JP7708616B2 - Railway pad protrusion amount abnormality detection system and abnormality detection method - Google Patents

Description

The present invention relates to a track pad protrusion amount abnormality detection system and an abnormality detection method that detects abnormalities in the amount of protrusion of track pads installed under the rails of a railway track.

Conventionally, in slab track, steel plate track pads (hereinafter simply called track pads) are installed under the rails to allow the rails laid along the track to slide as the rails expand and contract. Such track pads can shift in the longitudinal direction of the rails and become dislodged due to repeated expansion and contraction of the rails.
Therefore, a railway pad protrusion amount measuring device that measures the protrusion amount of the railway pad has been proposed (see, for example, Japanese Patent Application Laid-Open No. 2003-233664).

The track pad protrusion amount measuring device described in Patent Document 1 is equipped with a position detection unit that detects the installation position of the track pad by detecting parts other than the track pad that make up the fastening part of the rail (such as fastening bolts), and a photographing unit that takes an image of the fastening part of one rail from the side of the other rail. The photographing unit is controlled in synchronization with the output of the position detection unit to take an image, and the image is processed to calculate the protrusion amount of the track pad, thereby measuring the protrusion amount of the track pad when it is misaligned in the longitudinal direction of the rail.

JP 2017-187487 A

Conventionally, in the maintenance and management of track pads, if the protrusion amount measured by the track pad protrusion amount measuring device described above exceeds a preset value, repair work is uniformly performed. This makes periodic repair unnecessary and makes it possible to perform repairs according to the condition. However, in the method described above where the protrusion amount is determined to be abnormal under the same conditions and repairs are performed, the judgment criteria (threshold value) must be set low to ensure safety, which increases the number of track pads that are determined to need repair, which increases the burden of the repair work.

The present invention has been made in consideration of the above problems, and its purpose is to provide a system and method for detecting abnormalities in the amount of track pad protrusion that can reduce the number of track pads to be repaired without overlooking those that really need repair, and reduce the burden of repair work.
Another object of the present invention is to provide a system and method for detecting abnormalities in the amount of rail pad protrusion that can distinguish between abnormalities that require major repairs, such as adjusting the position of the rail pad or replacing it, and abnormalities that can be repaired by relatively simple repair work, such as adjusting the rail fastening device.

Prior to the present invention, the inventors investigated the number of track pad misalignments detected in a certain line section over the past two years to verify whether it would be possible to efficiently manage and repair track pads by judging under different conditions depending on the state of the track and the environment, rather than repairing track pads whose protrusions exceed the management value under the same conditions for the entire section. The number of times pad misalignments occurred for every 100 m of rail obtained from the investigation was shown in the graph in Figure 4 (A).

As a result, it was found that track pad misalignment was concentrated in certain sections. Therefore, the locations of sections where misalignment occurred frequently and rarely were investigated. As a result, as shown in Figure 4 (B), it was revealed that the sections where misalignment occurred frequently were long-rail movable sections where expansion joints are present and the rail expands and contracts with temperature changes, while the sections where misalignment occurred least were long-rail immobile sections where the rail axial force and the ballast longitudinal resistance force are balanced and the rail does not expand or contract. Additionally, the sections where misalignment occurred very rarely were tunnel sections. Regarding fastening devices for expansion joints, for example, there is an invention described in JP 2020-16010 A.

Therefore, to gain a more detailed understanding of the phenomena occurring in the movable rail section, we investigated the amount of movement of the rail and track pad at the point where the expansion joint is installed over a period of several months, along with the rail temperature. The amount of movement of the rail and pad was calculated from the number of pixels by processing images taken at specified times every few weeks using a camera installed under the vehicle, and the rail temperature was measured using a non-contact thermometer near the expansion joint, which was the target point of the investigation. In addition, we used as a reference the temperature measured by the temperature measuring device closest to the movement measurement point, out of the rail temperature measuring devices already installed at specified intervals along the rail. The results are shown in Figures 5 and 6.

