JP2013156161A - Method and system for detecting abnormality of axle bearing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for removing acoustic disturbance as much as possible in a railway vehicle (1), and while considering variation in rolling motion sound caused by dispersion in wheel tread states, detecting abnormality of an axel bearing, and to provide a bearing monitoring system for achieving the method.SOLUTION: A bearing monitoring method for detecting abnormality of an axel bearing (7) of a railway vehicle (1) includes: a step for installing a microphone (31) on a side of rails (11); a step for installing noise insulation equipment (51) between the rails and the microphone; a vehicle traveling sound detection step for detecting a vehicle traveling sound by the microphone (31); a rolling motion sound corresponding-value detection step for detecting a rolling motion sound corresponding-value by a rolling motion sound corresponding-value detector (71) attached to the rail; and an abnormality detection step for detecting abnormality of the axel bearing (7) on the basis of the vehicle traveling sound and the rolling motion sound corresponding-value.

Description

  The present invention relates to a method and system for detecting an abnormality in an axle bearing of a railway vehicle.

  Axle bearings, which are a part of the components of railway vehicles, are installed so as not to impede rotation of the axles while supporting the weight of the vehicle. If the surfaces of the inner ring and outer ring in contact with the roller in the axle bearing contact repeatedly while being loaded, mechanical fatigue may occur, and an abnormality such as peeling may occur.

  Such an abnormality in the axle bearing may cause problems such as heat generation, burn-in, and derailment of the vehicle, and it is important to detect such damage at an early stage.

  As a method for detecting an abnormality in an axle bearing, there is a conventional method for detecting an abnormality in an axle bearing by installing an acceleration pickup in each axle box that accommodates the axle bearing and detecting the acceleration. This method has not actually been used as a detection method due to the complexity and cost of having to install accelerometers in all the axle boxes.

  Currently, a method for detecting an abnormality of an axle bearing from a sound generated when the axle bearing is damaged has already been proposed. As one of them, software RailBAM (trademark) for detecting an abnormality of an axle bearing by a microphone array formed by a plurality of microphones installed beside a rail is generally sold by Vipac. In Non-Patent Document 1, an abnormality of an axle bearing is detected using this software.

Keith Bradon and four others, “PREDICTION CONDITION MONITORING OF RAILWAY ROLLIDNG STOCK”, (Australia), RTSA, June 20-23, 2004 (Conference Engineering Conference (Confer) Railway Engineering))

  However, RailBAM targets a freight car, and there is a problem that an abnormality of an axle bearing can be detected only by sound detected by an accompanying car. For example, if there is an operating sound of a vehicle under-floor device such as a drive unit or a ventilation / air conditioner, a rolling sound based on an excitation force caused by minute unevenness on the surface of the rail or wheel, or irregularity in the wheel tread This is because the sound generated by elements other than the axle bearing, such as impact noise, becomes an acoustic disturbance and becomes an obstacle when detecting an abnormality of the axle bearing.

  Accordingly, an object of the present invention is to eliminate the acoustic disturbance as much as possible in a railway vehicle and to consider the fluctuation of the rolling noise caused by the variation in the wheel tread state, and from the vehicle running sound detected by the microphone, An object of the present invention is to provide a method for detecting an abnormality and a bearing monitoring system.

According to the present invention,
A bearing monitoring method for detecting an abnormality in an axle bearing of a railway vehicle,
Installing a microphone on the side of the rail;
Installing a sound insulation device between the rail and the microphone, the sound insulation device installing a sound insulation device having an opening between the axle bearing and the microphone; and
A vehicle running sound detection stage for detecting a vehicle running sound with a microphone;
Rolling sound equivalent value detection that detects the rolling sound equivalent value representing the sound generated by the excitation force acting on the contact surface between the rail and the wheel by the rolling sound equivalent value detector installed on the rail Stages,
An abnormality detection stage for detecting an abnormality of the axle bearing based on the vehicle running sound and the rolling sound equivalent value;
including,
A bearing monitoring method is provided.

The bearing monitoring method further includes a step of passing the vehicle running sound and the rolling sound equivalent value through a band-pass filter of a frequency band determined according to the speed of the vehicle at the time of detection,
Preferably, the abnormality detection step is a step of detecting an abnormality of the axle bearing based on the vehicle running sound after passing through the band pass filter and the rolling sound equivalent value.

  This is because the noise generated by the axle bearing can be easily detected by cutting the frequency components other than the frequency component mainly included in the sound generated by the axle bearing.

The bearing monitoring method further includes a step of detecting when the axle bearing passes a predetermined detection position by an axle passage detection device installed on the rail,
The abnormality detection step is preferably a step of detecting an abnormality of the axle bearing based on the vehicle running sound and the rolling sound equivalent value at the time when the vehicle passes a predetermined detection position.

  When the axle bearing passes through the microphone and the rolling noise equivalent value detector, peaks are generated in the vehicle running sound and rolling noise equivalent value, and the position of those peaks is recognized by the axle passage detection device. This is because it becomes easier, and furthermore, it becomes easier to obtain the values of these peaks.

