JP7369631B2 - Uneven wear amount estimation system, uneven wear amount estimation method, placement determination method and program - Google Patents

Description

The present invention relates to an uneven wear amount estimation system, an uneven wear amount estimation method, a placement determination method, and a program.

In connection with the detection of uneven wear on the wheels of a railway vehicle, the vehicle uneven wear degree determination system described in Patent Document 1 measures the vibration of a track structure as the vehicle passes, and detects a specific frequency extracted from the measured vibration. The degree of uneven wear of the wheels of the vehicle is determined based on the components of the range and reference data for converting the vibration acceleration level into the amount of uneven wear.

In addition, the vehicle uneven wear degree determination system described in Patent Document 2 calculates the speed of the vehicle based on the vibration measurement value of the track structure, and determines whether the vehicle or a portion of the vehicle is determined based on the frequency component corresponding to the calculated speed. The timing of passing through a predetermined position is determined. Then, this vehicle uneven wear degree determination system determines the uneven wear degree of the wheels based on data in a time range corresponding to the detected passing timing.

JP2014-237348A Patent No. 6245466

When a vibration sensor is installed on a rail rather than on a railway structure such as an elevated bridge, the attenuation of vibration on the rail affects the estimation of the amount of uneven wear. It is preferable that the amount of uneven wear can be estimated with higher accuracy by reducing the influence of vibration damping in the rail on the estimation of the amount of uneven wear.

The present invention provides a system and method that can reduce the influence of vibration damping on the rail on uneven wear amount estimation when a vibration sensor is installed on the rail.

According to the first aspect of the present invention, the uneven wear amount estimation system includes at least three vibration sensors that are provided at intervals in the length direction of the rail and each detect vibrations generated by wheels passing on the rail. A power sum calculation unit that calculates a power sum of vibrations detected by the vibration sensor, and a conversion unit that converts the power sum into an uneven wear amount of the wheel.

The vibration sensor may be arranged such that a difference in the power sum depending on a position where an unevenly worn portion of the wheel contacts the rail is within a predetermined difference.

According to a second aspect of the present invention, there is provided a method for estimating uneven wear amount, in which at least three vibration sensors provided at intervals in the length direction of a rail each measure vibrations generated by wheels passing on the rail. The method includes a step of detecting, a step of calculating a power sum of vibrations detected by the vibration sensor, and a step of converting the power sum into an uneven wear amount of the wheel.

According to a fourth aspect of the invention, the program causes the computer to detect vibrations caused by at least three vibration sensors spaced along the length of the rail, each of which is caused by a wheel passing on the rail. This is a program for executing the steps of calculating the power sum of the vibrations caused by the vibration, and converting the power sum into the uneven wear amount of the wheel.

According to the present invention, when a vibration sensor is installed on a rail, it is possible to reduce the influence of vibration damping on the rail on uneven wear amount estimation.

FIG. 1 is a schematic block diagram showing a functional configuration of an uneven wear amount estimation system according to an embodiment. It is a figure showing an example of installation on a rail of an acceleration sensor concerning an embodiment. It is a figure showing an example of the processing procedure by which the uneven wear amount estimating device concerning an embodiment estimates the uneven wear amount of a wheel. FIG. 3 is a diagram showing an example of the relationship between the position of a sensor and vibrations caused by uneven wear of wheels according to the embodiment. It is a figure showing the 1st example in the case of installing one acceleration sensor concerning an embodiment on a rail. It is a figure showing the 2nd example in the case of installing one acceleration sensor concerning an embodiment on a rail. It is a figure showing the 1st example in the case of installing two acceleration sensors concerning an embodiment on a rail. It is a figure showing the 2nd example in the case of installing two acceleration sensors concerning an embodiment on a rail. FIG. 7 is a diagram illustrating an example of the distance between two acceleration sensors and the distance between the position of the acceleration sensor and the unevenly worn portion contact position when two acceleration sensors are arranged on the rail in the embodiment. FIG. 3 is a diagram illustrating an example of the relationship between the distance between two sensors, the distance from the unevenly worn portion contact position to the acceleration sensor, and uneven wear amount estimation accuracy according to the embodiment. FIG. 7 is a diagram illustrating an example of the distance between the acceleration sensors and the distance between the position of the acceleration sensor and the unevenly worn portion contact position when three acceleration sensors are arranged on the rail in the embodiment. FIG. 3 is a diagram showing an example of a level difference between the position of the acceleration sensor and a reference value according to the embodiment. FIG. 3 is a diagram illustrating an example of the arrangement of acceleration sensors according to the embodiment. It is a figure showing the 1st example of arrangement of the acceleration sensor concerning an embodiment. It is a figure which shows the 2nd example of arrangement|positioning of the acceleration sensor based on embodiment. FIG. 3 is a diagram illustrating an example of the relationship between sensor arrangement and power sum when three acceleration sensors according to the embodiment are arranged.

Hereinafter, embodiments of the present invention will be described, but the following embodiments do not limit the invention according to the claims. Furthermore, not all combinations of features described in the embodiments are essential to the solution of the invention.
FIG. 1 is a schematic block diagram showing the functional configuration of an uneven wear amount estimation system according to an embodiment of the present invention. As shown in FIG. 1, the uneven wear amount estimation system 1 includes an acceleration sensor 100 and an uneven wear amount estimation device 200. The uneven wear amount estimating device 200 includes a communication section 210, a display section 220, an operation input section 230, a storage section 280, and a control section 290. The control section 290 includes an uneven wear component extraction section 291 , a correction section 292 , a power sum calculation section 293 , a conversion section 294 , and a wheel arrival detection section 295 .

