JP2020012733A - Locus state evaluation method and evaluation device - Google Patents

Abstract

To provide an evaluation device or the like capable of determining a kind of a state change of a locus and detecting easily and inexpensively a state change of the locus, without being affected by change of travel speed of a railway vehicle.SOLUTION: A state evaluation device 100 comprises: an analysis part 10 into which travel speed of a railway vehicle 1 and vibration acceleration of a vehicle body 2 which are measured at each of a plurality of measurement times at a prescribed point on a locus 7 are input with the measurement time for performing multi-variable analysis; and an evaluation part 20 for evaluating the state change at the prescribed point of the locus. The analysis part extracts vibration acceleration components of a plurality of different frequency bands when detecting the state change in a prescribed point on the locus by the evaluation part, then, for every vibration acceleration component extracted, sets a second evaluation index expressed by the vibration acceleration component to an objective variable, then, sets input travel speed of the railway vehicle to a quantitative explanatory variable, and then sets at least the measurement time to a qualitative explanatory variable, for performing multivariable analysis. The evaluation part determines the kind of the state change, on the basis of the category score of the measurement time.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a method and an apparatus for evaluating a state of a track on which a railway vehicle travels. In particular, the present invention relates to an evaluation method and an evaluation apparatus capable of easily and inexpensively detecting a change in a track state without being affected by a change in a traveling speed of a railway vehicle and determining a type of the track state change. About.

  A track on which a railroad vehicle (hereinafter, simply referred to as a “vehicle”) travels repeatedly causes deterioration of a state such as a displacement of an installation position or a deformation of a tread surface due to abrasion. Therefore, it is necessary to understand the condition of the orbit and provide appropriate care.

  As a method of evaluating the state of the track, a method of measuring the displacement of the track using a dedicated track inspection vehicle, which is a non-business vehicle, is known. However, track inspection vehicles are expensive, so only some railway companies own them. In addition, the frequency of measurement is low, about several times a year, due to the limited number of holdings. For this reason, the state of the orbit has not always been grasped. In addition, there is no means for quantitatively evaluating the result of maintaining the track and the period during which the effect of the maintenance is maintained. Accordingly, at present, periodic maintenance (TBM: Time Based Maintenance) is performed instead of maintenance (CBM: Condition Based Maintenance) based on the evaluated track state as track maintenance.

  In order to solve the above-mentioned problems, a method has been proposed in which a sensing function is added to a normal business vehicle to evaluate the state of a track during business operation of the vehicle. For example, research is being conducted to apply a truck (PQ monitoring truck) capable of measuring wheel load and lateral pressure to an evaluation of a track condition by using the truck as a commercial vehicle. However, it is difficult to apply the PQ monitoring bogie to existing vehicles, and it is necessary to manufacture a new vehicle, so that the introduction cost increases. For this reason, there is a demand for a method that can more easily and inexpensively evaluate the state of the orbit.

  As a simpler and cheaper evaluation method, for example, a method described in Patent Document 1 has been proposed. The method described in Patent Literature 1 measures the vibration acceleration of the vehicle body to calculate the ride comfort level, and the ride comfort level at a specific point is calculated in advance with the average value and the standard deviation of the ride comfort level at the same point. Is determined to be degraded when the track state deviates from a predetermined range determined from the above (Claim 3, Paragraph 0023, 0024, etc. of Patent Document 1).

However, the method described in Patent Document 1 does not consider the traveling speed of the vehicle at all. Even at the same point on the track, the traveling speed of a commercial vehicle traveling at this point generally changes. If the traveling speed changes, the vibration acceleration of the vehicle body, and thus the riding comfort level, may change even if the state of the track does not change. In the method described in Patent Literature 1 that does not consider the traveling speed of the vehicle, the standard deviation of the riding comfort level at the same point obtained in advance includes a variation in the riding comfort level due to a change in the traveling speed of the vehicle. There is a risk. For this reason, the standard deviation increases, the range determined by the average value and the standard deviation also increases, and there is a possibility that a change in the orbital state cannot be detected unless a large change in the orbital state occurs.
Also, when the ride comfort level at a specific point deviates from a predetermined range, it is determined whether the cause is caused by deterioration of the track condition or a change in the traveling speed of the vehicle. It may not be possible.
Furthermore, in order to eliminate the influence of the change in the traveling speed of the vehicle, it is conceivable to keep the traveling speed of the vehicle constant when traveling at a specific point for evaluating the riding comfort level. It is not practical to impose constraints.

  In order to solve the above problems, the present inventors have proposed a method described in Patent Document 2. According to the method described in Patent Literature 2, it is possible to easily and inexpensively evaluate the state of a track without being affected by a change in traveling speed of a railway vehicle.

  However, although the method described in Patent Document 2 can detect a change in the state of the trajectory, it does not propose determining the type of the change in the state of the trajectory when the change occurs. Accordingly, in order to enable quick maintenance of the trajectory and the like, there is a demand for a method capable of not only detecting a change in the state of the track but also determining the type of the change in the state of the track.

JP-A-8-15098 International Publication No. WO 2017/159701

  SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems of the related art, and is capable of easily and inexpensively detecting a change in the state of a track without being affected by a change in a traveling speed of a railway vehicle. It is an object of the present invention to provide an evaluation method and an evaluation device capable of determining a type of a state change of an evaluation.