The bearing monitoring method includes:
The railway vehicle has a plurality of axle bearings;
When the axle bearing passes through the microphone, the vehicle running noise peaks.
Furthermore, when the axle bearing passes the rolling noise equivalent value detector, a peak occurs in the rolling noise equivalent value.
The abnormality detection step
Detecting a detection target vehicle travel sound peak value that is a peak value of the vehicle travel sound corresponding to an axle bearing of a target for detecting an abnormality;
Detecting an evaluation vehicle running sound peak value that is a peak value of a vehicle running sound corresponding to another vehicle bearing;
When the value obtained by subtracting the evaluation vehicle travel sound peak value from the detection target vehicle travel sound peak value is higher than a predetermined vehicle travel sound threshold, the detection target vehicle travel sound peak value is higher than the evaluation vehicle travel sound peak value. Tentatively determining that the cause is an axle bearing,
The detection target rolling sound equivalent value peak value, which is the peak value of the rolling sound equivalent value generated when the axle bearing corresponding to the detection target vehicle running sound peak passes over the rolling noise equivalent value detector, Detecting, and
An evaluation rolling sound equivalent value peak value that is a peak value of the rolling sound equivalent value generated when the axle bearing corresponding to the evaluation vehicle traveling sound peak passes over the rolling sound equivalent value detector is detected. Stages,
If it is detected that the value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is lower than a predetermined rolling sound equivalent value threshold value, it is determined that the provisional judgment is correct, If it is determined that there is an abnormality in the axle bearing and it is detected that the value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is equal to or greater than the predetermined rolling noise equivalent value threshold value Determining that the provisional determination is not correct and determining that the axle bearing is normal; and
It is preferable that it is a stage including.

  By including such an abnormality detection stage, it is possible to reduce erroneous detection due to fluctuations in rolling noise caused by variations in the wheel tread state, and thus improve the accuracy of the bearing monitoring method.

The bearing monitoring method includes:
The railway vehicle has a plurality of axle bearings;
When the axle bearing passes through the microphone, the vehicle running noise peaks.
Furthermore, when the axle bearing passes the rolling noise equivalent value detector, a peak occurs in the rolling noise equivalent value.
The abnormality detection step
Detecting a detection target vehicle travel sound peak value that is a peak value of the vehicle travel sound corresponding to an axle bearing of a target for detecting an abnormality;
Detecting an evaluation vehicle running sound peak value that is a peak value of a vehicle running sound corresponding to another vehicle bearing;
A step of taking the average value of the detected evaluation vehicle running sound peak value;
When the value obtained by subtracting the average value of the evaluation vehicle travel sound peak value from the detection target vehicle travel sound peak value is higher than a predetermined vehicle travel sound threshold, the detection target vehicle travel sound peak value is evaluated vehicle travel sound peak value. Tentatively determining that the axle bearing has a cause higher than the average value of
The detection target rolling sound equivalent value peak value, which is the peak value of the rolling sound equivalent value generated when the axle bearing corresponding to the detection target vehicle running sound peak value passes over the rolling noise equivalent value detector, Detecting, and
An evaluation rolling sound equivalent value peak value that is a peak value of the rolling sound equivalent value generated when the axle bearing corresponding to the evaluation vehicle traveling sound peak value passes over the rolling noise equivalent value detector is detected. Stages,
The stage of taking the average value of the detected evaluation rolling sound equivalent value peak value,
If it is detected that the value obtained by subtracting the average value of the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is lower than a predetermined rolling sound equivalent value threshold value, the provisional judgment is correct. It is determined that there is an abnormality in the axle bearing, and a value obtained by subtracting the average value of the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is equal to or greater than a predetermined rolling noise equivalent value threshold value. If it is detected, the provisional judgment is judged to be incorrect, and the axle bearing is judged to be normal.
It is preferable that it is a stage including.

  In the abnormality detection stage, by using the average value of the evaluation vehicle traveling sound peak value as a reference as described above, the erroneous detection due to the detection error of the vehicle traveling sound is further reduced, and the evaluation rolling sound equivalent value peak is described as described above. This is because, by using the average value as a reference, false detection due to fluctuations in rolling noise caused by variations in wheel tread condition is further reduced, and the accuracy of the bearing monitoring method is further improved.

The abnormality detection step
Removing the rolling sound equivalent value from the vehicle running sound, calculating the vehicle running sound from which the influence of the rolling noise is removed;
Detecting the abnormality of the axle bearing based on the vehicle running sound from which the influence of the rolling noise is removed;
It is preferable that the step includes

  This is because fluctuations in rolling noise due to variations in wheel tread surface conditions can be directly removed from the vehicle running sound from the vehicle running sound, which further improves the accuracy of the bearing monitoring method.

The sound insulation device is a substantially rectangular soundproof wall;
It is preferable that the height positions of the opening and the microphone are substantially equal to the height position of the axle box that houses the vehicle bearing.

  This is because the direct sound of the sound generated from the axle bearing to be detected reaches the microphone.

The height of the opening is approximately equal to the height of the axle box;
The length of the opening in the rail longitudinal direction is preferably shorter than the axial distance.

  This is because the direct sound from the axle box to be detected, and hence the axle bearing, is not blocked by the sound insulating portion of the sound insulation device, and the direct sound from the adjacent front and rear axle bearings is prevented from reaching the microphone. Here, the term “axial distance” refers to the distance between axles attached to the same carriage.

  It is preferable that the microphone and the rolling sound equivalent value detector are substantially aligned in a horizontal direction perpendicular to the longitudinal direction of the rail.

  This is because a vehicle running sound peak and a rolling sound equivalent value peak generated due to a certain axle bearing are detected at almost the same point, and subsequent data processing becomes easy.

  Further, if not only the microphone and the rolling sound equivalent value detector but also the axle passage detection device are installed, it is preferable that they are similarly aligned. This is because the subsequent data processing is further facilitated as described above.