The uneven wear amount estimation system 1 estimates the uneven wear amount of wheels of a railway vehicle. The railway vehicle referred to herein may be any vehicle that is equipped with wheels that can wear unevenly and runs on rails. The railway vehicle whose uneven wear amount is estimated by the uneven wear amount estimation system 1 may be operated as a single vehicle, or may be operated by connecting a plurality of vehicles.
A railway vehicle whose uneven wear amount is to be estimated by the uneven wear amount estimation system 1 is referred to as a target vehicle.

The acceleration sensors 100 are provided at intervals along the length of the rail, and each detects the acceleration generated when a wheel passes on the rail. However, the vibration sensor included in the uneven wear amount estimation system 1 is not limited to an acceleration sensor, and any sensor capable of measuring the magnitude of rail vibration may be used. For example, the uneven wear amount estimation system 1 may include a displacement sensor or a speed sensor as a vibration sensor.
The number of acceleration sensors 100 included in the uneven wear amount estimation system 1 may be three or more. By arranging three or more acceleration sensors 100 on the rail, it is possible to measure vibrations of approximately constant magnitude, regardless of the position where the unevenly worn portion of the wheel contacts the rail, as described below.

FIG. 2 is a diagram showing an example of installing the acceleration sensor 100 on a rail.
In the example of FIG. 2, three acceleration sensors 100 are provided at intervals along the length of the rail. It is preferable that the acceleration sensors 100 are arranged at intervals deviated from the circumference of the wheel so that the acceleration sensors 100 can pick up vibrations caused by uneven wear no matter where on the wheel the uneven wear occurs.

The uneven wear amount estimating device 200 estimates the uneven wear amount of the wheel using sensing data (acceleration measurement data) by the acceleration sensor 100. The uneven wear amount estimating device 200 is configured using a computer such as a personal computer (PC) or a workstation.

The communication unit 210 communicates with other devices. In particular, the communication unit 210 receives sensing data transmitted by each of the acceleration sensors 100.
The display unit 220 has a display screen such as a liquid crystal panel or an LED (Light Emitting Diode) panel, and displays various images. For example, the display unit 220 displays the uneven wear amount estimated by the uneven wear amount estimating device 200.
The operation input unit 230 includes input devices such as a keyboard and a mouse, and accepts user operations.

The storage unit 280 stores various data. The storage unit 280 is configured using a storage device included in the uneven wear amount estimating device 200. For example, the storage unit 280 stores in advance a conversion formula for converting the power sum calculated by the uneven wear amount estimating device 200 from the sensing data of the acceleration sensor 100 into the uneven wear amount of the wheel. This conversion formula is set in advance by, for example, a test. For example, a person in the railway industry or the like conducts a test using a railway vehicle, collects data, and sets a conversion formula by linearly approximating the correlation between the sum of power and the amount of uneven wear.
However, the method by which the storage unit 280 stores the correlation between the power sum and the amount of uneven wear is not limited to the method of storing it in the form of an equation. For example, the storage unit 280 may store a conversion table for converting the power sum into uneven wear amount.

The control section 290 controls each section of the uneven wear amount estimating device 200 to perform various processes. The control unit 290 is configured by a CPU (Central Processing Unit) included in the uneven wear amount estimating device 200 reading a program from the storage unit 280 and executing it.
The uneven wear component extraction unit 291 extracts uneven wear components from the sensing data of each acceleration sensor 100. The uneven wear component here is a component that is included in the sensing data of the acceleration sensor 100 and indicates uneven wear of the wheels.

For example, the uneven wear component extraction unit 291 frequency-converts the sensing data for each acceleration sensor 100. Then, the uneven wear component extraction unit 291 extracts, from among the obtained frequency components, a frequency component that is predetermined according to the speed of the target vehicle as a frequency that indicates uneven wear. The frequency component extracted by the uneven wear component extracting section 291 corresponds to an example of an uneven wear component. The method by which the uneven wear component extraction unit 291 converts the frequency of sensing data from the acceleration sensor 100 is not limited to a specific method. For example, the uneven wear component extraction unit 291 may frequency-convert the sensing data by Fourier transform such as FFT (Fast Fourier Transform).
Alternatively, the uneven wear component extraction unit 291 may apply a filter to the sensing data of the acceleration sensor 100 to extract a specific frequency component depending on the speed of the target vehicle.

The correction unit 292 corrects the uneven wear component extracted by the uneven wear component extraction unit 291 according to the speed of the target vehicle. The faster the speed of the target vehicle is, the larger the sensor value (measured value of acceleration) of the acceleration sensor 100 becomes. Therefore, the uneven wear component extracted by the uneven wear component extraction unit 291 becomes larger as the speed of the target vehicle increases even if the uneven wear amount of the wheels is the same.
In contrast, the correction unit 292 performs correction to reduce the influence of the speed of the target vehicle on the magnitude of the uneven wear component. Specifically, the correction unit 292 multiplies the correction coefficient so that the faster the speed of the target vehicle is, the smaller the value of the uneven wear component is.
The uneven wear component corrected by the correction unit 292 is referred to as a corrected uneven wear component.