The graphs in Figures 5(A), (B) and 6 show the results of measuring the absolute amount of movement of the "rail" and "rail pad" in three different rail movable sections, with the horizontal axis of the graph showing rail temperature arranged in chronological order, and the vertical axis showing the measured amount of movement. The bar graphs show the amount of pad movement, and the line graphs show the amount of rail movement. Figures 5 and 6 show that the movement of the pads is almost the same as the movement of the rails. It can also be seen that the movement of the rails is related to temperature. Note that a method of calculating the amount of movement of the rail pads from the number of pixels by processing an image taken by a camera installed under the vehicle is described in, for example, the aforementioned Patent Document 1, and a similar method can be used.

Next, the inventors plotted the measured rail movement on a coordinate system with temperature on the horizontal axis and track pad movement on the vertical axis, and illustrated the fluctuation trajectory by connecting each point in time series. The result is shown in Figure 7. Note that Figure 7 is based on measurements taken over a period of about nine months from the beginning of winter to the end of summer, and each of the trajectory lines A, B, and C is the fluctuation trajectory of the measurements taken at different points.
It can be seen from Figure 7 that the fluctuation trajectory of the pad movement amount resembles a hysteresis curve. The reason why the fluctuation trajectory of the pad movement amount shows hysteresis characteristics is that the track pad moves in response to the expansion and contraction of the rail, but the amount of expansion of the rail gradually increases from spring to summer and gradually contracts from autumn to winter.

The expansion of the rail causes the rail pad underneath to move in the positive direction, while the contraction of the rail causes the rail pad to move in the negative direction, but the rails move in each direction successively, reaching their maximum in summer or winter.Then, as soon as the rail starts to move in the opposite direction due to a reversal in temperature change, the speed of movement increases, and then the speed of movement gradually slows down.
According to this way of thinking, it can be estimated that in the missing parts of each of the trajectories A, B, and C shown in FIG. 7, i.e., in the period with no measurement values (autumn), the changes generally follow a line connecting the start point and the end point, that is, the changes overall follow a loop line similar to a hysteresis curve.

Also, from the graph in Figure 7, for example, the amount of movement of track line A increases to a maximum of 40 mm, but then starts to decrease and decreases to the same extent as the other tracks B and C, and it can be inferred that this is simply a repetition of increases and decreases due to installation conditions and seasonal factors, and that there is no particular abnormal movement. However, if a conventional abnormality judgment was made using an upper limit value (for example, 30 mm), even such track pads would be judged to be abnormal, leading to a situation where unnecessary repair work would be carried out.

On the other hand, the results of the above-mentioned verification conducted by the inventors revealed that the track pads to be detected behave as described above when they are pads in the movable rail section. Therefore, it was found that such track pads can be excluded from the scope of repair work by setting the upper limit value to a value larger than the conventional value (for example, 41 mm).

The present invention has been made based on the above findings, and in order to achieve the above object,
A track pad protrusion amount abnormality detection system comprising: an image collecting device that collects image data of the vicinity of the rail captured by an imaging device mounted on a vehicle while the vehicle is traveling; a rail temperature collecting device that measures the temperature of each part of the rail along the longitudinal direction of the rail and collects rail temperature data; and an abnormality detection device that detects abnormalities in the amount of track pad protrusion based on the image data collected by the image collecting device and the temperature data collected by the rail temperature collecting device,
The abnormality detection device includes:
a movement amount calculation means for calculating a movement amount of the rail pad by image processing the image data collected by the image collection device;
a storage means for storing a trajectory loop line serving as a preset judgment criterion based on information regarding the position of the rail movable section to be detected and a trajectory line in which a plurality of movement amounts of the track pad during a predetermined period obtained from the movement amount calculation means are plotted in chronological order on a coordinate system having temperature and movement amount as coordinate axes;
a movement amount determination means for determining whether the movement amount of the track pad in the rail movable section obtained from the movement amount calculation means is outside or inside the track loop line serving as a reference;
a section determination means for determining whether a track pad included in an image to be determined is in a movable rail section or an immovable rail section,
The movement amount determination means is configured to output an abnormality when it is determined that the movement amount obtained from the movement amount calculation means is outside the trajectory loop line serving as a determination criterion for a track pad determined by the section determination means to be in the rail movable section.

The track pad protrusion amount abnormality detection system configured as described above outputs an abnormality when the amount of movement of a track pad determined to be in a movable rail section is determined to be outside the track loop line that is the judgment criterion. This makes it possible to detect abnormalities using a different criterion from the abnormality judgment criterion for the amount of movement of a track pad in a stationary rail section. This makes it possible to reduce the number of track pads to be repaired without overlooking objects that really need repair, and to reduce the burden of repair work.

