JP2007022220A - Track condition analyzing method, track condition analyzing device, and track condition analyzing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a track condition detecting method and a track condition detecting device capable of inexpensively and precisely detecting a condition of the track with a simple constitution in a track condition detecting method, a track condition detecting device, and a track detecting program for detecting the condition of the track, on which a vehicle is traveling. <P>SOLUTION: This track condition analyzing method analyzes a condition of the track, on which the vehicle is traveling. Acceleration of the vehicle is acquired, and a multiple resolution is analyzed using a wavelet transform to the acquired acceleration. The condition of the track is analyzed from the analyzed results. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a track state analysis method, a track state analysis device, and a track state analysis program, and more particularly to a track state analysis method, a track state analysis device, and a track state analysis program for detecting the state of a track on which a vehicle travels.

  In track transportation systems such as railways, safety and passenger comfort depended on track accuracy. As for the accuracy of the track, it is important to ensure the accuracy during track construction and to maintain it after construction. In order to ensure accuracy during track construction and to maintain it after construction, an electric / track comprehensive inspection vehicle called Doctor Yellow has been developed.

  This electric / track comprehensive inspection vehicle has performed high-precision inspection using a large-scale measurement system in order to accurately obtain data that can be used for the maintenance of electricity / track. In addition, the electric / track comprehensive inspection vehicle needs to travel along the track on which the business vehicle travels, in addition to the driver, the attendant of the surveyor, and the business vehicle. And it lacked mobility.

  For this reason, there was a limit to running all the way to the sub trunk line and the local traffic line.

  In addition, as a measurement system, a measurement device installed on the ground detects the traveling state of the vehicle when the vehicle passes, and transfers and accumulates image information obtained by capturing the vehicle at a remote location in a lump. There is a system that analyzes changes in track and vehicle time from the obtained data (see Patent Document 1). In such a system, it takes time and labor to perform a measuring instrument, and the apparatus has a large structure, and only the state of a specific place can detect the state of the orbit.

Furthermore, as a measurement system, there is a system in which an acceleration sensor is provided in the axle box of the inspection vehicle, acceleration data measured by the acceleration sensor is subjected to continuous wavelet transform, and rail wave wear is detected based on the result (see Patent Document 2). . This system requires an inspection car, requires a driver and an attendant, and an operation plan for traveling on the track, is expensive, and lacks mobility. Further, even when a business vehicle is used, it is necessary to set an acceleration sensor in the axle box, and it is necessary to perform wiring from the axle box to the inside of the vehicle, resulting in a large-scale configuration.
JP 2003-182580 A Japanese Patent Application Laid-Open No. 2000-136988

  However, any of the conventional systems for detecting an orbital state has a large and expensive measurement system, and cannot be installed all over the sub trunk line and local traffic line.

  In addition, when the acceleration data is subjected to continuous wavelet transform and the rail wave wear is detected based on the result, the processing is complicated and the measurement system becomes large.

  The present invention has been made in view of the above points, and provides a track state analysis method, a track state analysis device, and a track state analysis program capable of accurately detecting a track state with a simple configuration and at a low cost. With the goal.

  The present invention is a trajectory state analysis method for analyzing the state of a trajectory on which a vehicle travels, obtains acceleration applied to the vehicle, performs multi-resolution analysis on the obtained acceleration using wavelet transform, and the analysis result From this, it is characterized by analyzing the state of the orbit.

  According to the present invention, the acceleration applied to the vehicle is acquired, multiresolution analysis is performed on the acquired acceleration using wavelet transform, and the state of the trajectory is analyzed from the analysis result, so that there is no large change in acceleration. Even in this case, as a result of the multi-resolution analysis, the analysis can be performed using the analysis result most reflecting the analyzed orbital state, and thus the orbital state can be reliably analyzed. Further, the processing can be realized by a small computer by using the discrete wavelet transform which is easy to process. In addition, by detecting the acceleration applied to the vehicle body, it becomes easy to install an acceleration sensor and so on, so that it can be installed in commercial vehicles. It can be analyzed accurately.