According to the present invention,
A bearing monitoring system for detecting an abnormality in an axle bearing of a railway vehicle,
A microphone installed on the side of the rail to detect vehicle running sound;
A sound insulation device installed between the rail and the microphone, the sound insulation device having an opening between the axle bearing and the microphone and a sound insulation portion between the rail and the microphone;
A rolling sound equivalent value detector installed on the rail for detecting a rolling sound equivalent value representing the sound generated from the rail;
An abnormality detection device that detects an abnormality of an axle bearing based on the vehicle running sound and the rolling sound equivalent value;
Comprising
A bearing monitoring system is provided.

  According to the above-described invention, in a railway vehicle, abnormalities in axle bearings are detected from vehicle running sound detected by a microphone while eliminating acoustic disturbance as much as possible and taking into account fluctuations in rolling noise caused by variations in wheel tread surface conditions. And a bearing monitoring system can be provided.

The top view which shows arrangement | positioning of each component for implementing the method for detecting the abnormality of an axle bearing based on embodiment by this invention. Sectional front view by the II-II line of FIG. 1 which shows arrangement | positioning of each component for implementing the method for detecting the abnormality of an axle bearing based on embodiment by this invention. The enlarged view of the A section of FIG. 2 for showing the positional relationship of the rail and the component attached to a rail based on embodiment by this invention. The front view from the rail vehicle side of the sound insulation apparatus and microphone which concern on embodiment by this invention. The figure which shows the structure of the analyzer by this invention. The time series graph of the assumption of a vehicle running sound and rolling sound equivalent value when an axle bearing is not damaged. The time series graph of the assumption of rolling sound equivalent value when not judging the vehicle running sound when it is temporarily judged that an axle bearing is damaged, and an axle bearing is abnormal. According to the third embodiment of the abnormality detection stage, after assuming the vehicle running sound from which the influence of the rolling noise is removed after removing the rolling noise equivalent value from the vehicle running sound when there is an abnormality in the axle bearing Series graph.

  The invention will be described in more detail with reference to the above-mentioned drawings. It should be noted that the drawings are drawn differently from the actual size, scale, and shape of the components in order to facilitate understanding of the present invention and simplify the description of the drawings.

  1 and 2 show the arrangement of each component when carrying out a method for detecting an abnormality in an axle bearing according to an embodiment of the present invention. Here, although the railway vehicle includes various other components, in FIG. 1 and FIG. 2, the description of components that are not directly related to the present invention is omitted for the sake of brevity.

  The railway vehicle 1 includes a wheel 3 for transmitting power to the rail, an axle 5 that couples the wheel 3, an axle bearing 7 that supports the axle 5, and an axle box 9 that accommodates the axle bearing 7. The railway vehicle 1 travels on two rails 11. In FIG. 1, the above-described components for one vehicle are shown, and two trolleys (not shown) are attached to the vehicle, and each trolley has two axles 5. One wheel 3, an axle bearing 7, and a shaft box 9 are provided at both ends.

  Further, FIG. 1 and FIG. 2 show a microphone 31 installed on the side of the rail 11 for detecting a vehicle running sound generated by the railway vehicle 1. The microphone 31 does not need to be a special microphone such as a directional microphone, and does not require special equipment.

  Further, in FIGS. 1 and 2, the sound insulation device 51 for insulating the direct sound of the sound generated by the components of the railway vehicle other than the axle bearings out of the vehicle running sound generated by the railway vehicle 1 is shown in FIG. It is shown that it is installed between the microphone 31.

  FIG. 3 is an enlarged view of a portion A in FIG. 2 and shows the rail 11 and components attached to the rail.

  FIG. 3 shows that the rolling sound equivalent value detector 71 is installed at the bottom of the rail 11. The rolling sound equivalent value detector 71 is based on the excitation force acting on the contact surface between the rail and the wheel among the sounds radiated into the atmosphere due to the vibration of the rail, sleeper, wheel and the like. The object is to detect a rolling sound equivalent value that is a physical quantity correlated with the generated rolling sound.

  In this embodiment, as shown in FIG. 3, a vehicle passage detection device 13 is installed in order to detect a point in time when the peak of the vehicle running sound and the rolling sound equivalent value occurs. Usually, the axle passage detection device 13 is installed on the side of the rail on the microphone side. In the embodiment, the vehicle passage detection device is a coil, and detects the passage of the axle by detecting a change in the magnetic field when the rotating wheel passes by. However, in the present invention, the vehicle passage detection device is not limited to this mechanism, and any device may be used as long as it can accurately detect the passage of the axle. The vehicle passage detection device 13 may be omitted.

  In this embodiment, it is assumed that the microphone 31 and the rolling sound equivalent value detector 71 are substantially aligned in the horizontal direction perpendicular to the longitudinal direction of the rail 11. Therefore, the peaks of the vehicle running sound and the rolling sound equivalent values in FIGS. 6 and 7 occur at almost the same point. However, the present invention is not limited to the positional relationship between the microphone 31 and the rolling sound equivalent value detector 71. This does not preclude a different arrangement due to circumstances such as the installation of the microphone 31 and the rolling sound equivalent value detector 71.

  In the embodiment, the rolling sound equivalent value detector 71 is an acceleration pickup. However, the rolling sound equivalent value detector 71 may be anything as long as it detects a physical quantity correlated with the rolling sound.