Also, preferably, the storage means stores an upper limit of the amount of protrusion of the rail pad,
When the movement amount determination means determines that the movement amount of the rail pad obtained from the movement amount calculation means is outside the trajectory loop line serving as a reference and exceeds the upper limit value, the movement amount determination means outputs a first abnormal state;
The movement amount determination means is configured to output a second abnormal state when it determines that the movement amount of the rail pad obtained from the movement amount calculation means is outside the reference trajectory loop line and does not exceed the upper limit value.
With this configuration, it is possible to distinguish and detect abnormalities that require major repair work, such as correcting the position of the track pad or replacing it, from abnormalities that can be repaired with relatively simple repair work, such as adjusting the rail fastening device.

Furthermore, preferably, the movement amount determination means is configured to output an abnormality when it determines that the movement amount obtained from the movement amount calculation means exceeds a predetermined threshold value which is smaller than the upper limit value for a track pad which has been determined by the section determination means to be in a rail immobile section.
According to this configuration, since abnormalities can be detected for track pads in immovable rail sections according to the same abnormality determination criteria as in the past, consistent maintenance management of track pads can be achieved.

In addition, a method for detecting an abnormality in the amount of protrusion of a railway pad having a configuration according to another aspect of the present invention is as follows:
A method for detecting abnormalities in the amount of protrusion of a track pad, comprising: collecting image data of the vicinity of the rail by an imaging device mounted on a vehicle while the vehicle is traveling, and temperature data of each part of the rail along the longitudinal direction of the rail; and detecting abnormalities in the amount of protrusion of a track pad based on the collected image data and temperature data,
collecting the image data and temperature data by photographing the vicinity of the rail in the rail movable section to be detected a plurality of times at predetermined intervals over a predetermined period of at least six months;
a step of calculating the amount of movement of the track pad by performing image processing on the collected image data of the rail movable section, and setting a reference trajectory loop line in advance for each rail movable section based on a trajectory line in which the calculated multiple amounts of movement of the track pad during the predetermined period are plotted in chronological order on a coordinate system having temperature and amount of movement as coordinate axes;
a step of calculating the amount of movement of the track pad by processing image data of the vicinity of the rail captured by an imaging device mounted on the vehicle;
determining in which rail movable section a track pad included in an image to be determined is located;
determining whether the amount of movement calculated for a rail pad in a rail movable section is outside or inside the track loop line set corresponding to the rail pad, and outputting an abnormality when the amount of movement is outside the track loop line;
The present invention is designed to have the following features.

According to the method for detecting abnormalities in the amount of track pad protrusion that has the above-mentioned steps, an abnormality is output when the amount of movement of a track pad that is determined to be in a movable rail section is determined to be outside the track loop line that is the judgment criterion, so that abnormalities can be detected using a different criterion from the abnormality judgment criterion for the amount of movement of a track pad in a stationary rail section. This makes it possible to reduce the number of track pads to be repaired without overlooking objects that really need repair, and to reduce the burden of repair work.

Also, preferably, when it is determined that the movement amount calculated for a rail pad in a rail movable section is outside the trajectory loop line set corresponding to the rail pad and exceeds a preset upper limit value of the rail pad protrusion amount, a first abnormal state is output,
When it is determined that the movement amount calculated for a track pad in the rail movable section is outside the trajectory loop line set for that track pad and does not exceed the upper limit value, a second abnormal state is output.
This method makes it possible to distinguish and detect abnormalities that require major repair work, such as correcting the position of the track pads or replacing them, from abnormalities that can be repaired with relatively simple repair work, such as adjusting the rail fastening device.

Further, preferably, the predetermined period is 9 months or more,
The locus loop line is a parallelogram circumscribing a locus line that represents a plurality of movements of the rail pad during the predetermined period in a time series on a coordinate system having temperature and movement as coordinate axes.
According to this method, it is possible to set a locus loop line as a judgment criterion that can obtain a good accuracy in detecting an abnormality relatively efficiently.

The track pad protrusion amount abnormality detection system and abnormality detection method according to the present invention can reduce the number of track pads to be repaired and the burden of repair work. It also has the effect of being able to distinguish and detect abnormalities that require major repairs such as adjusting or replacing the track pad position from abnormalities that can be repaired with relatively simple repair work such as adjusting the rail fastening device.