〔System configuration〕
FIG. 1 shows a system configuration diagram of an embodiment of the present invention.

  The track state detection system 100 according to the present embodiment includes, for example, a vehicle 112 that travels on a track 111 such as a railroad and a track state detection device 113.

  The vehicle 112 is a business vehicle and includes a carriage 121 and a vehicle body 122. Wheels 131 that rotate on the track 111 are rotatably attached to the carriage 121 and are suspended from the vehicle body 122 via suspensions 132.

  The track state detection device 122 is mounted on the vehicle body 122, detects the acceleration in the vertical direction applied to the vehicle body 122, that is, the arrow Z direction, and analyzes the detected acceleration to analyze the track state.

  FIG. 2 is a block diagram of the trajectory state detection device 113.

  The orbital state detection device 112 includes acceleration sensors 141 and 142, an interface 143, and a computer system 144.

  The acceleration sensor 141 detects acceleration applied in the vertical direction and the arrow Z direction. The acceleration detected by the acceleration sensor 141 is supplied to the interface 143. The acceleration sensor 142 detects acceleration applied in the front-rear direction and the arrow X direction. The acceleration detected by the acceleration sensor 142 is supplied to the interface 143.

  The interface 143 interfaces with the acceleration sensors 141 and 142 and the computer system 143, and inputs the acceleration detected by the acceleration sensors 141 and 142 to the computer system 144.

  The computer system 144 is composed of, for example, a personal computer system, and is composed of a processing device 151, a storage device 152, an input device 153, and a display device 154.

  The processing device 151 performs data processing based on the trajectory state analysis program installed in the storage device 152, and stores the acceleration supplied from the interface 143 in the storage device 153. The storage device 152 includes a hard disk drive, a memory, a disk drive, and the like. The storage device 152 is installed with an orbital state analysis program and stores acceleration detected by the acceleration sensors 141 and 142. The storage device 152 is also used as a working storage area for the processing device 151.

  The input device 153 includes a keyboard, a mouse, and the like, and is a device that inputs data and various commands to the processing device 151. The display device 154 includes a CRT, an LCD, and the like, and displays an analysis result of the orbital state analysis program executed by the processing device 151 and the like.

〔processing〕
FIG. 3 shows a process flowchart of the orbital state analysis program according to the embodiment of the present invention.

  When acquiring the acceleration data in step S1-1, the processing device 151 executes analysis processing in step S1-2. The analysis process is performed by multi-resolution analysis using discrete wavelet transform. The processing device 151 displays the analysis result on the display device 154 in step S1-3.

[Analysis processing]
Here, the details of the analysis processing in step S1-2 will be described. In the analysis processing of this embodiment, multiresolution analysis is performed on the acquired acceleration signal using discrete wavelet transform.

(Waveform analysis)
First, waveform analysis will be described.

  As the waveform analysis, Fourier analysis is common. Fourier analysis is an analysis method in which information in the time domain is Fourier transformed and replaced with information in the frequency domain. However, time information is missing in normal Fourier analysis.

  For this reason, short-time Fourier transform, so-called window Fourier transform, has been proposed. However, in the short-time Fourier transform, the time-frequency analysis of the signal can be performed, but a certain amount of knowledge and prediction for the analysis target is required. For this reason, considerable effort was required to detect and analyze unknown signals.

  Therefore, in this embodiment, wavelet transform is used for waveform analysis.

  The wavelet transform used in the present embodiment represents a local state of a waveform f (x) to be analyzed by translating and expanding and contracting a small wave ψ (x) called a mother wavelet, and the waveform is analyzed based on this. It will be.

  The wavelet transform formula is generally

Given in.