The power sum calculation unit 293 calculates the power sum of the accelerations detected by the acceleration sensor 100. Specifically, the power sum calculation unit 293 sums up the corrected uneven wear components for each acceleration sensor 100 for all acceleration sensors 100. Here, the square of the ratio between the frequency-converted acceleration amplitude of each frequency and the reference acceleration amplitude is referred to as the power ratio, and the power sum is the sum of the power ratios before decibel conversion.
The conversion unit 294 converts the power sum calculated by the power sum calculation unit 293 into an uneven wear amount of the wheel. For example, the conversion unit 294 calculates the uneven wear amount by inputting the power sum calculated by the power sum calculation unit 293 into a conversion formula stored in the storage unit 280. The uneven wear amount acquired by the conversion unit 294 is used as an estimated value of the uneven wear amount by the uneven wear amount estimation system 1.

The wheel arrival detection unit 295 detects the timing at which the wheel reaches the reference position for starting uneven wear amount estimation, and outputs a signal indicating the timing. The reference position for starting uneven wear amount estimation is predetermined as a position where the acceleration sensor 100 can acquire appropriate data for estimating the uneven wear amount of the wheel that has reached the reference position. For example, three acceleration sensors 100 may be arranged along the rail, and the position directly above the middle acceleration sensor 100 among the three may be determined as the reference position to start estimating the amount of uneven wear. Not limited. The reference position for starting uneven wear amount estimation is also simply referred to as the reference position.

The wheel arrival detection unit 295 may detect that the wheel has reached the reference position using the output signal (sensing data) of the acceleration sensor 100. Alternatively, the wheel arrival detection unit 295 may detect that the wheel has reached the reference position using information other than the output signal of the acceleration sensor 100. For example, the uneven wear amount estimation system 1 may include a photoelectric sensor. When the wheel reaches the reference position and blocks the light from the photoelectric sensor, the photoelectric sensor outputs a signal indicating that the light has been blocked, and the wheel arrival detection unit 295 uses the signal from the photoelectric sensor to determine whether the wheel is It may also be possible to detect that the reference position has been reached.

Here, the uneven wear amount estimation system 1 may estimate the uneven wear amount of the wheels in real time. Alternatively, the uneven wear amount estimation system 1 may perform the uneven wear amount estimation process with a time delay with respect to data acquisition, such as estimating the uneven wear amount of the wheels by batch processing.
When the uneven wear amount estimation system 1 estimates the uneven wear amount of the wheels in real time, the wheel arrival detection unit 295 detects in real time that the wheels have reached the reference position, and outputs a signal indicating the detection in real time. do.

On the other hand, when the uneven wear amount estimation system 1 estimates the uneven wear amount of wheels by batch processing, the wheel arrival detection unit 295 outputs, for example, the time when the wheels reach the reference position. The wheel arrival detection section 295 may detect in real time that the wheel has reached the reference position, and the storage section 280 may store the time at that time. Alternatively, the wheel arrival detection unit 295 may detect the time when the wheel reaches the reference position after the fact by data analysis.

When the wheel arrival detection unit 295 detects that the wheel has reached the reference position, the uneven wear amount estimation system 1 can estimate the uneven wear amount for each wheel. For example, the wheel arrival detection unit 295 may count the number of times that the wheel reaches the reference position. Thereby, the wheel arrival detection unit 295 can grasp which wheel from the front of the train or vehicle has reached the reference position, and therefore which wheel from the front is the target of uneven wear amount estimation. .

On the other hand, the uneven wear amount estimation system 1 is relatively less susceptible to the influence of vibration damping on the rail on uneven wear amount estimation. Thereby, there is a permissible range for the position of the wheel when the uneven wear amount estimation system 1 estimates the uneven wear amount. Therefore, an error is allowed in the detection by the wheel arrival detection unit 295 that the wheel has reached the reference position. For example, there may be a time lag between when the wheel arrival detection unit 295 detects the arrival of the wheel at the reference position and when the uneven wear estimation system 1 starts estimating the uneven wear amount. Even if the position deviates from the reference position, the influence on the estimation of the amount of uneven wear is relatively small.

FIG. 3 is a diagram showing an example of a processing procedure in which the uneven wear amount estimating device 200 estimates the uneven wear amount of a wheel. As described above, the uneven wear amount estimating device 200 may estimate the uneven wear amount of the wheels in real time, or may perform batch processing with a time delay relative to data acquisition. Therefore, the uneven wear amount estimating device 200 may perform the process shown in FIG. 3 in real time, or may perform the process with a time delay relative to data acquisition, such as by batch processing.

(Step S10)
The wheel arrival detection unit 295 detects the timing at which the wheel reaches the reference position, and outputs a signal indicating the timing.
When the uneven wear amount estimating device 200 estimates the uneven wear amount of the wheels in real time, the wheel arrival detection unit 295 detects in real time that the wheels have reached the reference position, and outputs a signal indicating the detection in real time. do. This signal is used as a trigger for the uneven wear estimation device 200 to start the process of estimating the uneven wear amount of the wheels (in the example of FIG. 3, the processing of steps S11 to S15).
On the other hand, when the uneven wear amount estimating device 200 estimates the uneven wear amount of a wheel by batch processing, the wheel arrival detection unit 295 outputs, for example, the time when the wheel reaches the reference position. This time is used as information indicating the starting position of data used by the uneven wear amount estimating device 200 to estimate the uneven wear amount of the wheel. For example, the uneven wear amount estimating device 200 stores data for a certain period of time from the time output by the wheel arrival detection unit 295 out of the historical data of the acceleration measurement values of the acceleration sensor 100 stored in the storage unit 280. Used as data for estimating the amount of uneven wear on wheels.