1 is a schematic diagram showing an example of the configuration of a rail pad protrusion amount measurement system according to an embodiment of the present invention; 11 is a graph showing an example of a trajectory loop line that serves as an abnormality determination criterion for the amount of movement of the rail pad in the system of the embodiment. 10 is a flowchart showing a procedure for a rail pad protrusion amount abnormality detection process in the embodiment. (A) is a graph showing the number of occurrences of track pad displacement detected over the past two years on a certain line section for every 100 m of rail, and (B) is a graph showing the relationship between the number of occurrences of track pad displacement in (A) in movable sections and immovable sections. Graph (A) shows the results of investigating the amount of movement of the rail and track pad at a certain point where an expansion joint is located, together with the rail temperature, over a period of several months. Graph (B) shows the results of investigating the amount of movement of the rail and track pad at another point, together with the rail temperature, over a period of several months. The graph shows the results of investigating the amount of rail and track pad movement at other locations over several months, along with the rail temperature. 1 is a graph showing a fluctuation trajectory obtained by plotting rail movement amounts measured in three rail movable sections on a coordinate system with temperature on the horizontal axis and movement amount on the vertical axis, and connecting each point in a chronological order.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a rail pad protrusion amount abnormality detection system and an abnormality detection method according to an embodiment of the present invention will be described in detail with reference to the drawings.
First, an embodiment of a rail pad protrusion amount abnormality detection system according to the present invention will be described with reference to Fig. 1. Fig. 1 is a schematic configuration diagram showing the abnormality detection system as a functional block diagram.

As shown in FIG. 1, the track pad protrusion amount abnormality detection system of this embodiment is composed of an image collection device 10 equipped with a control device 11 mounted on a vehicle such as a measurement vehicle that inspects the condition of the track, an image capture device 12 such as a CCD camera or CMOS camera that captures the area near the rail including the track pad area, a number of temperature measuring devices 21 installed at predetermined distances along the track to measure the temperature of the rail, a temperature data collection device 22 that collects image data from the image collection device 10 and temperature data measured by the temperature measuring devices 21, and an abnormality detection device 30 consisting of a data processing device that receives the temperature data collected by the temperature data collection device 22 and detects abnormalities in the amount of movement of the track pad based on these data.

The imaging device 12 is equipped with a pair of cameras mounted under the vehicle, and as described in Patent Document 1, in order to photograph the track pad portion (the fastening portion of the rail) from an oblique direction, it is preferable to place the cameras above the rail on the opposite side to the rail on which the track pad is located. The timing of photographing can be determined by detecting parts such as fastening bolts on the track using a laser displacement meter or the like.
The image collecting device 10 includes a storage device 13 that stores data such as images captured by the imaging device 12, a position detector 14 such as a speed generator for grasping the vehicle position, a timer 15 that keeps track of the current time, and a wireless communication device 16. The abnormality detecting device 30 includes an arithmetic processing device 31, a wireless communication device 32, a storage device 33, a display device 34, etc.

The control device 11 of the image collecting device 10 and the arithmetic processing device 31 of the abnormality detecting device 30 can be configured with a normal microprocessor or microcomputer having a CPU (Central Processing Unit), ROM, RAM, etc., and have the function of controlling each part in cooperation with the program stored in the ROM and the CPU. In addition, the arithmetic processing device 31 is configured with a program having the function of calculating the pad protrusion amount (movement amount) from the number of pixels based on image data of the track pad photographed by a camera.
The control device 11 of the image collection device 10 is programmed to grasp the position of the vehicle based on information from the position detector 14, and when it approaches the movable rail section, to photograph the track pad area with the imaging device 12 and store the data of the photographed image in the memory device 13 together with the time data of the timer 15.

The storage device 33 of the image collection device 10 also stores position information of the rail movable section and a reference trajectory loop line L as shown in Fig. 2, which is set for each rail movable section. Here, the reference trajectory loop line L is formed by drawing a circumscribed parallelogram of the variation trajectory line of the variation amount based on the pad movement amount calculated based on image data of the track pad collected in advance over a period of six months or more.
Incidentally, the fluctuation trajectory line T in Fig. 2 corresponds to the fluctuation trajectory line A in Fig. 7. For trajectories with small loops such as trajectories B, C, and D in Fig. 7, a parallelogram obtained by enlarging the position of the upper side of the circumscribed parallelogram of the movement amount trajectory by moving it to an upper limit value (determination threshold value) max for determining an abnormality in the movement amount and by moving the position of the lower side to a lower limit value min may be set as the reference trajectory loop line.