  In equation (1), ψ ((x−a) / a) is an operation for (1 / a) the frequency shifted in time (phase) by b. The transformation by these operations is called continuous wavelet transformation. As is clear from the equation (1), the continuous wavelet transform requires a considerable amount of calculation and overlaps information.

  In order to simplify this, a wavelet transform called a discrete wavelet transform has been proposed.

  The discrete wavelet transform is a discretization of a and b in Equation (1).

It is expressed by the formula. This discrete wavelet transform generally has a feature that the amount of information is smaller than that of the continuous wavelet transform, but it is efficient and can be analyzed with high accuracy. Therefore, in this embodiment, waveform analysis is performed by this discrete wavelet transform. In this embodiment, the multi-resolution analysis is performed using the discrete wavelet transform at this time. Analyzing.

[Multi-resolution analysis]
Next, multiresolution analysis using discrete wavelet analysis will be described.

  FIG. 4 is a diagram for explaining the multi-resolution analysis.

  Assuming that the waveform to be analyzed is S, for example, as shown in FIG. 4, low-frequency components (approximation) a1, a2, a3, a4 and high-frequency components (detail) d1, d2, d3 having different frequency bands are used. , D4 is decomposed into eight hierarchical structures.

At this time, the analysis target waveform S in the hierarchical structure shown in FIG.
S = a4 + d4 + d3 + d2 + d1 (3)
It is represented by

〔Analysis result〕
Next, an analysis result by multi-resolution analysis using the discrete wavelet transform will be described.

[During normal driving (at high speed)]
First, analysis results during normal traveling and traveling at 40 km / h will be described.

[Results of acceleration detection in axle box]
FIG. 5 is a diagram showing the time variation of the vertical acceleration of the axle box, and FIG. 6 is a diagram showing the power spectrum density of the vertical acceleration of the inner rail side axle box. 5A shows the change over time of the vertical acceleration on the inner gauge side, and FIG. 5B shows the time change of the vertical acceleration on the outer gauge side. In FIG. 6, the characteristics shown in black indicate the characteristics of the curved portion, and the characteristics shown in gray indicate the characteristics of the straight line.

  FIG. 5 and FIG. 6 show the detection results of the vertical acceleration of the axle box during a curve run at a normal speed of 40 km / h. As shown in FIG. 5 (A), the inner track side vibrates violently, but as shown in FIG. 5 (B), the outer track side does not vibrate very much. This is a characteristic of wavy wear. In addition, as shown in FIG. 5 (B), it is thought that the vibration on the outer track side is transmitted through the shaft on the inner track side.

  At this time, in the power spectral density (PSD) of the vertical acceleration of the axle box, as shown in FIG. 6, the acceleration in the curve section is larger in all frequencies than in the straight section. Further, it can be seen that a lot of frequency components near 160 Hz are included.

  Wave wear at a traveling speed of 40 km / h is about 110 to 220 Hz from the wavelength. Therefore, in FIG. 6, it can be easily estimated that the increase in the frequency component in the vicinity of 160 Hz is the influence of the wavy wear.

[Detection results of acceleration in the vehicle body]
FIG. 7 is a diagram showing the time variation of the vertical acceleration of the vehicle body, and FIG. 8 is a diagram showing the power spectrum density of the vertical acceleration of the vehicle body. In FIG. 8, the characteristic shown in black indicates the curve part, and the characteristic shown in gray indicates the characteristic of the straight line part.

  With the time change of the vertical acceleration shown in FIG. 7, it is not possible to detect a change in wavy wear. Further, in the power spectral density shown in FIG. 8, the frequency component in the vicinity of 160 Hz is increased similarly to the power spectral density of the vertical acceleration of the axle box, and it can be easily estimated that it is an influence of wave wear. However, the position cannot be characterized by the power spectral density shown in FIG.