(Step S11)
The uneven wear component extraction unit 291 performs frequency conversion on the sensing data (time series data) of the acceleration sensor 100. The uneven wear component extraction unit 291 performs the process of step S11 for each acceleration sensor 100.
After step S11, the process advances to step S12.

(Step S12)
The uneven wear component extraction unit 291 extracts uneven wear components from the data obtained in step S11 (sensing data after frequency conversion). The uneven wear component extraction unit 291 performs the process of step S12 for each acceleration sensor 100.
Here, the faster the speed of the target vehicle is, the faster the rotation speed of the wheels becomes. When there is uneven wear on the wheels, the faster the speed of the subject vehicle is, the shorter the time interval in which the unevenly worn portion hits the rail, and the higher the frequency of the uneven wear component becomes. Therefore, the uneven wear component extraction unit 291 acquires speed information of the target vehicle, and extracts a frequency component predetermined according to the speed of the target vehicle as a frequency indicating uneven wear.
The relationship between the speed of the target vehicle and the frequency extracted by the uneven wear component extraction unit 291 is calculated in advance by, for example, a person. This frequency can be found by taking the reciprocal of the time required for the wheel to travel approximately 100 millimeters (mm) to 1000 millimeters.
After step S12, the process proceeds to step S13.

(Step S13)
The correction unit 292 corrects the uneven wear component extracted by the uneven wear component extraction unit 291 according to the speed of the target vehicle. The correction unit 292 performs the process of step S13 for each acceleration sensor 100.
As described above, the correction unit 292 multiplies the correction coefficient so that the faster the speed of the target vehicle is, the smaller the value of the uneven wear component is. Thereby, the correction unit 292 reduces the influence of the speed of the target vehicle on the magnitude of the uneven wear component.
After step S13, the process advances to step S14.

(Step S14)
The power sum calculation unit 293 calculates the power sum by summing the corrected uneven wear components for each acceleration sensor 100 for all acceleration sensors 100.
Note that when the acceleration sensors 100 are provided on each of the left and right rails, the power sum calculation unit 293 calculates the power sum for each rail by summing the corrected uneven wear components of all the acceleration sensors 100 for each rail. .
After step S14, the process proceeds to step S15.

(Step S15)
The conversion unit 294 inputs the power sum calculated by the power sum calculation unit 293 into a conversion formula, and calculates an estimated value of the amount of uneven wear.
Note that when the acceleration sensor 100 is provided on each of the left and right rails, the conversion unit 294 calculates the estimated value of the amount of uneven wear for each rail using the power sum calculated for each rail by the power sum calculation unit 293. do. Thereby, the acceleration sensor 100 estimates the amount of uneven wear for each of the left and right wheels of the target vehicle.
After step S15, the uneven wear amount estimating device 200 ends the process of FIG. 3.

Next, the number and arrangement of acceleration sensors 100 will be explained.
FIG. 4 is a diagram showing an example of the relationship between the position of the sensor and vibrations caused by uneven wear of the wheels.
In FIG. 4, the upper part of the rail 910 and the wheels 920 of the target vehicle are shown. In FIG. 4, the same wheel 920 rotates two revolutions, and the unevenly worn portion (the unevenly worn portion) contacts the rail 910 three times. Arrows B11 and B12 indicate the traveling direction of the wheels 920 (therefore, the traveling direction of the target vehicle).

The horizontal axis in FIG. 4 indicates the position of the rail 910 in the length direction. In FIG. 4, positions X11 to X15 are shown. The unevenly worn portion of the wheel 920 contacts the rail 910 at positions X11, X13, and X15, respectively. Position X12 is an intermediate position between position X11 and position X13, and will be used in the explanation below. Position X14 is an intermediate position between position X13 and position X15, and will be used in the explanation below.
The position where the worn part of the wheel 920 contacts the rail 910 is referred to as the unevenly worn part contact position.

A line L11 shows an example of vibration of the rail 910 when the unevenly worn portion of the wheel 920 contacts the rail 910 at a position X11. A line L12 shows an example of vibration of the rail 910 when the unevenly worn portion of the wheel 920 contacts the rail 910 at a position X13. A line L13 shows an example of vibration of the rail 910 due to uneven wear when the unevenly worn portion of the wheel 920 contacts the rail 910 at position X15.
The vertical axis in FIG. 4 indicates acceleration, and lines L11, L12, and L13 are shown with the center of vibration (position of zero acceleration) shifted so that lines L11, L12, and L13 can be distinguished.