As shown by the ● mark in Fig. 2, when the upper limit value max is exceeded, the rail pad becomes easily dislodged, so repair work such as returning the rail pad to its original position or replacing it is performed. The upper limit value max of the pad movement amount was previously set to a value such as 30 mm, but in this embodiment, it is set to a value such as 40 mm. This reduces the number of pads that are judged to be abnormal, and reduces the frequency of large-scale repair work such as adjusting the position of the pad or replacing it.
Furthermore, the abnormality detection system of this embodiment has a function of determining that an abnormality has occurred if the detected pad movement amount is outside the reference locus loop line L, as shown by the ▲ mark in Figure 2, even if it does not exceed the upper limit value max.

It has been known that the amount of movement of the track pads suddenly increases within the upper limit max in a shorter period than usual due to factors such as deterioration of the variable pads that support the pads, deterioration of the CA mortar under the slab, and factors such as bridge tilt, sudden earthquakes, and warping of the slab due to temperature changes. In addition, there are also cases where the amount of movement of the track pads increases suddenly even without the above phenomena, and when the cause is investigated, it is found that the bolts of the rail fastening device have become loose.
In such a case, the pad movement amount will be a value as shown by the ▲ mark in Fig. 2. Therefore, in the abnormality detection system of this embodiment, even if the amount of movement does not exceed the upper limit value max but is significantly outside the fluctuation trajectory line T of the fluctuation amount as shown by the ▲ mark in Fig. 2, it is possible to determine that an abnormality has occurred based on the reference trajectory loop line L. Then, the cause of the displacement of the pad position is investigated, and if the cause is, for example, a loose bolt of the rail fastening device, repair work such as tightening the bolt is performed. Even if the cause of the abnormality is not a loose bolt of the rail fastening device, work to eliminate the cause is performed.

Next, a method for detecting an abnormality in the amount of protrusion of a rail pad according to an embodiment of the present invention will be described in detail with reference to the flow chart of FIG.
In the method for detecting abnormality in the amount of track pad protrusion of this embodiment, first, for a period of at least six months, preferably nine months to one year, images of the track pad portion at the moving end of the rail movable section are taken by the image collecting device 10 at approximately the same predetermined time each time in each rail movable section, and the image data is stored (step S1). In parallel with this, rail temperature data measured by temperature measuring devices 21 arranged along the track is collected by the temperature data collecting device 22.

Next, the image data of the track pad accumulated in step S1 is image processed to calculate the amount of movement of the track pad, and from the calculated amount of movement and the rail temperature data collected by the temperature data collection device 22, the measurement data of the temperature measuring device closest to the time when the image data was captured and closest to the movable section captured is extracted, and the measurement data is plotted on the temperature-amount of movement coordinates as shown in Figure 2. Then, a fluctuation trajectory line T of the amount of fluctuation is drawn by connecting the plotted points in chronological order (step S2). Next, a parallelogram circumscribing the fluctuation trajectory line T of the amount of fluctuation is drawn and set as the reference trajectory loop line L (step S3).

Then, a vehicle equipped with the image collection device 10 is driven, and at the same time as above, the image collection device 10 takes images of the track pad portion at the moving end side in each movable rail section, and stores the image data and shooting position information (movable section information) (step S4). When shooting of all sections is completed, the image collection device 10 transmits the stored data to the anomaly detection device 30 (step S5). The anomaly detection device 30 then receives the image data and reads the rail temperature data for the corresponding time and in the vicinity of the movable section from the temperature data collection device 22 (step S6).

