CN110758456A

CN110758456A - Wheel rail health state monitoring system and method

Info

Publication number: CN110758456A
Application number: CN201911084741.1A
Authority: CN
Inventors: 郝秋实; 王艳; 沈毅; 章欣; 王康伟
Original assignee: Harbin Institute of Technology
Current assignee: Harbin Institute of Technology
Priority date: 2019-11-08
Filing date: 2019-11-08
Publication date: 2020-02-07
Anticipated expiration: 2039-11-08
Also published as: CN110758456B

Abstract

The invention relates to a system and a method for monitoring the health state of a wheel rail, comprising the following steps: a plurality of sensors mounted on the side of the rail, wherein each sensor is configured to collect a signal from that location; the system comprises a plurality of signal processing units, a plurality of monitoring units and a plurality of wheel and rail monitoring units, wherein the signal processing units are arranged to be capable of processing signals acquired by a plurality of sensors, comprise program modules such as the monitoring method, signal preprocessing, signal denoising, damage detection, damage classification, damage positioning, damage severity evaluation and wheel and rail service life prediction, and are used for realizing wheel and rail health state monitoring; the wheel track health state monitoring control center comprises a control, display and early warning unit; and data transmission between the sensors and the signal processing unit, between the signal processing units and the monitoring control center. The invention is suitable for the whole process monitoring of the health state of the wheel rail under the conditions of different wheel rail contact states, driving speeds, loads and the like.

Description

Wheel rail health state monitoring system and method

Technical Field

The invention relates to a wheel rail health state monitoring system and a method for realizing wheel rail health state monitoring in the system. The scheme of the invention is particularly suitable for the whole process monitoring of the health state of the wheel rail under the conditions of different wheel rail contact states, driving speeds, loads and the like.

Background

Wheels and rails are the basic components of a railway system, and the guarantee of the health state of the wheels and rails is of great importance to the safety of the railway. At present, wheel rail damage detection is carried out under non-operating conditions, wheel damage detection mainly depends on factory quality inspection and regular return detection, and steel rail damage detection is mainly based on a steel rail flaw detection vehicle and a hand-push type flaw detector. Therefore, the existing wheel-rail damage detection consumes a large amount of manpower and material resources, is slow in speed and low in efficiency, cannot reflect the evolution process and all information of the damage in real time, and cannot ensure the safe operation of the train under the rapid development of the high-speed railway. The wheel rail health state monitoring not only can improve the flaw detection efficiency, but also is beneficial to understanding the damage generation mechanism and the whole development process, and is convenient for the graded diagnosis and the timely treatment of the damage, so the real-time monitoring of the wheel rail health state becomes the development direction for ensuring the railway safety.

The wheel rail damage is generated along with changes of stress, displacement, vibration, sound and the like, so the purpose of monitoring the wheel rail damage can be achieved by installing the sensor. The risk of installing the sensor on the wheel running at high speed is high, the safety factor is low, but the changes of stress, displacement, vibration, sound and the like generated by damage can be transmitted not only in the wheel and the rail, but also between the wheel and the rail through the contact of the wheel and the rail, so that the sensor is installed on the side surface of the steel rail to form the installation mode with the highest reliability at present, and the health state monitoring of the infrastructure related to the wheel, the wheel and the steel rail can be simultaneously realized. At this time, the collection of the wheel and rail damage signals is performed synchronously, and the sensor cannot distinguish whether the damage signals come from the wheel or the rail. Therefore, when the same system is adopted to monitor the health state of the wheel rail, a proper monitoring method is needed to distinguish the damage signals of the wheel and the steel rail, and the wheel and the steel rail can be further processed in a targeted manner.

In addition, the mounted sensor passively receives a damage signal generated when the wheel rail damage occurs, so that the wheel rail health state monitoring is irrelevant to external conditions and only relevant to internal factors such as the type, the position, the severity and the like of the damage. Therefore, the sensor arranged on the side surface of the steel rail is suitable for monitoring the health state of the wheel rail under the conditions of different wheel rail contact states, driving speeds, loads and the like.

Disclosure of Invention

The invention aims to provide a wheel rail health state monitoring system and a wheel rail health state monitoring method, which can improve the detection efficiency of wheel rail damage, and can monitor the wheel rail health state under the conditions of different wheel rail contact states, driving speeds, loads and the like in the whole process, so as to finally ensure the safety of a wheel rail system.

