TW201533439A - Moving train wheel abnormality detection system and method - Google Patents

Abstract

A moving train wheel abnormality detection system and a method are disclosed. The system includes a vibration sensor, a microphone, and a dynamic signal analyzer. The vibration sensor is disposed at a rail base for capturing a mixed vibration signal while a train passes through. The microphone captures sound signals around the vibration sensors. The dynamic signal analyzer receives the mixed vibration signal transmitted from the vibration sensor and the sound signals transmitted from the microphones, eliminates the sound signals from the mixed vibration signal for obtaining an original vibration signal, analyzes the original vibration signal to obtain multi-level vibration signals having a plurality of vibration signal centralizing intervals, displaying the vibration signal centralizing intervals of the vibration signal of each level, and displays a corresponding detection threshold curve according to the vibration signal of each level; wherein, the vibration signal centralizing intervals of the vibration signal of each level correspond to the wheels respectively. The relative magnitudes relationship between each vibration signal centralizing interval of the vibration signal of each level and the corresponding detection threshold curve can determine whether an abnormality occurs in each wheel.

Description

Train wheel abnormality detection system and method

The invention relates to a wheel abnormality detecting system and method, in particular to a running train wheel abnormality detecting system and method.

Trains (such as the MRT, etc.) cause wheel tread wear due to friction during driving, thus making the wheel diameter smaller. Due to the friction and peeling of the tread caused by the brakes and the relative movement of the wheel and the rail surface during cornering, the tread rolling circle is chord-shaped (flat). The abrasion and wear of the wheel, especially the wheel impact, the additional impact vibration caused by the train during the train will affect the safety and service life of the train and track facilities. It must be solved in the development of various trains such as MRT and high-speed train. Detection problems.

Accordingly, it is an object of the present invention to provide a method for detecting an abnormality of a train wheel during traveling.

Therefore, the traveling wheel abnormality detecting method of the present invention is suitable for detecting whether an abnormality occurs in a plurality of wheels of the train while the train is running, and includes: a step of extracting signals, using a vibration sensing provided on a track basis , when the train passes, draw a mixed vibration signal, and simultaneously use a microphone to capture the sound signal around the vibration sensor, following And transmitting the mixed vibration signal and the sound signal to a dynamic signal analyzer; and the step of subtracting the signal, using the dynamic signal analyzer to subtract the sound signal from the mixed vibration signal to obtain an original vibration signal; In the signal step, the dynamic signal analyzer is used to analyze the multi-level vibration signal from the original vibration signal, wherein each of the vibration signals includes a plurality of vibration signal concentration intervals, and the vibration signal concentration intervals of each level of the vibration signal respectively correspond to And a display step of displaying, on a display of one of the dynamic signal analyzers, the vibration signal concentration intervals of each level of the vibration signal, and displaying a corresponding detection threshold curve for each level of the vibration signal, wherein The relative magnitude relationship between each vibration signal concentration interval of each level of the vibration signal and its corresponding detection threshold curve indicates whether an abnormality has occurred in each wheel.

Another object of the present invention is to provide a traveling wheel abnormality detecting system for traveling.

Therefore, the traveling wheel abnormality detecting system of the present invention is suitable for detecting whether an abnormality occurs in a plurality of wheels of the train by sensing vibration on a track when a train is running, and includes a vibration sensor, Microphone and a dynamic signal analyzer. The vibration sensor is disposed on the track base and is configured to capture a mixed vibration signal. The microphone is used to capture the sound signal around the vibration sensor. The dynamic signal analyzer is configured to measure the mixed vibration signal transmitted by the vibration sensor and the sound signal transmitted from the microphone when the train passes, and subtract the sound signal from the mixed vibration signal to obtain an original vibration signal. And analyzing the multi-level vibration signals including the plurality of vibration signal concentration sections from the original vibration signal, and displaying each level of vibration The vibration signal concentration interval of the signal, and for each level of the vibration signal, a corresponding detection threshold curve is displayed, wherein the vibration signal concentration intervals of each level of the vibration signal respectively correspond to the wheels, and the vibration signals of the respective levels are respectively The relative magnitude relationship between the vibration signal concentration interval and its corresponding detection threshold curve indicates whether an abnormality has occurred in each wheel.