[Analysis result by multi-resolution analysis using discrete wavelet transform]
FIG. 9 shows a waveform diagram of multi-resolution analysis during normal running according to an embodiment of the present invention. 9A is a low frequency component a4, FIG. 9B is a high frequency component d4, FIG. 5C is a high frequency component d3, FIG. 5D is a high frequency component d2, and FIG. 5E is a high frequency component d1. The waveform diagram of is shown. Here, the sampling frequency is 2 kHz, the high frequency component d1 is 500 to 1 kHz, the high frequency component d2 is 250 to 500 Hz, the high frequency component d3 is 125 to 250 Hz, the high frequency component d4 is 62.5 to 125 Hz, and the low frequency component. a4 shows a waveform corresponding to a frequency band of 62.5 Hz or less.

  The high-frequency components d1 and d4 shown in FIGS. 9E and 9B do not change significantly with respect to time, whereas the high-frequency components shown in FIGS. 9C and 9D. A relatively large change is seen in d3 and d2. In particular, the high-frequency component d3 shown in FIG. 9C has a characteristic of 125 to 250 Hz, and matches the frequency of 160 Hz of wave wear. Therefore, it becomes possible to detect wave wear by paying attention to the high frequency component d3 shown in FIG.

  It can be seen that the high-frequency component d3 shown in FIG. 9C shows a change substantially equivalent to the acceleration of the axle box shown in FIG. Since FIG. 9 shows the change with respect to time, the time can be specified, and thus the position can be specified.

[Low speed]
Next, an analysis result during traveling at a low speed and traveling at 20 km / h will be described.

[Results of acceleration detection in axle box]
FIG. 10 is a diagram showing the time change of the vertical acceleration on the inner rail side of the axle box, and FIG. 11 is a diagram showing the power spectrum density of the vertical acceleration of the inner rail side axle box.

  FIG. 10 and FIG. 11 both show the detection results of the vertical acceleration of the axle box when traveling on a curve at a low speed and 20 km / h. As shown in FIG. 10, it can be seen that the inner rail side of the axle box vibrates violently, and the feature of wave wear appears.

  At this time, it can be seen that the power spectral density (PSD) of the vertical acceleration of the axle box includes a lot of frequency components near 80 Hz as shown in FIG.

  The wavy wear at a traveling speed of 20 km / h is about 55 to 110 Hz, which is a half of the frequency when traveling at high speed. Therefore, in FIG. 11, it can be easily estimated that the increase in the frequency component in the vicinity of 80 Hz is an influence of wave wear.

[Detection results of acceleration in the vehicle body]
FIG. 12 is a diagram showing the time variation of the vertical acceleration of the vehicle body, and FIG. 13 is a diagram showing the power spectrum density of the vertical acceleration of the vehicle body.

  With the time change of the vertical acceleration shown in FIG. 12, the change of the wavy wear cannot be detected. Further, in the power spectral density shown in FIG. 13, the frequency component in the vicinity of 80 Hz is increased similarly to the power spectral density of the vertical acceleration of the axle box, and it can be easily estimated that it is an influence of wave wear. However, the position cannot be characterized by the power spectral density shown in FIG.

[Analysis result by multi-resolution analysis using discrete wavelet transform]
FIG. 14 is a waveform diagram of multi-resolution analysis during low-speed running according to an embodiment of the present invention. 14A is a low frequency component a4, FIG. 14B is a high frequency component d4, FIG. 14C is a high frequency component d3, FIG. 14D is a high frequency component d2, and FIG. 14E is a high frequency component d1. The waveform diagram of is shown. Here, the sampling frequency is 2 kHz, the high frequency component d1 is 500 to 1 kHz, the high frequency component d2 is 250 to 500 Hz, the high frequency component d3 is 125 to 250 Hz, the high frequency component d4 is 62.5 to 125 Hz, and the low frequency component. a4 shows a waveform corresponding to a frequency band of 62.5 Hz or less.