Next, the abnormality detection device 30 calculates the amount of movement of the rail pad from the image data for each movable section of interest (step S7). Then, it is determined whether the calculated amount of movement is outside the reference trajectory loop line L shown in Fig. 2 (step S8). If the result is No, that is, if it is determined that the calculated amount of movement is inside the reference trajectory loop line L shown in Fig. 2, the process returns to step S7 and the amount of movement of the rail pad is calculated from the image data for the next movable section.
Furthermore, if the abnormality detection device 30 determines in step S8 that the calculated movement amount is outside the reference locus loop line L shown in FIG. 2 as Yes, the abnormality detection device 30 proceeds to step S9 and determines whether the calculated movement amount exceeds the upper limit value max.

If it is determined in step S9 that the amount of movement exceeds the upper limit max (Yes), the process proceeds to step S10, where an abnormality that requires repair such as correction of the pad position or replacement is displayed on the display device 34 together with the position information (distance in kilometers) of the corresponding location. If it is determined in step S9 that the amount of movement does not exceed the upper limit max (No), the process proceeds to step S11, where an abnormality that requires adjustment of the rail fastening device (tightening of the bolts) is displayed on the display device 34 together with the position information (distance in kilometers) of the corresponding location.
Then, the process proceeds to step S12 to determine whether or not the process has been completed for all the rail pads in the movable rail section, and if the process is determined to be not completed (No), the process returns to step S7 to repeat the above process. On the other hand, if the process is determined to be completed for all the rail pads in the movable rail section (Yes) in step S12, the abnormality detection process is terminated.

As described above, the abnormality detection process can distinguish between serious abnormalities that require major correction of pad position, replacement, etc., and minor abnormalities that require adjustment of the rail fastening device (tightening of bolts, etc.), and can instruct the worker on the repair content corresponding to each abnormality.
In addition, by setting the upper limit value max to a value larger than before, it is possible to reduce the frequency with which it is determined that a serious abnormality has occurred in the rail movable section, requiring significant adjustment of the pad position, replacement, etc., thereby reducing the burden on the repair work department.

The above abnormality detection process is for track pads in movable rail sections, and for track pads in immovable rail sections, a threshold value (e.g. 30 mm) smaller than the upper limit max for determining movable rail sections is used as the conventional threshold value for judgment. Position information for movable rail sections is stored in the storage device 33 of the abnormality detection device 30, and when the calculation processing device 31 judges that the position of the track pad in the read image is in a movable rail section, it performs abnormality detection processing according to the flowchart in Figure 3, and when the position of the track pad in the image is not in a movable rail section, it judges it to be a track pad in an immovable section and judges the presence or absence of an abnormality using the conventional threshold value.

Although the invention made by the inventor has been specifically described above based on an embodiment, the present invention is not limited to the above embodiment. For example, in the above embodiment, it was described that in the track pad protrusion amount measuring device 10, the image of the track pad taken by the camera (12) is stored together with distance data and time data, but when the time when the measurement is started or ended, or the image data obtained in a series of measurements is stored in a single file, the time when the file was created or updated may be transmitted to the abnormality detection device 30 together with the image data as the acquisition time of the image of the track pad.

In the above-described case, the abnormality detection device 30 calculates the time when the image of each track pad was taken from the received time data and distance data, and obtains temperature information measured by the temperature measuring device 21 closest to the target track pad at that time from the rail temperature collection device 22, thereby making it possible to detect abnormalities in the amount of movement of the track pad in a manner similar to the abnormality detection process shown in FIG. 3.
In the above embodiment, the rail temperature is detected by temperature measuring devices 21 installed on the rails at predetermined intervals along the track. However, it is also possible to mount a non-contact temperature measuring device such as an infrared radiation temperature sensor that uses infrared rays or the like under the vehicle and use that temperature measuring device to measure the temperature of the side of the rail near the track pad when the track pad is photographed.

10 Image collection device 11 Control device (control means)
12 Imaging device (photographing means)
13 Storage device (storage means)
14 Position detector (position detection means)
15 Timer for measuring time 16 Wireless communication device 21 Temperature measuring device (rail temperature detection means)
22 Temperature data collection device (rail temperature collection device)
30 Abnormality detection device 31 Processing device 32 Wireless communication device 33 Storage device 34 Display device

Claims (6)