The present invention will be described with reference to the accompanying drawings. A first aspect of the invention provides a system for monitoring the health of a wheel and rail, the diagram of which is shown in fig. 1, in which a wheel 102 is in rolling contact with a rail 101. The system comprises a plurality of sensors 103a-i mounted on the side of the rail 101, each sensor being arranged to acquire a signal from that location. The system further comprises a plurality of signal processing units 104 arranged to be able to process signals acquired by the plurality of sensors. In particular, it is required that the number of sensors associated with one signal processing unit is not less than 2, and the distance between two farthest spaced sensors is equivalent to the wheel circumference. The structure of the signal processing unit is shown in fig. 2, wherein one signal processing unit may include a processor 1041 and a memory 1042, which are respectively used for executing and storing program modules such as signal preprocessing, signal denoising, flaw detection, flaw classification, flaw localization, flaw severity evaluation, wheel-rail service life prediction, and the like. The system further includes a wheel track health status monitoring control center 105, the structure diagram of which is shown in fig. 3, and the wheel track health status monitoring control center includes a control unit 1051, a display unit 1052 and an early warning unit 1053, and is used for sending instructions to the signal processing unit 104, displaying monitoring results and giving early warning to dangerous states. The system further includes data transmission, either wired or wireless, between the plurality of sensors 103a-i and the signal processing units 104a-c, between the plurality of signal processing units 104a-c, and between the plurality of signal processing units 104a-c and the monitoring control center 105. Through transmission, signals collected by the sensors can be uploaded to the signal processing units related to the sensors, data exchange can be carried out between adjacent signal processing units, result data processed by the signal processing units can be uploaded to the control center, and the control center can send instructions to the signal processing units.

A flowchart of the method for monitoring the health status of the wheel track is shown in fig. 4, and the method can be used as a program module to be embedded into a signal preprocessing module or a signal denoising module in a signal processing unit, and then divided into three steps, which are described by taking the signal processing unit 104a and the associated sensors 103a-c in fig. 1 as an example, and the specific steps are as follows:

the method comprises the following steps: judging whether the damage exists

Input vehicle speed V, wheel radius R and signals s collected by sensors 103a-c_a-c(t) from s_a-c(t) judging whether or not the damage is present. When s is_a-cWhen the background noise amplitude in (t) is smaller, the existence of a damage signal can be obviously seen, if the damage signal exists, the second step is executed, if the damage signal does not exist, the second step is finished, and the monitoring method can be embedded into a signal preprocessing module in a signal processing unit; when s is_a-c(t) when the background noise amplitude is large enough to submerge the damage signal, the monitoring method can be used as an independent module behind the signal denoising module, the noise amplitude after denoising is obviously reduced, whether the damage signal exists can be judged from the signal amplitude, if the damage signal exists, the second step is executed, and if the damage signal does not exist, the second step is ended;

step two: calculating period

When the wheel runs for one circle, the ratio of the advancing distance of the wheel to the vehicle speed is the running period, so that the running period of the wheel can be calculated through the vehicle speed V and the wheel radius R

When the wheel has damage which can generate a damage signal when the wheel is in pressure contact with the steel rail, the damage signals of the sensors 103a-c in adjacent time periods can present periodicity consistent with the running period of the wheel; when there is a rail defect, the signal of the defect in the adjacent time interval of the sensors 103a-c will be independent of the operation cycle, thereforeThe damage signal s passing through two adjacent time periods_a-c(t) calculating the signal period

Wherein A is_a,t1、A_b,t1、A_c,t1Amplitude, A, of the flaw signal collected for the

sensors

103a, 103b, 103c during time period t1_a,t2、A_b,t2、A_c,t2Amplitude of the flaw signal collected for the

sensors

103a, 103b, 103c at time period t2, t2>t1, max { } is taken to be maximum,for the time corresponding to the maximum amplitude of the damage signal in sensors 103a-c during time period t1,

is the time corresponding to the maximum amplitude of the damage signal in sensors 103a-c at time period t 2;