Next, the rolling sound equivalent value will be described. Here, it is known that the acoustic power W radiated from the vibrating object is given by the following equation.
W (t) = ρ ₀ c ₀ Sσv ² (t)
Where ρ ₀ : air density
c ₀ : speed of sound in the air
S: Radiation area of the object
σ: Radiation efficiency
v (t): Vibration speed Here, the vibration speed is the vibration speed in the vertical direction. This is because the exciting force between the rail and the wheel is applied in that direction. Therefore, the sound level Lv radiated from the vibrating object when the arbitrary sound power W ₀ and the arbitrary vibration velocity value v ₀ are set as a reference is
Lv (t) = W (t) / W ₀
= 10 log (ρ ₀ c ₀ Sσv ² (t) / ρ ₀ c ₀ Sσv ₀ ² )
= 10 log (v ² (t) / v ₀ ² )
= 10 logv ² (t) + C
Here, C: constant (= −10 log ₁₀ v ₀ ² )
It becomes. In this embodiment, the above Lv (t) is approximated as a rolling sound equivalent value. However, the rolling sound equivalent value may be any value as long as it correlates with the rolling sound.

  FIG. 4 is a diagram showing the shape of the sound insulation device 51 from the front.

  The sound insulation device 51 includes an opening 53 that is opened in a slit shape, and a sound insulation portion 55 that is another part and is intended to isolate sound from the railcar 1 side. In the embodiment, the sound insulation device 51 is a flat plate, and is a soundproof wall installed on the ground. Further, in the embodiment, the opening 53 is substantially rectangular and is substantially at the same height as the axle box 9, and the height thereof is substantially equal to the height of the axle box 9. According to this aspect, the direct sound generated at the axle bearing 7 and reaching the microphone 31 is passed through the opening 53, while the direct sound to the microphone 31 generated by other components is blocked. It is intended. Further, it is preferable that the length of the sound insulating device 51 in the rail longitudinal direction is longer than the length of one rail vehicle 1 and the length of the opening 53 in that direction is shorter than the axial distance. This is to prevent direct sound from the adjacent axle bearing 7 from reaching the microphone 31. Accordingly, the shape, material, and the like of the sound insulation device 51 may be any as long as the above object can be achieved.

  FIG. 5 is a diagram showing the configuration of the analysis apparatus according to the present invention. This analysis device includes a CPU (computer) 91 that performs the necessary analysis using the above detection result as an input, and an output device 93 that outputs the analysis result. The CPU 91 indicates that the vehicle running sound and the rolling sound equivalent value detected by the microphone 31 and the rolling sound equivalent value detector 71 and optionally the vehicle detected by the vehicle passage detection device 13 have passed. A signal is taken in and data such as a graph necessary for an abnormality detection stage described later is created. Next, the CPU 91 outputs the data created by the CPU to an output device 93 such as a display or a printer so that the user can see the data.

  Further, since the sound generated by the axle bearing tends to appear in a certain frequency band, the vehicle running sound and the rolling sound equivalent value may be passed through a band-pass filter having a certain frequency band. This makes it easier to detect damage to the axle bearing. The frequency band is not greatly affected by the abnormal state of the axle bearing 7, but is determined mainly according to the speed of the vehicle at the time of detection. The bandpass filter may be an existing frequency filter such as an octave band filter and a 1/3 octave band filter, or may be a bandpass filter for a specific frequency band defined by a user.

  With the above configuration, when the railway vehicle 1 travels on the rail 11, the vehicle traveling sound and the rolling sound equivalent value are detected by the microphone 31 and the rolling sound equivalent value detector 71. 6 and 7 show time series graphs of assumed vehicle running sound and rolling sound equivalent values for one vehicle. As can be seen from these figures, when the tread of the wheel, and thus the axle bearing 7, passes in front of the microphone 31, a peak occurs in the vehicle running sound, and when it passes over the rolling sound equivalent value detector, it corresponds to the rolling sound. A peak occurs in the value. As described above, FIG. 6 and FIG. 7 show the vehicle running sound and rolling sound equivalent value for one vehicle, and such a phenomenon occurs even when another vehicle runs.

  FIG. 6 shows a model case when there is no abnormality in the axle bearing. The behavior of the vehicle running sound and the rolling noise equivalent value when the rolling noise that can be generated by the axle bearing is not detected from the vehicle running sound is shown. In this case, at each peak, the rolling noise occupies a very large proportion of the vehicle running sound, and the behavior of the vehicle running sound substantially coincides with the behavior corresponding to the rolling noise.

  FIG. 7 shows the behavior of the vehicle running sound and the rolling sound equivalent value when one of the four vehicle running sound peaks in this embodiment is larger than the other peaks. In the rolling sound equivalent value graph of FIG. 7, the solid line and the dotted line indicate the model case when the axle bearing has an abnormality and when there is no abnormality, respectively.

  From this, the abnormality detection stage which detects the abnormality of an axle bearing using the time series data of the vehicle running sound and the rolling sound equivalent value detected by the configuration of the above embodiment will be described in several embodiments. To do.

  In the first embodiment at the abnormality detection stage, an abnormality of the axle bearing is detected by the following analysis method.

  First, the detection target vehicle travel sound peak 33 that is the peak of the vehicle travel sound corresponding to the axle bearing that is the target of detecting an abnormality from the vehicle travel sound is selected, and further from the peak of the vehicle travel sound corresponding to the other vehicle bearing The evaluation vehicle running sound peak 35 is selected.

  The evaluation target vehicle running sound peak 35 is the same railway vehicle as the railway vehicle 1 to which the method of detecting an abnormality of the axle bearing is applied, so that the rolling noise does not vary due to wheel variations. You may select from the vehicle running sound previously detected in the same environment using the same new railway vehicle which is new or equivalent.