In the following, an example will be described in which the distance attenuation rate of the rail 910 is 10 decibels per meter (dB/m) and the circumference of the wheel 920 is 2.7 meters (m). If the distance attenuation rate is 10 dB/meter and the circumference of the wheel 920 is 2.7 meters, the magnitude of vibration (as measured by the acceleration sensor 100) is 1.35 meters, which corresponds to half a rotation of the wheel 920. (acceleration amplitude) becomes approximately one-fifth. At 2.7 meters, which corresponds to one rotation of the wheel 920, the magnitude of vibration is approximately 1/20th.
However, the distance attenuation rate of the rail and the circumference of the wheel targeted by the uneven wear amount estimation system 1 are not limited to specific values.

FIG. 5 is a diagram showing a first example in which one acceleration sensor 100 is installed on the rail 910 in the example shown in FIG. In the example of FIG. 5, the acceleration sensor 100 is installed at position X13. In this case, acceleration sensor 100 measures a relatively large acceleration as shown by line L12 when the unevenly worn portion of wheel 920 contacts rail 910 at position X13. In this respect, it is possible to extract the uneven wear component from the sensing data of the acceleration sensor 100 and estimate the amount of uneven wear with relatively high accuracy.

FIG. 6 is a diagram showing a second example in which one acceleration sensor 100 is installed on the rail 910 in the example of FIG. In the example of FIG. 6, the acceleration sensor 100 is installed at position X12. In this case, the acceleration sensor 100 detects that even when the unevenly worn portion of the wheel 920 contacts the rail 910 at any of the positions X11, Unable to measure acceleration. In this respect, the accuracy of estimating the uneven wear amount from the sensing data of the acceleration sensor 100 becomes low.
In this way, when one acceleration sensor 100 is installed on the rail 910, the accuracy of estimating the amount of uneven wear decreases depending on the relationship between the position of the acceleration sensor 100 and the contact position of the unevenly worn part. Therefore, a plurality of acceleration sensors 100 are installed on the rail 910 to increase the accuracy of estimating the amount of uneven wear.

FIG. 7 is a diagram showing a first example in which two acceleration sensors 100 are installed on the rail 910 in the example shown in FIG. In the example of FIG. 7, acceleration sensors 100 are installed at positions X13 and X15, respectively.
In the following, the two acceleration sensors 100 will be distinguished using reference numerals 100a and 100b. Furthermore, when the three acceleration sensors 100 are to be distinguished, the symbols 100a, 100b, and 100c are used.
In the example of FIG. 7, the acceleration sensor 100 at position X13 is referred to as acceleration sensor 100a, and the acceleration sensor 100 at position X15 is referred to as acceleration sensor 100b.

In the example of FIG. 7, when the unevenly worn portion of the wheel 920 contacts the rail 910 at position X13, the acceleration sensor 100a measures a relatively large acceleration. Further, when the unevenly worn portion of the wheel 920 contacts the rail 910 at the position X15, the acceleration sensor 100b measures a relatively large acceleration.
When frequency-converting time-series sensing data as in the process described with reference to FIG. 3, a relatively large uneven wear component is extracted from both the sensing data of the acceleration sensor 100a and the sensing data of the acceleration sensor 100b. When these corrected uneven wear components are summed, the power sum becomes larger than the actual value, and the estimated value of the uneven wear amount becomes larger than the actual uneven wear amount.

FIG. 8 is a diagram showing a second example in which two acceleration sensors 100 are installed on the rail 910 in the example shown in FIG. In the example of FIG. 8, acceleration sensors 100 are installed at positions X12 and X14, respectively. In the example of FIG. 8, the acceleration sensor 100 at position X12 is referred to as acceleration sensor 100a, and the acceleration sensor 100 at position X14 is referred to as acceleration sensor 100b.
In the example of FIG. 8, in both acceleration sensors 100a and 100b, when the unevenly worn portion of the wheel 920 contacts the rail 910 at any of the positions X11, X13, or X15, the vibration is sufficiently damped and Unable to measure large accelerations. Even if the corrected uneven wear components of these two acceleration sensors 100 are summed, the power sum becomes smaller than the actual value, and the estimated value of the uneven wear amount becomes smaller than the actual uneven wear amount.

In this way, there are cases in which the amount of uneven wear is overestimated and cases in which it is underestimated. In order to estimate the amount of uneven wear with high accuracy, it is necessary to determine the number and arrangement of the acceleration sensors 100 while paying attention to both cases of overestimating and underestimating the amount of uneven wear.
Below, first, regarding the case where two acceleration sensors 100 are installed on the rail 910, the distance (interval) between the two acceleration sensors 100, the distance between the position of the acceleration sensor 100 and the contact position of the unevenly worn part, and the unevenly worn part contact position. The relationship with the estimation accuracy of quantity will be explained. Next, regarding the case where three acceleration sensors 100 are installed on the rail 910, the distance between the two acceleration sensors 100, the distance between the position of the acceleration sensor 100 and the contact position of the unevenly worn part, and the estimation accuracy of the amount of uneven wear. Explain the relationship between

FIG. 9 is a diagram showing an example of the distance between the two acceleration sensors 100 and the distance between the position of the acceleration sensor 100 and the unevenly worn portion contact position when the two acceleration sensors 100 are arranged on the rail. The horizontal axis in FIG. 9 indicates the position of the rail 910 in the length direction, and the traveling direction of the target vehicle is + (positive). Of the two acceleration sensors 100 shown in FIG. 9, the acceleration sensor 100 on the near side (the one with the smaller position value) in the traveling direction of the target vehicle is defined as the acceleration sensor 100a, and the acceleration sensor on the back side (the one with the larger position value) 100 is an acceleration sensor 100b.