step three: periodic comparison

Comparing the operating periods T_wAnd signal period T_sIf the signal period is approximately equal to the operation period, the signal is a wheel damage signal, and if the difference between the signal period and the operation period is larger, the signal is a steel rail damage signal. The monitoring method is further explained below with reference to the positions of the sensors 103a-c and the wheel and rail damage signals collected by the sensors 103a-c at different time intervals. Fig. 5 is a schematic diagram of a damaged wheel generating a damaged signal when passing through the sensors 103a-c, fig. 6(a) is a schematic diagram of wheel damaged signals collected by the sensors 103a-c during a time period t1, and fig. 6(b) is a schematic diagram of wheel damaged signals collected by the sensors 103a-c during a time period t 2. Setting the position of the wheel damage which is firstly in pressure contact with the steel rail as p1 (close to the sensor 103a), and intercepting a damage signal acquired by the sensors 103a-c as a signal section t 1; after the wheel rolls for one circle, the position where the pressure contact occurs for the second time is p2 (close to the sensor 103c), and the signal segment t2 is intercepted from the damage signals collected by the sensors 103 a-c. Amplitude is generated when signal is propagated in steel railThe attenuation is generated, p1 is closest to the sensor 103a, and the amplitude of the signal acquired by the sensor 103a is maximum at the time of t 1; since the signal propagates from left to right, the sensor 103a receives the impairment signal first and the sensor 103c receives the impairment signal the latest. Similarly, during the time period t2, the sensor 103c closer to p2 receives the damage signal first and has the largest amplitude, and then the signal gradually propagates from right to left to the sensor 103b and the sensor 103 a. From the periodicity of the wheel roll, it can be seen that the difference between the time of occurrence of the maximum amplitude of sensor 103c during time t2 and the time of occurrence of the maximum amplitude of sensor 103a during time t1 should be approximately equal to the operating cycle. Fig. 7 is a schematic diagram of a damaged signal generated by a damaged steel rail when the damaged signal is collected by the sensors 103a-c, fig. 8(a) is a schematic diagram of a steel rail damaged signal collected by the sensors 103a-c in a time period t1, and fig. 8(b) is a schematic diagram of a steel rail damaged signal collected by the sensors 103a-c in a time period t 2. Setting the position of the rail damage as p1 (close to the sensor 103a), and intercepting the damage signals collected by the sensors 103a-c as a signal section t 1; and after a period of time, the damage at the position p1 is further expanded to generate a damage signal, and the damage signals collected by the sensors 103a-c are intercepted to form a signal segment t 2. Since p1 is closest to sensor 103a, the signal collected by sensor 103a-c in both time periods t1 and t2 is the same pattern as the signal collected by sensor 103a-c in the first time period and has the largest amplitude. However, the rail flaw signal does not have a periodicity that coincides with the operating cycle, and therefore the difference between the maximum amplitude occurrence time of the sensor 103c during the period t2 and the maximum amplitude occurrence time of the sensor 103a during the period t1 is not equal to the operating cycle.

A third aspect of the invention provides a computer program comprising software instructions which, when executed by a computer, implement the program modules of the monitoring method, signal preprocessing, signal de-noising, lesion detection, lesion classification, lesion localization, lesion severity assessment and wheel-track service life prediction.

Compared with the prior art, the invention not only improves the wheel-rail damage detection efficiency, but also has the following advantages:

1) the system is suitable for monitoring the health state of wheels, wheels and steel rails and infrastructure related to the steel rails;

2) the device is suitable for monitoring the health state of the wheel rail under the conditions of different wheel rail contact states, driving speeds, loads and the like;

3) the method is suitable for monitoring the whole process of the health state evolution of the wheel track.

Drawings

FIG. 1 is a diagram of a wheel track health monitoring system;

fig. 2 is a diagram showing the structure of the signal processing unit 104;

fig. 3 is a structural diagram of the wheel track health status monitoring control center 105;

FIG. 4 is a flow chart of a method for monitoring health status of wheel tracks;

FIG. 5 is a schematic diagram of a damaged wheel generating a damaged signal as it passes over sensors 103 a-c;

FIG. 6 is a schematic of wheel damage signals collected by sensors 103a-c during time period (a) t1 and during time period (b) t 2;

FIG. 7 is a schematic representation of a signal of a flaw generated by a flaw-containing rail as it is being acquired by sensors 103 a-c;

FIG. 8 is a schematic diagram of a rail flaw signal collected by the sensors 103a-c, showing a time period (a) t1 and a time period (b) t 2;

FIG. 9 is a graph of the signals collected by three sensors 103a-c during two periods of travel;

fig. 10 shows the signals acquired by the three sensors 103a-c during two periods of time in the absence of a vehicle.