  The following comparison is performed using the values of these peaks. When a value obtained by subtracting the evaluation vehicle travel sound peak value from the detection target vehicle travel sound peak value is higher than a predetermined vehicle travel sound threshold, it is temporarily determined that the cause is the axle bearing. At this stage, there may be a case where the rolling noise is increased due to the variation of the wheels, so at this stage, it is not determined that the axle bearing is abnormal. In addition, when the value obtained by subtracting the evaluation vehicle travel sound peak value from the detection target vehicle travel sound peak value is equal to or less than the predetermined vehicle travel sound threshold, it is determined that the cause is not in the axle bearing.

  The vehicle running sound threshold is when there are abnormalities in various detection conditions (window function, time constant, overlapping, etc.), vehicle speed, and axle bearings related to detecting vehicle running sound and rolling sound equivalent values. This is a value determined in consideration of a detection rate indicating how much probability can be detected, and is an arbitrary value determined by the user.

  Next, from the time-series graph of the rolling sound equivalent value, the rolling sound equivalent value generated when the axle bearing corresponding to the detection target vehicle running sound peak 33 passes over the rolling sound equivalent value detector. The detection target rolling sound equivalent value peak 73, which is a peak, is extracted, and the rolling noise representative generated when the axle bearing corresponding to the evaluation vehicle traveling sound peak value passes over the rolling sound equivalent value detector is extracted. An evaluation rolling sound equivalent value peak 75 that is a peak is extracted.

  As in the case of the vehicle running sound, the evaluation rolling sound equivalent value peak 75 is the same railway vehicle as the railway vehicle 1 to which the method of detecting an abnormality of the axle bearing is applied, and is caused by variations in wheels. You may extract from the rolling sound equivalent value previously detected in the same environment using the same railway vehicle which is new or equivalent so that the rolling sound equivalent value does not fluctuate.

  Finally, the following comparison is performed using the values of these peaks. On the other hand, if it is detected that a value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is lower than a predetermined rolling sound equivalent value threshold value, the provisional judgment is correct. Judge that there is an abnormality in the axle bearing. This is because it can be determined that the rolling noise increases due to the variation in the wheels, and that this does not increase the vehicle running sound peak 35 to be evaluated. On the other hand, if it is detected that the value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is greater than or equal to a predetermined rolling sound equivalent value threshold, the provisional judgment is not correct. It is determined that there is no abnormality in the axle bearing. This is because it is possible to determine that the rolling noise is increased due to the variation in the wheels, and that this is a result of increasing the vehicle running sound peak 35 to be evaluated.

  Note that the rolling sound equivalent value threshold is a value determined in consideration of various factors in the same manner as the vehicle running sound threshold, and is an arbitrary value determined by the user.

  As described above, by comparing the difference between the detection target rolling sound equivalent value peak value and the evaluation rolling sound equivalent value peak value, it is possible to detect an abnormality in the axle bearing while considering the variation between the wheels.

  Here, FIG. 7 is used to show a process of actually executing the above-described abnormality detection stage of the first embodiment. First, the user looks at the vehicle running sound graph and confirms a peak that is larger than the other peaks (third from the left in the vehicle running sound graph of FIG. 7). In this case, this peak is selected as the detection target vehicle running sound peak 33. The occurrence of a peak larger than the other peaks is because there is a suspicion that the axle bearing is abnormal. Next, another vehicle traveling sound peak is selected as the evaluation vehicle traveling sound peak 35. In this case, it is assumed that the peak (fourth from the left in the vehicle running sound graph of FIG. 7) of the same cart is selected. Next, as described above, if the detection target vehicle traveling sound peak value is higher than a predetermined vehicle traveling sound threshold with respect to the evaluation vehicle traveling sound peak value, there is an abnormality in the axle bearing corresponding to the large peak of the traveling vehicle sound. Temporarily judge that there is. There is a certain difference between the detection target vehicle travel sound peak value and the evaluation vehicle travel sound peak value, and it is assumed here that the difference exceeds the vehicle travel sound threshold. Therefore, in this case, it is assumed that a provisional judgment is made that there is an abnormality in the axle bearing. Next, as described above, the rolling noise equivalent value peak generated when the axle bearing corresponding to the detection target vehicle running sound peak 33 and the evaluation vehicle running sound peak 35 passes over the rolling noise equivalent value detector. A detection target rolling sound equivalent value peak 73 and an evaluation rolling sound equivalent value peak 75 (third and fourth peaks from the left in the rolling sound equivalent value graph of FIG. 7) are extracted. Then, on the one hand, if it is detected that the value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value as described above is lower than the predetermined rolling sound equivalent value threshold value, It is determined that the provisional determination is correct, and it is determined that there is an abnormality in the axle bearing. This indicates the case of the solid line of the rolling sound equivalent value graph of FIG. 7, and the detection target rolling sound equivalent value peak value and the evaluation rolling sound equivalent value peak value have substantially the same value on the graph. Therefore, in this case, there is almost no difference between both peak values, and here it is assumed that the rolling sound equivalent value threshold is not exceeded. Therefore, in this case, it is determined that the provisional determination is correct, and it is determined that there is an abnormality in the axle bearing. On the other hand, if it is detected that the value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is equal to or greater than the predetermined rolling sound equivalent value threshold as described above, It is judged that the judgment was not correct, and it is judged that there is no abnormality in the axle bearing. This indicates the case of the dotted line in the rolling sound equivalent value graph of FIG. In this case, there is a certain difference between the detection target rolling sound equivalent value peak value and the evaluation rolling sound equivalent value peak value, and the difference here exceeds the rolling sound equivalent value threshold value. Shall. Therefore, in this case, it is determined that the provisional determination is not correct, and it is determined that there is no abnormality in the axle bearing. In this way, the abnormality of the axle bearing is detected in the abnormality detection stage of the first embodiment.