1‧‧‧Vibration sensor

2‧‧‧ microphone

3‧‧‧Dynamic Signal Analyzer

4‧‧‧Original vibration signal

40‧‧‧ Train passage interval

41‧‧‧First level vibration signal

42‧‧‧Second level vibration signal

43‧‧‧third level vibration signal

44‧‧‧fourth level vibration signal

45‧‧‧5th level vibration signal

5‧‧‧Detection threshold curve

71~74‧‧‧Steps

8‧‧‧ Track

81‧‧‧ Track base

9‧‧‧Webmaster's Office

Other features and effects of the present invention will be apparent from the following description of the drawings. FIG. 1 is a system architecture diagram illustrating a preferred embodiment of the train wheel anomaly detection system of the present invention; FIG. Is a perspective view illustrating a measurement position map of the vibration sensor in the present invention; FIG. 3 is a flow chart illustrating a preferred embodiment of the method for detecting an abnormality of a train wheel in the present invention; FIG. 4 is a schematic view illustrating the present invention The original vibration signal and the vibration signals of each level are displayed by the dynamic signal analyzer in the invention; and FIG. 5 is a schematic diagram showing the original vibration signal and the vibration signals of each level displayed by the dynamic signal analyzer in the present invention, wherein the train is displayed The fifth section of the front wheel of the car and the rear wheel of the sixth section of the car have abnormal conditions.

The train wheel abnormality detecting system and method of the present invention proposes a train (such as MRT, etc.) wheel state detecting technology based on the following theory. When the train passes the measuring station, the wheel vibrates as the wheel is in contact with the track, and the wheel acts as a force to the track. Due to measurement In the process, the track is not replaced, so it can be assumed that the dynamic stiffness of the measurement environment has not changed, so the vibration of the track measurement is directly proportional to the force applied by the wheel. The magnitude of the force exerted by the wheel on the track is related to the contour geometry of the wheel at the same speed of each train. This is also the reason why vibration measurement can indirectly analyze the state of the train wheel. In the embodiment of the present invention, each class of MRT trains passing through the measurement station is composed of six cars, and each car has four wheels (two on each side). The effect of the detection technique of the present invention is that it can analyze the vibration state generated by the wheels of each car when the train passes, and thus can provide information on whether the vehicle has abnormal detection.

Referring to Figures 1, 2, and 4, a preferred embodiment of the traveling wheel abnormality detecting system of the present invention is suitable for detecting the train by sensing the vibration on a track base 81 while a train (not shown) is traveling. Whether or not an abnormality has occurred in a plurality of wheels (not shown). As shown in FIG. 1, the traveling train wheel abnormality detecting system includes a vibration sensor 1 (for example, an acceleration gauge), a microphone 2, and a dynamic signal analyzer 3.

The vibration sensor 1 is disposed on the track base 81, electrically connected to the dynamic signal analyzer 3, and used to capture a mixed vibration signal.

The microphone 2 is electrically connected to the dynamic signal analyzer 3 and is used to capture the sound signal of the vibration sensor 1 .

The dynamic signal analyzer 3 is configured to measure the mixed vibration signal transmitted by the vibration sensor 1 and the sound signal transmitted from the microphone 2 when the train passes, and subtract the sound signal from the mixed vibration signal to obtain a sound signal. The original vibration signal 4 (shown in FIG. 4) analyzes each of the plurality of vibration signal concentration intervals from the original vibration signal 4 (12 vibrations as shown in FIG. 4) The multi-level vibration signal of the signal concentration interval (the first, second, third, fourth, and fifth-order vibration signals 41, 42, 43, 44, 45 shown in FIG. 4), and the vibrations of each level of the vibration signal are displayed. The signal is concentrated in the interval, and a corresponding detection threshold curve 5 is displayed for each level of the vibration signal. As shown in FIG. 4, since the 12 vibration signal concentration intervals of each level of the vibration signal respectively correspond to the front and rear wheels of the six cars, the relative relationship between the vibration signal concentration intervals of the vibration signals of the respective levels and their corresponding detection threshold curves 5 is shown. The size relationship can indicate whether an abnormality has occurred in each wheel. Further, as shown in FIG. 1, the dynamic signal analyzer 3 can be disposed, for example, in the station room 9 for the examiner to monitor whether the wheel is abnormal.

Referring to FIG. 3, a preferred embodiment of the method for detecting abnormality of a running train wheel of the present invention includes a capture signal step 71, a subtraction signal step 72, an analysis signal step 73, and a display step 74.

Referring to FIG. 1 , FIG. 3 and FIG. 4 , the stepping signal step 71 uses the vibration sensor 1 to capture a mixed vibration signal while the train passes, and simultaneously captures the vibration sensor 1 by using the microphone 2 . The surrounding sound signal, and then the mixed vibration signal and sound signal are transmitted to the dynamic signal analyzer 3, wherein the vibration sensor has a signal sampling rate greater than 10 kHz. Since the vibration sensor 1 senses the mixed vibration signal from time to time, in the preferred embodiment, the trigger threshold can be set to determine when the train passes. In this embodiment, the trigger threshold value is used. For example, it can be 0.2 to 0.3 g. That is, when the mixed vibration signal sensed by the vibration sensor 1 reaches the trigger threshold value, it is the measurement starting point (when the measurement time point is zero seconds).