As shown in FIG. 9, the distance between two acceleration sensors 100 is expressed as d. Furthermore, the difference between the position of the acceleration sensor 100a and the unevenly worn portion contact position is expressed as xa. |xa| (absolute value of xa) indicates the distance between the acceleration sensor 100a and the unevenly worn portion contact position. When the value of the position of the acceleration sensor 100a is larger than the contact position of the unevenly worn part (that is, when it is on the back side in the traveling direction of the target vehicle), the sign of xa is +. When the value of the position of the acceleration sensor 100a is smaller than the contact position of the unevenly worn part (that is, when it is on the near side in the traveling direction of the target vehicle), the sign of xa is -.
The distance with a sign, which is the difference between the position of the acceleration sensor 100a and the contact position of the unevenly worn part as a reference, is also referred to as the distance from the unevenly worn part contact position to the acceleration sensor 100a.

FIG. 10 is a diagram showing an example of the relationship between the distance between two sensors, the distance from the unevenly worn part contact position to the acceleration sensor 100a, and the uneven wear amount estimation accuracy. The horizontal axis in FIG. 10 indicates the distance xa (unit: meters) from the unevenly worn portion contact position to the acceleration sensor 100a. The vertical axis indicates the distance d (unit: meters) between the two acceleration sensors 100.
In FIG. 10, the level difference from the reference value of the power sum calculated from the sensing data of the two acceleration sensors 100 is expressed as a contour line in decibels for various distances xa and d, with the amount of uneven wear of the wheels 920 being constant. It is shown in In the example of FIG. 10, the power sum is calculated from the sensing data of the two acceleration sensors 100 using the method described with reference to FIG. Furthermore, when the distance xa from the unevenly worn part contact position to the acceleration sensor 100a is 0 (that is, when the unevenly worn part contact position is directly above the acceleration sensor 100a), the method described with reference to FIG. The corrected uneven wear component calculated from the sensing data of the acceleration sensor 100a is used as a reference value, and this reference value is set to 0 decibel (dB).

In order to estimate the uneven wear amount with high accuracy, it is preferable that the power sum is constant regardless of the distance xa from the uneven wear portion contact position to the acceleration sensor 100a. If the goal is for the difference between the power sum and the reference value to be within ±2 decibels, the range surrounded by line L21 will be the range that satisfies the goal.
In FIG. 10, no matter where a line parallel to the horizontal axis is placed, as exemplified by line L22, the entire line cannot be included within the target range. Therefore, no matter how you set the distance d between the two acceleration sensors 100, depending on the distance xa from the unevenly worn part contact position to the acceleration sensor 100a, the difference between the power sum and the reference value will be within the target ±2 decibel range. It doesn't fit in.

FIG. 11 is a diagram showing an example of the distance between the acceleration sensors 100 and the distance between the position of the acceleration sensor 100 and the unevenly worn portion contact position when three acceleration sensors 100 are arranged on the rail. The horizontal axis in FIG. 11 indicates the position of the rail 910 in the length direction, and the traveling direction of the target vehicle is + (positive). Two of the three acceleration sensors 100 shown in FIG. 11 are the same as those in FIG. 9. Specifically, of the two acceleration sensors 100, the acceleration sensor 100 on the near side in the traveling direction of the target vehicle (the one with the smaller position value) is set as the acceleration sensor 100a, and the acceleration sensor 100 on the back side (the one with the larger position value) The sensor 100 is assumed to be an acceleration sensor 100b.

As in the case of FIG. 9, the distance between acceleration sensors 100a and 100b is expressed as d. Further, the difference between the position of the acceleration sensor 100a and the contact position of the unevenly worn part as a reference (that is, the distance from the unevenly worn part contact position to the acceleration sensor 100a) is expressed as xa.
Further, the third acceleration sensor 100 will be referred to as an acceleration sensor 100c. Based on the position of the acceleration sensor 100a, the difference between the position of the acceleration sensor 100c and the position of the acceleration sensor 100c is expressed as _d3 .

|d ₃ | (absolute value of d ₃ ) indicates the distance between the acceleration sensor 100a and the acceleration sensor 100c. When the position of the acceleration sensor 100c has a larger value than the position of the acceleration sensor 100a (that is, when the position is further back in the traveling direction of the target vehicle), the sign of _d3 is +. When the position of the acceleration sensor 100c has a smaller value than the position of the acceleration sensor 100a (that is, when it is on the near side in the traveling direction of the target vehicle), the sign of _d3 is -.
This signed distance, which is the difference between the position of the acceleration sensor 100c and the position of the acceleration sensor 100a, is also referred to as the distance from the acceleration sensor 100a to the acceleration sensor 100c.

FIG. 12 is a diagram showing an example of the level difference between the position of the acceleration sensor 100c and the reference value. The horizontal axis in FIG. 12 indicates the distance xa from the unevenly worn portion contact position to the acceleration sensor 100a. The vertical axis indicates the level difference in decibels of the corrected uneven wear component based on the sensing data of the acceleration sensor 100c from the reference value.
Line L31 shows an example where the distance _d3 from acceleration sensor 100a to acceleration sensor 100c is 1 meter. A line L32 shows an example where the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is 0. Line L33 shows an example where the distance _d3 from acceleration sensor 100a to acceleration sensor 100c is -1 meter.