  In the second embodiment at the abnormality detection stage, an abnormality of the axle bearing is detected by the following analysis method.

  First, the detection target vehicle travel sound peak 33 that is the peak of the vehicle travel sound corresponding to the axle bearing that is the target of detecting an abnormality from the vehicle travel sound is selected, and further from the peak of the vehicle travel sound corresponding to the other vehicle bearing A plurality of evaluation vehicle traveling sound peaks 35 are selected.

  As in the first embodiment in the abnormality detection stage, the evaluation vehicle traveling sound peak 35 is selected from vehicle traveling sounds that have been detected in advance using a new or equivalent identical railway vehicle. Also good. Furthermore, in order to average the value of the evaluation vehicle running sound peak 35 to be described later, it is preferable to select many peaks at this time. This is because the accuracy of detection increases. The evaluation vehicle travel sound peak 35 may be selected from the same vehicle as the vehicle on which the axle bearing on which the detection target vehicle travel sound peak 33 is based, and in addition to that of another vehicle. You may choose.

  The following comparison is performed using the values of these peaks. When the value obtained by subtracting the average value of the plurality of evaluation vehicle travel sound peak values from the detection target vehicle travel sound peak value is higher than a predetermined vehicle travel sound threshold, it is temporarily determined that the cause is the axle bearing. Similar to the first embodiment at the abnormality detection stage, at this stage, the rolling noise may be increased due to variations in the wheels, so at this stage, it is not determined that the axle bearing is abnormal. In addition, when the value obtained by subtracting the average value of the plurality of evaluation vehicle traveling sound peak values from the vehicle traveling sound peak value to be detected is equal to or lower than the predetermined vehicle traveling sound threshold value, it is determined that the cause is not in the axle bearing. become.

  Next, from the time-series graph of the rolling sound equivalent value, the rolling sound equivalent value generated when the axle bearing corresponding to the detection target vehicle running sound peak 33 passes over the rolling sound equivalent value detector. The detection target rolling sound equivalent value peak 73 that is a peak is extracted, and the rolling bearing generated when the axle bearing corresponding to each of the plurality of evaluation vehicle traveling sound peaks 35 passes over the rolling noise equivalent value detector. An evaluation rolling sound equivalent value peak 75 that is a peak of the moving sound representative is extracted.

  The plurality of evaluation rolling sound equivalent value peaks 75 are detected in advance using a new or equivalent same railway vehicle in the same manner as in the first embodiment in the abnormality detection stage. You may extract from the rolling sound equivalent value.

  Finally, the following comparison is performed using the values of these peaks. On the other hand, if it is detected that a value obtained by subtracting the average value of the plurality of evaluation rolling sound equivalent value peak values from the detection target rolling sound equivalent value peak value is lower than a predetermined rolling sound equivalent value threshold value, The provisional judgment is judged to be correct, and it is judged that there is an abnormality in the axle bearing. This is because it can be determined that the rolling noise is not increased due to the variation in the wheels. On the other hand, if it is detected that the value obtained by subtracting the average value of the plurality of evaluation rolling sound equivalent value peak values from the detection target rolling sound equivalent value peak value is equal to or greater than a predetermined rolling sound equivalent value threshold value, It is determined that the provisional judgment is not correct, and it is determined that there is no abnormality in the axle bearing. This is because it can be determined that the rolling noise has increased due to variations in the wheels.

  As described above, by using the average value of the plurality of evaluation rolling sound equivalent values peak values as a reference for detecting the abnormality of the axle bearing, it is possible to reduce the variation among the wheels as compared with the first embodiment in the abnormality detection stage. It is possible to detect an abnormality in the axle bearing after taking into consideration.

  Here, as in the case of the abnormality detection stage of the first embodiment, FIG. 7 is used to show a process of actually executing the abnormality detection stage of the second embodiment described above. First, the user looks at the vehicle running sound graph and confirms a peak that is larger than the other peaks (third from the left in the vehicle running sound graph of FIG. 7). In this case, this peak is selected as the detection target vehicle running sound peak 33 as in the case of the abnormality detection stage of the first embodiment. Next, another vehicle traveling sound peak is selected as the evaluation vehicle traveling sound peak 35. In this case, it is assumed that the peaks of the same vehicle (1st, 2nd and 4th from the left in the vehicle running sound graph of FIG. 7) are selected. Next, as described above, if the detection target vehicle travel sound peak value is higher than a predetermined vehicle travel sound threshold with respect to the average value of the evaluation vehicle travel sound peak value, the axle bearing corresponding to the large peak of the vehicle travel sound Is temporarily determined to be abnormal. There is a certain difference between the detection target vehicle traveling sound peak value and the average value of the evaluation vehicle traveling sound peak value, and here, it is assumed that the difference exceeds the vehicle traveling sound threshold. Therefore, in this case, it is assumed that a provisional judgment is made that there is an abnormality in the axle bearing. Next, as described above, the rolling sound equivalent value generated when the axle bearing corresponding to the detection target vehicle traveling sound peak 33 and each evaluation vehicle traveling sound peak 35 passes over the rolling sound equivalent value detector. The detection target rolling sound equivalent value peak 73 (the third peak from the left in the rolling sound equivalent value graph of FIG. 7) and the respective evaluation rolling sound equivalent value peak 75 (corresponding to the rolling sound of FIG. 7). The first, second and fourth peaks from the left in the value graph are extracted. Next, on the other hand, as described above, it is detected that the value obtained by subtracting the average value of the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is lower than the predetermined rolling sound equivalent value threshold value. If so, it is determined that the provisional determination is correct and it is determined that there is an abnormality in the axle bearing. This indicates the case of the solid line of the rolling sound equivalent value graph of FIG. 7, and the detection target rolling sound equivalent value peak value and the average value of the evaluation rolling sound equivalent value peak value are (each evaluation rolling In this case, there is almost no difference between the two peak values, and it is assumed that the rolling sound equivalent value threshold is not exceeded. Therefore, in this case, it is determined that the provisional determination is correct, and it is determined that there is an abnormality in the axle bearing. On the other hand, if it is detected that the value obtained by subtracting the average value of the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is equal to or greater than a predetermined rolling sound equivalent value threshold as described above. It is determined that the provisional determination is not correct, and it is determined that there is no abnormality in the axle bearing. This indicates the case of the dotted line in the rolling sound equivalent value graph of FIG. In this case, there is a certain difference between the detection target rolling sound equivalent value peak value and the average value of the evaluation rolling sound equivalent value peak value. It shall be beyond. Therefore, in this case, it is determined that the provisional determination is not correct, and it is determined that there is no abnormality in the axle bearing. In this manner, the abnormality of the axle bearing is detected in the abnormality detection stage of the second embodiment.