Then, as shown in the subtraction signal step 72, the dynamic signal analyzer 3 is used to subtract the sound signal from the mixed vibration signal to obtain the original vibration signal 4.

Then, as shown in the analysis signal step 73, the dynamic signal analyzer 3 is used to analyze the multi-level vibration signal from the original vibration signal 4. As shown in FIG. 4, in the preferred embodiment, the multi-level vibration signals are first, second, third, fourth, and fifth-order vibration signals 41, 42, 43, 44, and 45, and the vibration signals of the various levels are obvious. There are 12 vibration signal concentration intervals. As shown in FIG. 4 for the train passing section 40 indicated by the first-stage vibration signal 41, the 12 vibration signal concentration sections sequentially represent the first section of the train to the front and rear sections of the sixth section of the train from left to right. The vibration generated when the sensor 1 is vibrated. For example, the 0.95 seconds indicated in FIG. 4 is the time difference between the front and rear wheels of the first car passing through the vibration sensor 1, and the 0.27 seconds indicated in FIG. 4 indicates the first car rear wheel and the second car front wheel. By the time difference of the vibration sensor 1.

Then, as shown in the display step 74, the display (not shown) of the dynamic signal analyzer 3 displays the vibration signal concentration intervals of each level of the vibration signal, and displays the corresponding detection for the vibration signals of each level. The threshold curve 5, wherein the relative magnitude relationship between each vibration signal concentration interval of each level of the vibration signal and the detection threshold curve 5 indicates whether an abnormality has occurred in each wheel.

That is, the identification of the wheel state is based on the detection threshold curve 5 of the vibration signals of each stage shown in FIG. 4 as the detection standard. The threshold value of detecting the threshold curve 5 is based on the long-term collection of data of more than 1000 trips, based on Set by 6 standard deviation methods. If the vibration signal characteristics of each level are within the range of the detection threshold curve 5, it means that the wheel is normal, and vice versa, it is regarded as abnormal. The threshold value of the detection threshold curve 5 can be, for example, 7 g, that is, if the vibration signal characteristic is 7 g. Hereinafter, the wheel is normal, and if the vibration signal characteristic is greater than 7 g, the wheel is regarded as abnormal.

In more detail, since the multi-level vibration signals in the preferred embodiment include the first, second, third, fourth and fifth levels of vibration signals 41, 42, 43, 44, 45, the dynamic signal analyzer 3 The display shows that the first, second, third, fourth and fifth level vibration signals 41, 42, 43, 44, 45 generated by a particular wheel are not greater than the detection threshold curve 5, indicating that the particular wheel has no abnormality. When the display shows that only the first-order vibration signal 41 is greater than the detection threshold curve 5, and the second, third, fourth and fifth-order vibration signals 42, 43, 44, 45 do not exceed the detection threshold curve 5, the specific The wheel is slightly damaged. When the display shows that the first and second vibration signals 41, 42 are larger than the detection threshold curve 5, and the third, fourth and fifth vibration signals 43, 44, 45 do not exceed the detection threshold curve 5, the degree of damage is slightly Add serious, and so on. The most extreme case is when the display shows that the first, second, third, fourth and fifth level of vibration signals 41, 42, 43, 44, 45 are greater than the detection threshold curve 5, indicating that the particular wheel is severely damaged.

Therefore, the present invention allows the inspector to know which of the front or rear wheels of the car is abnormal based on the abnormal lateral position of the first-order vibration signal 41. For example, as shown by the first stage vibration signal 41 of FIG. 5, an abnormal condition occurs in the fifth car front wheel and the sixth car rear wheel of the train. In addition, the first, second, third, fourth and fifth vibration signals 41, 42 are integrated. 43, 44, 45 can be seen, because the first, second and third stage vibration signals 41, 42 and 43 of the front wheel of the fifth car exceed the detection threshold curve 5, and the rear wheel of the sixth car has only the first stage vibration. The signals 41, 42, and 43 exceed the detection threshold curve 5, so the inspector can further judge that the damage degree of the front wheel of the fifth car is more serious than that of the rear wheel of the sixth car.

In summary, the traveling wheel abnormality detecting system and method of the present invention can provide the detection information of whether the wheel has abnormality by analyzing whether the vibration characteristic value of the running train exceeds the detection threshold curve, in particular, the detecting personnel do not need to be complete. Under the detection knowledge and education training, the vehicle detection work can be accurately performed to determine the abnormal condition and damage degree of each train car wheel, so the object of the present invention can be achieved.

