment is facilitated. The infrared sensor is selected as the trigger sensor, because the rail vehicle can cut infrared rays when passing through the infrared sensor, pulse signals are generated, and whether the rail vehicle passes through can be judged by whether the pulse signals are generated. The infrared sensor is installed at a position about 3 meters away from the array, the infrared sensor is guaranteed to be vertically aligned with the front during installation, the direction of a coming vehicle is judged through the infrared sensor, the infrared sensor device, the two microphone arrays and the industrial camera are connected with the data processor, the vehicle can be captured by the infrared sensor when passing through, the industrial camera is started simultaneously, the positioning axle box shoots, the outline of the axle box is extracted through an edge detection algorithm, the vehicle speed is calculated, then the information is transmitted to the server through a 5G network, the information acquired by the microphone arrays is transmitted to the data processor to be processed, and then the information is transmitted to the server through 5G communication equipment, when the rail vehicle passes through another infrared sensor, the data acquisition is stopped, and the system recovers a standby state to wait for the next rail vehicle to arrive.

As shown in fig. 6, the process of acquiring and processing data by the data acquisition and processing module specifically includes:

when no train passes through, the whole system is in a standby state, signals of only four infrared sensors are subjected to real-time data acquisition and processing in the standby state, and when a railway vehicle is about to pass through, data acquired by one infrared sensor can change, so that the coming direction can be determined, and the whole system is triggered to start working. The system comprises a data acquisition and storage module and an industrial camera, wherein a machine vision detection module shoots an axle box photo, an edge detection technology is used for extracting the outline of an axle box bearing cover, the instantaneous speed of a vehicle is calculated by utilizing the position change of the outline, the industrial camera is positioned above the center of an array, the length of a track shot by a lens of the industrial camera is ensured to be 5m by focusing, when the industrial camera shoots the bearing cover and extracts the circle center of the bearing cover, the acoustic signal data are intercepted, and when the bearing cover leaves the shooting range of the industrial camera, the data interception is stopped. And after the train completely passes through the infrared sensor in the other direction, transmitting the intercepted data to a fault diagnosis module.

As shown in fig. 7, the process of performing fault diagnosis on the bearing of the axle box by the fault diagnosis module specifically includes:

firstly, receiving bearing sound signal data of a data acquisition module, processing the data by using a delay summation beam forming algorithm, then carrying out envelope processing on the sound signals, judging whether a bearing has a fault or not by judging whether the bearing has the fault or not according to the characteristic frequency of the fault of the bearing, forming a corresponding fault message if the bearing has the fault, and fusing speed information for uploading; and if the fault does not exist, not uploading the information.

As shown in fig. 8, the microphone array of the present invention corresponds to x in the figure, the sound wave comes in the form of plane wave, and first reaches the first microphone on the left side of the array, the microphone interval of the present invention is 0.1m, and the time delay between the first microphone and the second microphone is

Wherein c is the sound velocity, theta is the included angle between the plane wave direction and the direction perpendicular to the array direction and is a quantity related to the vehicle speed, the collection of the sound signal is started after the infrared sensor is triggered, a touch screen is arranged on the data acquisition unit, the distance between the infrared sensor and the center of the array and the distance between the infrared sensor and the track which is close to the array are input from the touch screen, the distance between the infrared sensor and the center of the microphone array is s and is set to be about 3m, taking the case that a train passes through the array from left to right, the distance between the track which is close to the microphone array is d, the range is 3-5 m, and the expression of theta is as follows:

where n is the sampling rate of the acquisition device, i.e., the number of times data is acquired a second.

While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.