  In the third embodiment of the abnormality detection stage, an abnormality of the axle bearing is detected by converting the rolling sound equivalent value detected by the following analysis method into a direct sound and correcting the vehicle running sound by the sound. To do.

Using the rolling sound equivalent value Lv, the vehicle running sound from which the influence of the rolling noise is removed is calculated by removing the rolling noise equivalent value from the vehicle running sound according to the following equation.
Lb (t) = 10 log (10 ^{Ld (t) / 10} −10 ^{Lv (t) / 10} )
Where Ld: vehicle running sound
Lb: Vehicle running sound with the influence of rolling noise removed

  Specifically, in the case of FIG. 7, when there is an abnormality in the axle bearing (when the rolling sound equivalent value graph of FIG. The coming Lb is assumed to draw a graph as shown in FIG. As described above, the rolling sound equivalent value including the fluctuation of the rolling noise due to the variation of the wheels is directly removed from the vehicle running sound, so that the peak sound generated by the abnormal axle bearing finally remains. Become. Therefore, it is possible to detect that there is an abnormality in the axle bearing, and as a result, the object of the present invention can be achieved.

  From this, the bearing monitoring system for implementing the said method is demonstrated.

  Since the equipment and detection process of the bearing monitoring system are the same as those described above, a description thereof is omitted here.

  In the above-described abnormality detection stage, the user has detected an abnormality by looking at the graph, but in this bearing monitoring system, the vehicle running sound threshold value and the rolling sound equivalent value threshold value are input in advance to the abnormality detection device, In this example, the abnormality detection device executes the abnormality detection stage of any one of the first to third embodiments of the abnormality detection stage using the detected vehicle running sound and rolling sound equivalent value. Since it is set, the abnormality of the axle bearing 7 can be automatically detected by the abnormality detection device. Note that the abnormality detection device is the CPU 91 in this example.

  The bearing monitoring method can be automatically executed by the bearing monitoring system described above.

DESCRIPTION OF SYMBOLS 1 Rail vehicle 3 Wheel 5 Axle 7 Axle bearing 9 Axle box 11 Rail 13 Vehicle passage detection device 31 Microphone 33 Vehicle sound detection peak 35 Evaluation vehicle travel sound peak 51 Sound insulation device 53 Opening portion 55 Sound insulation portion 71 Rolling sound equivalent value Detector 73 Detection target rolling sound equivalent value peak 75 Evaluation rolling sound equivalent value peak 91 CPU
93 Output device

Claims (10)