FIG. 13 is a diagram showing an example of the arrangement of the acceleration sensor 100c. The horizontal axis in FIG. 13 indicates the distance xa (unit: meters) from the unevenly worn portion contact position to the acceleration sensor 100a. The vertical axis indicates the distance d (unit: meters) between the acceleration sensors 100a and 100b. In FIG. 13, as in the case of FIG. 10, the level of the power sum calculated from the sensing data of the acceleration sensors 100a and 100b is calculated from the reference value for various distances xa and d, with the amount of uneven wear of the wheels 920 being constant. The difference is shown by contour lines expressed in decibels.

Based on FIGS. 12 and 13, the acceleration sensor 100 is arranged as follows.
(1) The distance d between the acceleration sensors 100a and 100b from the area A11 is set in the range of 0.25 to 2.4 meters.
(2) In area A11, the distance _d3 from acceleration sensor 100a to acceleration sensor 100c is set so that the sum of powers based on sensing data of three acceleration sensors 100 does not become too large. Specifically, the acceleration sensor 100c is set to -8 decibels or less. From FIG. 12, the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is set to be 0.8 meters or more or -0.8 meters or less.

(3) By adding the acceleration sensor 100c, 0 decibel is added to the power sum shown in FIG. 13 in areas A13 and A14. At this time, in area A12, the sum of powers based on the sensing data of the three acceleration sensors 100 is made to be within ±2 decibels. Therefore, the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is set to about 0.8 to 1.07 meters or -1.05 to -0.8 meters.

A: When the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is 0.8 to 1.07 meters,
- In region A15, the sum of powers based on the sensing data of the three acceleration sensors 100 becomes too large;
- At the upper left end of area A13, the sum of powers based on the sensing data of the three acceleration sensors 100 becomes too small;
- In area A14, the sum of powers based on sensing data of the three acceleration sensors 100 is -2 decibels or more;
Considering these three factors, the distance d between the acceleration sensors 100a and 100b is set to 1.7 to 1.8 meters. In addition, in FIG. 13, the point where d=1.7 meters is indicated by a line.

B: When the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is -1.05 to -0.8 meters,
- In area A16, the sum of powers based on the sensing data of the three acceleration sensors 100 becomes too large;
- At the lower right end of area A14, the sum of powers based on the sensing data of the three acceleration sensors 100 becomes small;
- In area A13, the sum of powers based on sensing data of the three acceleration sensors 100 is -2 decibels or more;
Considering these three factors, the distance d between the acceleration sensors 100a and 100b is set to about 1 meter. In addition, in FIG. 13, the location where d=1 meter is indicated by a line.
In case B, strictly speaking, the condition that the sum of powers based on the sensing data of the three acceleration sensors 100 is within ±2 decibels is slightly not satisfied in some parts, but it is almost satisfied.

(4) When the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c exceeds 1.07 meters, an area appears in the area A12 where the sum of the powers from the sensing data of the three acceleration sensors 100 is less than -2 decibels. . The area is small until the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is about 1.15 meters, and if this is avoided, the sum of powers based on the sensing data of the three acceleration sensors 100 will be within ±2 decibels. At this time, the distance d between the acceleration sensors 100a and 100b is 2 to 2.4 meters. The lower limit value of the distance d at this time depends on the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c.

FIG. 14 is a diagram showing a first example of the arrangement of the acceleration sensor 100.
In the arrangement shown in FIG. 14, the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is within the range of 0.8 to 1.15 meters.
When the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is 1.07 meters or less, the distance d between the acceleration sensors 100a and 100b is set within the range of 1.7 to 1.8 meters.
When the distance _d3 from acceleration sensor 100a to acceleration sensor 100c is greater than 1.07 meters, the distance d between acceleration sensors 100a and 100b is set within the range of 2 to 2.4 meters. In this case, the lower limit value of the distance d between the acceleration sensors 100a and 100b changes depending on the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c.

FIG. 15 is a diagram showing a second example of the arrangement of the acceleration sensor 100.
In the arrangement shown in FIG. 15, the distance _d3 from the acceleration sensor 100a to the acceleration sensor 100c is set within the range of -1.07 to -0.8 meters. Further, the distance d between the acceleration sensors 100a and 100b is set to 1 meter.
As described above, strictly speaking, the arrangement of FIG. 15 slightly does not satisfy the condition that the sum of powers based on the sensing data of the three acceleration sensors 100 is within ±2 decibels. From this point of view, the arrangement shown in FIG. 14 is preferable.

If the error is further allowed (that is, if the allowable range of the power sum based on the sensing data of the three acceleration sensors 100 is widened), the range of the installation intervals of the acceleration sensors 100 will be expanded. Conversely, if the error is made smaller (that is, if the allowable range of the power sum based on the sensing data of the three acceleration sensors 100 is narrowed), there will be no arrangement that satisfies the condition.

FIG. 16 is a diagram showing an example of the relationship between sensor arrangement and power sum when three acceleration sensors 100 are arranged. The horizontal axis in FIG. 16 indicates the distance xa (unit: meters) from the unevenly worn portion contact position to the acceleration sensor 100a. The vertical axis indicates the distance d (unit: meters) between the acceleration sensors 100a and 100b.
In FIG. 16, the level difference from the reference value of the power sum calculated from the sensing data of the two acceleration sensors 100 is expressed as a contour line in decibels for various distances xa and d, with the amount of uneven wear of the wheels 920 being constant. It is shown in Further, the distance from sensor 100a to sensor 100c is constant at 1.09 meters.