Detailed Description

The following describes a specific embodiment of the present invention with reference to the following drawings, taking the monitoring of the health status of the wheel and rail based on the acoustic emission sensor as an example: three acoustic emission sensors 103a-c are mounted on the side of the rail in the manner shown in fig. 1, all connected to a signal processing unit 104 comprising a processor and a memory, which is connected to a computer 105 as a monitoring control center to form a monitoring system, and the monitoring method is stored and executed as a program module by the signal processing unit 104. Wherein the horizontal spacing of adjacent acoustic emission sensors is 1m, x in fig. 5 and 7_b＝1m、x_c2 m. Wheel radius R is 0.45m, driving speed V is 10m/s and no vehicle is availableThe measured signal is processed as follows.

Executing the step one: and judging whether the damage exists or not.

When the wheel 102 rolls stably on the rail 101 at a speed of 10m/s, the three sensors 103a-c acquire signals at two periods as shown in fig. 9, where the period t1 corresponds to 0 to 1ms and the period t2 corresponds to 282.86ms to 283.86 ms. In fig. 9, the background noise in the driving process cannot be ignored, but the amplitude is relatively small, the damage signal is clearly visible, and the existence of the damage is judged. In a time period t1, the acoustic emission sensor 103a firstly measures a damage signal and has the largest amplitude, and the acoustic emission sensor 103c finally measures a damage signal and has the smallest amplitude, which indicates that the damage occurrence position p1 is closest to the acoustic emission sensor 103a and farthest from the acoustic emission sensor 103 c; at the time of t2, the acoustic emission sensor 103c first measures the damage signal and has the largest amplitude, which indicates that the damage occurrence position p2 is closest to the acoustic emission sensor 103c and farthest from the acoustic emission sensor 103 a.

The signals acquired by the three sensors 103a-c in the absence of a vehicle at two time intervals are measured by re-timing as shown in fig. 10, where the time interval t1 corresponds to 96.22ms to 97.0392ms and the time interval t2 corresponds to 99.01ms to 99.8292 ms. In fig. 10, the amplitude of the background noise is very small, the damage signal is clearly visible, and the existence of the damage is judged. In two time periods of t1 and t2, signals collected by the sensors 103a-c have the same mode, and all the acoustic emission sensors 103a firstly measure damage signals and have the maximum amplitude, which indicates that the positions of two damage occurrences are kept unchanged relative to the three sensors, and are uniformly closest to the acoustic emission sensors 103a and farthest from the acoustic emission sensors 103 c.

And (5) executing the step two: and calculating the period.

When the wheel runs for one circle, the ratio of the advancing distance of the wheel to the vehicle speed is the running period. When the vehicle speed V is 10m/s and the wheel radius R is 0.45m, calculating the running period of the wheel

Get T_w282.7 ms. When the wheel is damaged, the damage can be generated when the wheel is in pressure contact with the steel railGenerating a damage signal, the damage signal for adjacent time periods of the sensors 103a-c will exhibit a periodicity consistent with the wheel operating cycle; when the rail has damage, the damage signals of the sensors 103a-c in adjacent time periods are independent of the operation cycle, so that the damage signals s in two adjacent time periods can pass_a-c(t) calculating the signal period

Wherein A is_a,t1、A_b,t1、A_c,t1Is the amplitude, A, of the damage signal of the

sensor

103a, 103b, 103c during the time period t1_a,t2、A_b,t2、A_c,t2The amplitude of the damage signal of the

sensor

103a, 103b, 103c during the time period t2, t2>t1, max { } is taken to be maximum,for the time corresponding to the maximum amplitude of the damage signal in the acoustic emission sensors 103a-c at time t1,is the time corresponding to the maximum amplitude of the damage signal in the acoustic emission sensor 103a-c at time t 2. Respectively calculating the damaged signal period T in the running process_swAnd the period T of the damage signal in the absence of vehicle_srTo obtain

T_sw＝283.09-0.2308＝282.8592(ms)，

T_sr＝99.07-96.277＝2.793(ms)。

And step three is executed: and (5) comparing the periods.

Period T of the damage signal_sw、T_srAnd the operating period T_wComparing to obtain

T_sw≈T_w，

T_sr≠T_w，

The measured damage signal is generated by wheel damage in the driving process, and the measured damage signal in the no-vehicle condition is a steel rail damage signal which can be further expanded and generated by steel rail damage at the same position. After the source of the damage signal is determined, the signal processing unit can enter the next step of processing, and finally the wheel track health state monitoring is achieved.