A bearing monitoring method for detecting an abnormality in an axle bearing (7) of a railway vehicle (1),
Installing a microphone (31) beside the rail (11);
In the stage of installing the sound insulation device (51) between the rail (11) and the microphone (31), the sound insulation device (51) has an opening (between the axle bearing (7) and the microphone (31)). Installing a sound insulation device (51) having 53);
A vehicle running sound detection step of detecting a vehicle running sound by the microphone (31);
Rolling sound equivalent value representative of rolling sound generated based on the excitation force acting on the contact surface between the rail (11) and the wheel (3) is detected as the rolling sound equivalent value installed on the rail. A rolling sound equivalent value detection stage detected by a vessel (71);
An abnormality detection step of detecting an abnormality of the axle bearing (7) based on the vehicle running sound and the rolling sound equivalent value;
including,
Bearing monitoring method. And passing the vehicle running sound and the rolling sound equivalent value through a band-pass filter of a frequency band determined according to the speed of the vehicle at the time of detection,
The abnormality detection step is a step of detecting an abnormality of the axle bearing (7) based on the vehicle running sound and the rolling sound equivalent value after passing through a band pass filter.
The bearing monitoring method according to claim 1. And a step of detecting by the axle passage detection device (13) installed on the rail (11) when the axle bearing (7) passes a predetermined detection position,
The abnormality detection step is a step of detecting an abnormality of the axle bearing (7) based on the vehicle running sound and the rolling sound equivalent value at the time of passing through a predetermined detection position.
The bearing monitoring method according to claim 1 or 2. The railway vehicle (1) has a plurality of axle bearings;
When the axle bearing (7) passes through the microphone (31), a peak occurs in the vehicle running sound,
Furthermore, when the axle bearing (7) passes through the rolling noise equivalent value detector (71), a peak occurs in the rolling noise equivalent value.
The abnormality detection step
Detecting a detection target vehicle travel sound peak value, which is a peak value of the vehicle travel sound corresponding to the axle bearing (7) to be detected for abnormality,
Detecting an evaluation vehicle running sound peak value that is a peak value of a running vehicle sound corresponding to another vehicle bearing (7);
When the value obtained by subtracting the evaluation vehicle travel sound peak value from the detection target vehicle travel sound peak value is higher than a predetermined vehicle travel sound threshold, the detection target vehicle travel sound peak value is higher than the evaluation vehicle travel sound peak value. Tentatively determining that the cause is the axle bearing (7);
The detection target rolling sound equivalent value that is the peak value of the rolling sound equivalent value generated when the axle bearing (7) corresponding to the detection target vehicle running sound peak passes over the rolling noise equivalent value detector. Detecting a peak value;
Evaluation rolling sound equivalent value peak value, which is the peak value of the rolling sound equivalent value generated when the axle bearing (7) corresponding to the evaluation vehicle running sound peak passes over the rolling noise equivalent value detector. Detecting the stage,
If it is detected that the value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is lower than a predetermined rolling sound equivalent value threshold value, it is determined that the provisional judgment is correct, It is determined that there is an abnormality in the axle bearing (7), and it is detected that the value obtained by subtracting the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is equal to or greater than a predetermined rolling noise equivalent value threshold value. If so, determining that the provisional determination is not correct and determining that there is no abnormality in the axle bearing (7);
Including the stage,
The bearing monitoring method according to claim 1. The railway vehicle (1) has a plurality of axle bearings (7);
When the axle bearing (7) passes through the microphone (31), a peak occurs in the vehicle running sound,
Furthermore, when the axle bearing (7) passes through the rolling noise equivalent value detector (71), a peak occurs in the rolling noise equivalent value.
The abnormality detection step
Detecting a detection target vehicle travel sound peak value, which is a peak value of the vehicle travel sound corresponding to the axle bearing (7) to be detected for abnormality,
Detecting a value of an evaluation vehicle traveling sound peak that is a peak value of a vehicle traveling sound corresponding to another vehicle bearing (7);
A step of taking the average value of the detected evaluation vehicle running sound peak value;
When the value obtained by subtracting the average value of the evaluation vehicle travel sound peak value from the detection target vehicle travel sound peak value is higher than a predetermined vehicle travel sound threshold, the detection target vehicle travel sound peak value is evaluated vehicle travel sound peak value. Tentatively determining that the axle bearing (7) is higher than the average value of
The detection target rolling sound equivalent value that is the peak value of the rolling sound equivalent value generated when the axle bearing (7) corresponding to the detection target vehicle running sound peak passes over the rolling noise equivalent value detector. Detecting a peak value;
An evaluation rolling sound equivalent value that is a peak value of the rolling sound equivalent value generated when the axle bearing (7) corresponding to each of the evaluation vehicle traveling sound peaks passes over the rolling sound equivalent value detector. Detecting a peak value;
The stage of taking the average value of the detected evaluation rolling sound equivalent value peak value,
If it is detected that the value obtained by subtracting the average value of the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is lower than a predetermined rolling sound equivalent value threshold value, the provisional judgment is correct. It is determined that the axle bearing (7) is abnormal, and a value obtained by subtracting the average value of the evaluation rolling sound equivalent value peak value from the detection target rolling sound equivalent value peak value is a predetermined rolling noise equivalent value threshold value. If it is detected that the above, the provisional judgment is judged to be incorrect, and the axle bearing (7) is judged to be normal.
Including the stage,
The bearing monitoring method according to claim 1. The abnormality detection step
Removing the rolling sound equivalent value from the vehicle running sound, calculating the vehicle running sound from which the influence of the rolling noise is removed;
Detecting the abnormality of the axle bearing based on the vehicle running sound from which the influence of the rolling noise is removed;
Including the stage,
The bearing monitoring method according to claim 1. The sound insulation device (51) is a substantially rectangular sound barrier;
The height positions of the opening (53) and the microphone (31) are substantially equal to the height position of the axle box (9) that houses the vehicle bearing (7).
The bearing monitoring method according to claim 1. The height of the opening (53) is substantially equal to the height of the axle box (9);
The length of the opening (53) in the rail longitudinal direction is shorter than the axial distance.
The bearing monitoring method according to claim 1. The microphone (31) and the rolling sound equivalent value detector (71) are substantially aligned in a horizontal direction perpendicular to the longitudinal direction of the rail.
The bearing monitoring method according to claim 1. A bearing monitoring system for detecting an abnormality in an axle bearing (7) of a railway vehicle (1),
A microphone (31) installed on the side of the rail (11) for detecting vehicle running sound;
A sound insulation device (51) installed between a rail (11) and a microphone (31), having an opening (53) between the axle bearing (7) and the microphone (31) and the rail (11). ) And a microphone (31), and a sound insulation device (51) having a sound insulation portion (55);
A rolling sound equivalent value detector (71) installed on the rail (11) for detecting a rolling sound equivalent value representing the sound generated from the rail (11);
An abnormality detection device for detecting an abnormality of the axle bearing (7) based on the vehicle running sound and the rolling sound equivalent value;
Comprising
Bearing monitoring system.

Images