  The high frequency components d1 and d2 shown in FIGS. 14 (E) and 14 (D) do not change significantly with respect to time, whereas the high frequency components shown in FIGS. 14 (C) and 14 (B). A relatively large change is seen in d3 and d4. In particular, the high frequency component d4 shown in FIG. 14B has a characteristic of 62.5 to 125 Hz, and coincides with the frequency of 80 Hz of wavy wear. Therefore, during low-speed traveling, it is possible to detect wave wear by paying attention to the high-frequency component d4 shown in FIG.

  It can be seen that the high-frequency component d4 shown in FIG. 14B shows a change substantially equivalent to the acceleration of the axle box shown in FIG. Since FIG. 14 shows changes with respect to time, the time can be specified, and thus the position can be specified. At this time, by integrating the longitudinal acceleration in the direction of the arrow X detected by the acceleration sensor 142 twice, the distance can be obtained and the position can be corrected.

〔result〕
In this way, by detecting the vertical acceleration of the vehicle body and performing multi-resolution analysis using discrete wavelet transform, it is possible to detect wavy wear based on changes with time. For this reason, while being able to detect wavy wear reliably, the position which has generate | occur | produced can be detected.

〔effect〕
According to the present embodiment, since the acceleration sensor may be attached to the vehicle body, the attachment can be performed much more easily than the axle box or the like. Therefore, it can be easily attached to the floor of a business vehicle or the like.

  Further, since the analysis can be performed by a relatively simple process such as discrete wavelet transform, the analysis can be performed using a personal computer or the like.

  Therefore, the state of the trajectory can be analyzed accurately with a simple configuration and at a low cost. At this time, a threshold value may be set in advance in the analysis result, and an abnormality may be notified when the threshold value is exceeded.

[Application example]
FIG. 15 shows a system configuration diagram of a first application example of the present invention. In the figure, the same components as in FIG.

  The track state analysis system 200 of this application example has a configuration in which a computer 213 in which acceleration sensors 141 and 142 and a data recorder 211 are mounted on a vehicle 111 and a track state analysis program is installed in a data processing center 212 is provided. The acceleration data detected by the acceleration sensors 141 and 142 when the vehicle 111 is traveling is stored in the data recorder 211, and after the vehicle 112 is traveling, the data stored in the data recorder 211 is transferred to the storage medium 214 or the like and stored. The medium 214 is brought into the data processing center 212, taken into the computer 213, and multi-resolution analysis using the above-described discrete wavelet transform is performed to analyze the orbital state.

  According to this application example, a business vehicle or the like can be used, and the track state can be analyzed simply by installing a personal computer in which the track state analysis program is installed. Therefore, a system can be constructed at a low cost with a simple configuration.

  FIG. 16 shows a system configuration diagram of a second application example of the present invention. In the figure, the same components as those in FIGS. 1 and 15 are denoted by the same reference numerals, and the description thereof is omitted.

  In the track state analysis system 300 of this application example, acceleration sensors 141 and 142 and a data communication device 311 are provided in a vehicle 111, a data communication device 312 and a computer 213 are provided in a data processing center 212, and the acceleration sensors 141 and 142 detect them. The data communication device 311 transmits the measured acceleration to the data communication device 312 provided in the data processing center 212. In the data processing center 212, the data received by the data communication device 321 is supplied to the computer system 221. The computer system 213 stores the acceleration received by the data communication device 321. The computer system 213 performs multi-resolution analysis on the stored acceleration using the above-described discrete wavelet transform, and analyzes the orbital state.

  According to this application example, by performing data communication between the vehicle 111 and the computer system 213 of the data processing center 212, acceleration can be acquired in real time and analysis processing can be performed.

[Others]
In this embodiment, the acceleration sensor is provided on the vehicle body, but the acceleration sensor may be provided on the axle box.

  In this embodiment, an appropriate analysis result is selected from the multi-resolution analysis result. However, an optimal analysis result may be selected from the traveling speed or the like.