In addition, the wheel track health state monitoring system and method provided by the invention can be based on an acoustic emission sensor, a stress sensor, a displacement sensor, a vibration sensor, an acceleration sensor, an ultrasonic sensor and the like.

It should be recognized that embodiments of the present invention can also be implemented or realized in computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer readable memory. The methods may be implemented in a computer program using standard programming techniques, including a non-transitory computer readable storage medium configured with the computer program, where the storage medium so configured causes a computer to operate in a specific and predefined manner, according to the methods and figures described in the detailed description. Each program may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language. Furthermore, the program can be run on a programmed application specific integrated circuit for this purpose.

Further, the method may be implemented in any type of computing platform operatively connected to a suitable interface, including but not limited to a personal computer, mini computer, mainframe, workstation, networked or distributed computing environment, separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, and the like. Aspects of the invention may be embodied in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optically read and/or write storage medium, RAM, ROM, or the like, such that it may be read by a programmable computer, which when read by the storage medium or device, is operative to configure and operate the computer to perform the procedures described herein. Further, the machine-readable code, or portions thereof, may be transmitted over a wired or wireless network. The invention described herein includes these and other different types of non-transitory computer-readable storage media when such media include instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor. The invention also includes the computer itself when programmed according to the methods and techniques described herein.

The above description is only a preferred embodiment of the present invention, and the present invention is not limited to the above embodiment, and the present invention shall fall within the protection scope of the present invention as long as the technical effects of the present invention are achieved by the same means. The invention is capable of other modifications and variations in its technical solution and/or its implementation, within the scope of protection of the invention.

Claims

1. A wheel track health status monitoring system, the structure diagram of which is shown in figure 1, is characterized in that:

a plurality of sensors 103a-i mounted on the side of the rail 101, wherein each sensor is arranged to acquire a signal from that location;

a plurality of signal processing units 104, arranged to be able to process the signals acquired by the plurality of sensors. In particular, it is required that the number of sensors associated with one signal processing unit is not less than 2, and wherein the distance between two farthest spaced sensors corresponds to the wheel circumference;

the structure of the signal processing unit is shown in fig. 2, wherein one of the signal processing units may include a processor 1041 and a memory 1042, which are respectively used for executing and storing program modules such as signal preprocessing, signal denoising, flaw detection, flaw classification, flaw location, flaw severity evaluation, wheel-rail service life prediction, and the like;

a wheel-track health status monitoring control center 105, the structure diagram of which is shown in fig. 3, including a control unit 1051, a display unit 1052 and an early warning unit 1053, for sending instructions to the signal processing unit 104, displaying monitoring results and giving early warning to dangerous status;

the data transmission between the sensors 103a-i and the signal processing units 104a-c, between the signal processing units 104a-c, and between the signal processing units 104a-c and the monitoring control center 105 can be wired transmission or wireless transmission, by which the signals collected by the sensors can be uploaded to the signal processing units associated therewith, data exchange can be performed between adjacent signal processing units, the result data processed by the signal processing units can be uploaded to the control center, and the control center can send instructions to the signal processing units.

2. A method for monitoring the health status of a wheel and rail system, the flow chart of which is shown in fig. 4, when the signal processing unit 104a and the plurality of sensors 103a-c associated therewith in fig. 1 are taken as an example, the method includes the following steps:

the method comprises the following steps: judging whether the damage exists

step two: calculating period

When a wheel is damaged and a signal is generated when the wheel is in pressure contact with the rail, the signal will appear adjacent to the sensors 103a-cThe operating period is consistent with the periodicity; when the rail has damage, the damage signals of the sensors 103a-c in adjacent time periods are independent of the operation cycle, so that the damage signals s in two adjacent time periods can pass_a-c(t) calculating the signal period

Wherein A is_a,t1、A_b,t1、A_c,t1Amplitude, A, of the flaw signal collected for the sensors 103a, 103b, 103c during time period t1_a,t2、A_b,t2、A_c,t2Amplitude of the flaw signal collected for the sensors 103a, 103b, 103c at time period t2, t2>t1, max { } is taken to be maximum,for the time corresponding to the maximum amplitude of the damage signal in sensors 103a-c during time period t1,

step three: periodic comparison

Comparing the operating periods T_wAnd signal period T_sIf the signal period is approximately equal to the operation period, the signal is a wheel damage signal, and if the difference between the signal period and the operation period is larger, the signal is a steel rail damage signal.

3. A computer program comprising software instructions which, when executed by a computer, implement the method of any one of claims 1 to 2.