FIG. 16 shows an example in which one more sensor 100 is added to the example in which there are two sensors 100 shown in FIG. 10, resulting in three sensors 100. The example in FIG. 16 is similar to the example in FIG. 10 except that there are three sensors.
In the example of FIG. 16, when d=1.875 meters, there is a portion where the value slightly exceeds ±2 decibels, but it is approximately within the range of ±2 decibels. If the spacing between the sensors 100 is further adjusted, it can be kept within the range of ±2 decibels.
In addition, in FIG. 16, the point where d=1.875 meters is indicated by a line.

As described above, at least three acceleration sensors 100 are provided at intervals in the length direction of the rail, and each detects the acceleration generated by the wheel 920 passing on the rail 910. The power sum calculation unit 293 calculates the power sum of the accelerations detected by the acceleration sensor 100. The conversion unit 294 converts the power sum calculated by the power sum calculation unit 293 into an uneven wear amount of the wheel.
According to the uneven wear amount estimation system 1, at least three acceleration sensors 100 are provided at intervals in the length direction of the rail 910, so that regardless of the position where the unevenly worn portion of the wheel 920 contacts the rail 910, approximately Can measure vibrations of a certain magnitude. Thereby, the uneven wear amount estimation system 1 can reduce the influence of vibration damping in the rail 910 on uneven wear amount estimation, and in this respect, the uneven wear amount can be estimated with high accuracy.

Further, the acceleration sensor 100 is arranged so that the difference in the sum of power depending on the position where the unevenly worn portion of the wheel 920 contacts the rail 910 is within a predetermined difference. As a result, in the uneven wear amount estimation system 1, a power sum of approximately constant magnitude can be obtained regardless of the position where the unevenly worn portion of the wheel 920 contacts the rail 910. According to the uneven wear amount estimation system 1, it is possible to reduce the influence of vibration damping in the rail on uneven wear amount estimation, and in this respect, uneven wear amount can be estimated with high accuracy.

In addition, the difference between the reference value of the power sum based on the sensing data of the two acceleration sensors 100 is calculated based on the distance between the two acceleration sensors 100 and the position where the reference acceleration sensor 100 and unevenly worn parts of the wheels 920 touch the rail 910. The distance to is shown by contour lines on the graph. The relationship between the graph shown by contour lines, the distance between the reference acceleration sensor 100 and the position where the unevenly worn part of the wheel 920 contacts the rail 910, and the magnitude of the acceleration measured by the third acceleration sensor 100 Based on this, the arrangement of the acceleration sensor 100 is determined.
According to this arrangement determination method, a power sum of approximately constant magnitude can be obtained regardless of the position where the unevenly worn portion of the wheel 920 contacts the rail 910. If this placement determination method is used, it is possible to reduce the influence of vibration damping on the rail on estimating the amount of uneven wear, and in this respect, the amount of uneven wear can be estimated with high accuracy.

Note that a computer or hardware-readable recording medium is used to store a program for realizing all or part of the processing performed by the uneven wear amount estimating device 200 and the processing for determining the arrangement of the acceleration sensor 100. The program recorded on the recording medium may be read into a computer system or hardware and executed to perform the processing of each part. Note that the "computer system" herein includes hardware such as an OS and peripheral devices.
Furthermore, the term "computer system" includes the homepage providing environment (or display environment) if a WWW system is used.
Furthermore, the term "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into computer systems. Further, the above-mentioned program may be one for realizing a part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in the computer system.

Although the embodiments of the present invention have been described above in detail with reference to the drawings, the specific configuration is not limited to these embodiments, and may include design changes within the scope of the gist of the present invention.

1 Uneven wear amount estimation system 100 Acceleration sensor 200 Uneven wear amount estimation device 210 Communication section 220 Display section 230 Operation input section 280 Storage section 290 Control section 291 Uneven wear component extraction section 292 Correction section 293 Power sum calculation section 294 Conversion section 295 Wheel Arrival detection part

Claims (4)

at least three vibration sensors spaced apart along the length of the rail and each detecting vibrations caused by wheels passing on the rail;
a power sum calculation unit that calculates a power sum of vibrations detected by the vibration sensor;
a conversion unit that converts the power sum into an uneven wear amount of the wheels;
Uneven wear amount estimation system equipped with The uneven wear amount estimation system according to claim 1, wherein the vibration sensor is arranged such that a difference in the power sum depending on a position where an unevenly worn portion of the wheel contacts the rail is within a predetermined difference. at least three vibration sensors spaced apart along the length of the rail, each detecting vibrations caused by wheels passing on the rail;
a step of calculating a power sum of vibrations detected by the vibration sensor;
a step of converting the power sum into an amount of uneven wear of the wheels;
A method for estimating uneven wear amount including to the computer,
Calculating the power sum of vibrations that are detected by at least three vibration sensors spaced apart in the length direction of the rail, each of which is generated by a wheel passing on the rail;
a step of converting the power sum into an amount of uneven wear of the wheels;
A program to run.

Images