  Furthermore, in this embodiment, the state of the rail of the railway is described as an example of the track. However, the track that can be analyzed in this embodiment is not limited to the rail of the rail, and other tracks can also be analyzed. It goes without saying that a similar analysis can be performed.

It is a system configuration figure of one example of the present invention. It is a block block diagram of the orbital state detection apparatus 111. It is a process flowchart of one Example of this invention. It is a figure for demonstrating multi-resolution analysis. It is a figure which shows the time change of the vertical acceleration of an axle box. It is a figure which shows the power spectrum density of the vertical acceleration of an inner track side axle box. It is a figure which shows the time change of the vertical acceleration of a vehicle body. It is a figure which shows the power spectrum density of the vertical acceleration of a vehicle body. It is a wave form diagram of the multi-resolution analysis at the time of normal driving of one example of the present invention. It is a figure which shows the time change of the vertical acceleration of the inner track side of an axle box. It is a figure which shows the power spectrum density of the vertical acceleration of an inner track side axle box. It is a figure which shows the time change of the vertical acceleration of a vehicle body. It is a figure which shows the power spectrum density of the vertical acceleration of a vehicle body. It is a wave form diagram of the multiresolution analysis at the time of low speed driving of one example of the present invention. It is a system configuration | structure figure of the 1st application example of this invention. It is a system configuration | structure figure of the 2nd application example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 System 111 Track, 112 Vehicle, 113 Track state analysis device 121 Car, 122 Car body 131 Wheel, 132 Suspension 141, 142 Acceleration sensor, 143 Interface 144 Computer system 151 Processing device, 152 Storage device, 153 Input device, 154 Display device

Claims (12)

A track state analysis method for analyzing a track state in which a vehicle travels,
An acceleration acquisition procedure for acquiring acceleration applied to the vehicle;
An orbital state analysis method comprising: an analysis procedure for performing multi-resolution analysis on the acceleration acquired by the acceleration acquisition procedure using wavelet transform and analyzing the state of the orbit from the analysis result. The trajectory state analysis method according to claim 1, wherein the acceleration acquisition procedure detects an acceleration applied in a vertical direction of a vehicle body of the vehicle. 3. The orbital state detection method according to claim 1, wherein the analysis procedure uses discrete wavelet transform as the wavelet transform. 4. The method according to claim 1, wherein the analysis procedure analyzes the state of the trajectory using an analysis result reflecting a state of the trajectory to be detected among the results of the multi-resolution analysis. Orbit state detection method described in the item. A track state detection device for detecting a state of a track on which a vehicle travels,
An acceleration detector for detecting acceleration applied to the vehicle;
An orbital state detection apparatus comprising: an analysis unit that performs multi-resolution analysis on the acceleration detected by the acceleration detection unit using wavelet transform and analyzes the state of the orbit from the analysis result. 6. The track state detection device according to claim 5, wherein the acceleration detection unit detects an acceleration applied to a vehicle body of the vehicle. The orbital state detection device according to claim 5 or 6, wherein the pre-analysis unit uses discrete wavelet transform as the wavelet transform. The said analysis part analyzes the state of the said orbit using the analysis result in which the state of the orbit to be detected was reflected as a result of multi-resolution analysis. Orbital state detection device. On the computer,
An acceleration acquisition procedure for acquiring the acceleration applied to the vehicle traveling on the track;
A computer-readable program for performing multi-resolution analysis on the acceleration acquired in the acceleration acquisition procedure using wavelet transform and executing the analysis procedure for analyzing the state of the trajectory from the analysis result . The computer-readable program according to claim 9, wherein the acceleration acquired in the acceleration acquisition procedure is an acceleration applied to a vehicle body of the vehicle. The computer-readable program according to claim 9 or 10, wherein the analysis procedure uses discrete wavelet transform as the wavelet transform. 12. The computer according to claim 9, wherein the analysis procedure detects the state of the trajectory using an analysis result reflecting a state to be detected as a result of multi-resolution analysis. A readable program.