A track pad protrusion amount abnormality detection system comprising: an image collecting device that collects image data of the vicinity of the rail captured by an imaging device mounted on a vehicle while the vehicle is traveling; a rail temperature collecting device that measures the temperature of each part of the rail along the longitudinal direction of the rail and collects rail temperature data; and an abnormality detection device that detects abnormalities in the amount of track pad protrusion based on the image data collected by the image collecting device and the temperature data collected by the rail temperature collecting device,
The abnormality detection device includes:
a movement amount calculation means for calculating a movement amount of the rail pad by image processing the image data collected by the image collection device;
a storage means for storing information on the position of the rail movable section to be detected , which is a section where an expansion joint is interposed and the rail expands and contracts with temperature changes, and a trajectory loop line that serves as a preset judgment criterion based on a trajectory line that represents multiple amounts of movement of the track pad in a predetermined period obtained from the movement amount calculation means in time series on a coordinate system with temperature and amount of movement as coordinate axes;
a movement amount determination means for determining whether the movement amount of the track pad in the rail movable section obtained from the movement amount calculation means is outside or inside the track loop line serving as a reference;
a section determination means for determining whether a track pad included in an image to be determined is in the rail movable section or in a rail immobile section where the rail axial force and the ballast longitudinal resistance force are balanced and the rail does not expand or contract,
a track pad protrusion amount abnormality detection system, characterized in that the movement amount determination means outputs an abnormality when it determines that the movement amount obtained from the movement amount calculation means is outside the trajectory loop line that is the determination criterion for a track pad that has been determined by the section determination means to be in the rail movable section. The storage means stores an upper limit value of the amount of protrusion of the rail pad,
when the movement amount determination means determines that the movement amount of the rail pad obtained from the movement amount calculation means is outside the trajectory loop line serving as a reference and exceeds the upper limit value, it outputs a first abnormal state;
The rail pad protrusion amount abnormality detection system according to claim 1, characterized in that the movement amount determination means outputs a second abnormal state when it determines that the movement amount of the rail pad obtained from the movement amount calculation means is outside the trajectory loop line serving as a reference and does not exceed the upper limit value. The track pad protrusion amount abnormality detection system according to claim 2, characterized in that the movement amount determination means outputs an abnormality when it determines that the movement amount obtained from the movement amount calculation means exceeds a predetermined threshold value that is smaller than the upper limit value for a track pad that is determined by the section determination means to be in the rail immobile section. A method for detecting abnormality in the amount of protrusion of a track pad, comprising: collecting image data of the vicinity of the rail by an imaging device mounted on a vehicle while the vehicle is traveling, and temperature data of each part of the rail along the longitudinal direction of the rail; and detecting abnormality in the amount of protrusion of a track pad based on the collected image data and temperature data,
A step of collecting the image data and temperature data taken multiple times at predetermined intervals over a predetermined period of at least six months , in the vicinity of the rail in the rail movable section, which is a section where an expansion joint is interposed and the rail expands and contracts due to temperature changes;
a step of calculating the amount of movement of the rail pad by performing image processing on the collected image data of the rail movable section, and setting a reference trajectory loop line in advance for each of the rail movable sections based on a trajectory line in which the calculated multiple amounts of movement of the rail pad during the predetermined period are plotted in chronological order on a coordinate system having temperature and amount of movement as coordinate axes;
A step of calculating the amount of movement of the track pad by processing image data of the vicinity of the rail captured by an imaging device mounted on the vehicle;
A step of determining in which of the rail movable sections a track pad included in an image to be determined is located;
determining whether the amount of movement calculated for a rail pad in the rail movable section is outside or inside the track loop line set for the rail pad, and outputting an abnormality when the amount of movement is outside the track loop line;
A method for detecting abnormality in the amount of rail pad protrusion, comprising: a first abnormal state is output when it is determined that the movement amount calculated for the rail pad in the rail movable section is outside the trajectory loop line set corresponding to the rail pad and exceeds a preset upper limit value of the protrusion amount of the rail pad;
The method for detecting abnormality in the amount of protrusion of a track pad according to claim 4 , characterized in that when it is determined that the amount of movement calculated for a track pad in the movable rail section is outside the trajectory loop line set for that track pad and does not exceed the upper limit value, a second abnormal state is output. The predetermined period is 9 months or more,
The locus loop line is
The method for detecting abnormality in the amount of protrusion of a rail pad according to claim 5, characterized in that the amount of movement of the rail pad during the specified period is a parallelogram circumscribed on a trajectory line that is expressed in time series on a coordinate system having temperature and amount of movement as coordinate axes.