The above is only the preferred embodiment of the present invention, and the scope of the present invention is not limited thereto, that is, the simple equivalent changes and modifications made by the patent application scope and patent specification content of the present invention, All remain within the scope of the invention patent.

1‧‧‧Vibration sensor

2‧‧‧ microphone

3‧‧‧Dynamic Signal Analyzer

8‧‧‧ Track

9‧‧‧Webmaster's Office

Claims (10)

A method for detecting an abnormality of a train wheel during driving is suitable for detecting whether an abnormality occurs in a plurality of wheels of the train while the train is running, and comprising the following steps: a step of extracting signals, using a vibration sensing provided on a track base , when the train passes, captures a mixed vibration signal, and simultaneously uses a microphone to capture the sound signal around the vibration sensor, and then transmits the mixed vibration signal and the sound signal to a dynamic signal analyzer; The step of subtracting the signal, using the dynamic signal analyzer, subtracting the sound signal from the mixed vibration signal to obtain an original vibration signal; and analyzing the signal step, using the dynamic signal analyzer to analyze the original vibration signal The level vibration signal, wherein each level of the vibration signal comprises a plurality of vibration signal concentration intervals, and the vibration signal concentration intervals of each level of the vibration signal respectively correspond to the wheels; and a display step in one of the dynamic signal analyzers The vibration signal concentration interval of each level of the vibration signal is displayed on the display, and the vibration information is transmitted for each level. , Displays a corresponding detection threshold curve, wherein the relative magnitude relationship between the vibration signal levels of concentration interval corresponding to the vibration signal detection threshold curve indicating whether each of the wheels is abnormal. The traveling train wheel abnormality detecting method according to claim 1, wherein the vibration sensor is an acceleration gauge. A method for detecting an abnormality of a running train wheel according to claim 1, wherein The signal sampling rate of the vibration sensor is greater than 10 kHz. The method for detecting an abnormality of a running train wheel according to claim 1, wherein the multi-level vibration signals include first, second, third, fourth and fifth vibration signals, wherein a display of the dynamic signal analyzer displays a specific wheel When the first, second, third, fourth and fifth vibration signals generated are not greater than the detection threshold curve, it indicates that the specific wheel has no abnormality, and when the display shows that only the first level vibration signal is greater than the detection threshold curve, it indicates that A particular wheel is slightly damaged. When the display shows that the first, second, third, fourth and fifth vibration signals are greater than the detection threshold curve, it indicates that the particular wheel is seriously damaged. The method for detecting an abnormality of a running train wheel according to claim 1, wherein the dynamic signal analyzer determines when the train passes by a trigger threshold. A traveling train wheel abnormality detecting system is suitable for detecting whether an abnormality occurs in a plurality of wheels of a train by sensing vibration on a track when a train is running, and comprising: a vibration sensor disposed at The track base is used to capture a mixed vibration signal; a microphone is used to capture the sound signal around the vibration sensor; and a dynamic signal analyzer is used to measure the vibration sensing when the train passes The mixed vibration signal transmitted from the device and the sound signal transmitted from the microphone, the sound signal is subtracted from the mixed vibration signal to obtain an original vibration signal, and the plurality of vibration signals are analyzed from the original vibration signal. The multi-level vibration signal in the concentrated section, the vibration signal concentration interval showing each level of the vibration signal, and a corresponding detection threshold curve for each level of the vibration signal, wherein the vibration signal concentration interval of each level of the vibration signal Corresponding to the wheels, the relative magnitude relationship between each vibration signal concentration interval of each level of the vibration signal and its corresponding detection threshold curve indicates whether an abnormality has occurred in each wheel. The traveling train wheel abnormality detecting system according to claim 6, wherein the vibration sensor is an acceleration gauge. The traveling train wheel abnormality detecting system according to claim 6, wherein the vibration sensor has a signal sampling rate greater than 10 kHz. The traveling train wheel abnormality detecting system according to claim 6, wherein the multi-level vibration signals include first, second, third, fourth and fifth level vibration signals, wherein a display of the dynamic signal analyzer displays a specific wheel When the first, second, third, fourth and fifth vibration signals generated are not greater than the detection threshold curve, it indicates that the specific wheel has no abnormality, and when the display shows that only the first level vibration signal is greater than the detection threshold curve, it indicates that A particular wheel is slightly damaged. When the display shows that the first, second, third, fourth and fifth vibration signals are greater than the detection threshold curve, it indicates that the particular wheel is seriously damaged. The traveling train wheel abnormality detecting system according to claim 6, wherein the dynamic signal analyzer determines a time when the train passes by a trigger threshold.