  In order to solve the above-mentioned problems, the present inventors have conducted intensive studies based on the results of various tests, as described in Patent Document 2, and determined the state of the track such as the ride comfort level caused only by the change in the measurement time. If the change of the evaluation index (the evaluation index expressed by using the vibration acceleration of the vehicle body or the like, the first evaluation index) related to the change can be extracted, the change of the evaluation index caused only by the change of the extracted measurement time is the change of the state of the track. I focused on what is considered to represent. Then, if multivariate analysis is performed using the evaluation index related to the state of the track such as the ride comfort level as the objective variable and the running speed and measurement time of the railway vehicle as explanatory variables, the effect of the change in the running speed on the change in the evaluation index And the effect of the change of the measurement time on the change of the evaluation index, and found that the change of the evaluation index caused only by the change of the measurement time can be extracted.

Furthermore, the present inventors have earnestly studied a method of determining the type of change in the state of the orbit, and as a result, have found the following items (1) to (3).
(1) Examples of the type of track state change include deterioration of the surface properties of the rails forming the track, loosening of the ballast of the track, displacement of the joints of the rails forming the track, and the like.
(2) Depending on the type of the track state change as described above, a difference occurs in the frequency band of the vibration acceleration generated in the vehicle body or the like due to the track state change. The frequency band is often a unique frequency band for each type of orbital state change.
(3) Accordingly, when a change in the state of the trajectory is detected, a plurality of vibration acceleration components in different frequency bands are extracted from the vibration acceleration, and an evaluation index ( If the second evaluation index is used, it can be expected that the type of the state change of the trajectory can be determined.

The present invention has been completed based on the above findings of the present inventors.
That is, in order to solve the problem, the present invention is a method for evaluating the state of a track on which a railway vehicle travels, and at a predetermined point on the track, the traveling speed of the railway vehicle and the traveling speed at a plurality of measurement times. A first procedure for measuring a vibration acceleration of a vehicle component that is any of a vehicle body, a bogie, and an axle box provided in a railway vehicle, and the vibration acceleration of the vehicle component measured in the first procedure is represented. A second procedure of performing a multivariate analysis using the first evaluation index as an objective variable, the traveling speed of the railway vehicle measured in the first procedure as a quantitative explanatory variable, and at least the measurement time as a qualitative explanatory variable; A third step of detecting a state change at the predetermined point on the trajectory based on the category score of the measurement time obtained by the multivariate analysis of the second step; and When a state change at the predetermined point is detected, vibration acceleration components of a plurality of different frequency bands are extracted from the vibration acceleration of the vehicle component measured in the first procedure, and the extracted different frequency bands are extracted. For each vibration acceleration component, a second evaluation index represented using the vibration acceleration component is used as an objective variable, and the traveling speed of the railway vehicle measured in the first procedure is used as a quantitative explanatory variable, and at least the measurement timing Based on a category score of the measurement timing obtained for each of the vibration acceleration components of the different frequency bands extracted by the multivariate analysis of the fourth procedure, using the multivariate analysis of the fourth step as a qualitative explanatory variable. And a fifth step of determining a type of a state change at the predetermined point on the trajectory.

According to the present invention, in the first procedure, the traveling speed of the railway vehicle and the vibration acceleration of the vehicle components (any one of the vehicle body, the bogie, and the axle box) are measured. The running speed is measured for brake control and the like even in a normal business vehicle. Further, the vibration acceleration is measured to perform the anti-sway control in a railway vehicle equipped with a so-called anti-sway control device. Therefore, the running speed of the railway vehicle and the vibration acceleration of the vehicle components can be measured simply and inexpensively without the necessity of newly manufacturing the vehicle.
The first evaluation index (for example, the riding comfort level) represented by using the vibration acceleration of the vehicle component is considered to be an evaluation index related to the state of the track. According to the present invention, in the second procedure, the first evaluation index related to the state of the trajectory is used as an objective variable, the traveling speed is used as a quantitative explanatory variable, and at least the measurement timing of the vibration acceleration and the traveling speed is qualitatively determined. Perform multivariate analysis as explanatory variables. Specifically, since the objective variable is a quantitative variable and the explanatory variable is both a quantitative variable and a qualitative variable, a multivariate analysis referred to as a so-called extended quantification class 1 is performed.
In general, it is known that the category score of a qualitative variable obtained by a multivariate analysis indicates the degree of contribution to an objective variable. Accordingly, the category score of the measurement time obtained by the multivariate analysis in the second procedure of the present invention indicates the contribution of the measurement time to the first evaluation index. In other words, the change in the first evaluation index caused only by the change in the measurement time can be extracted from the category score in the measurement time. Therefore, according to the present invention, in the third procedure, it is possible to detect a state change at a predetermined point on the trajectory based on the category score at the measurement time.
As described above, according to the present invention, it is possible to easily and inexpensively detect the state of a track without being affected by a change in the traveling speed of a railway vehicle.

Further, according to the present invention, when a state change at the predetermined point on the track is detected in the third procedure, the vibration acceleration of the vehicle component measured in the first procedure is changed to a plurality of different frequencies in the fourth procedure. A vibration acceleration component of a frequency band is extracted, and for each of the extracted vibration acceleration components of different frequency bands, a second evaluation index represented using the vibration acceleration component is set as an objective variable, and the same as in the case of the first evaluation index Perform multivariate analysis. For a plurality of different frequency bands from which the vibration acceleration component is extracted, a unique frequency band corresponding to the type of the state change of the trajectory to be determined may be set.
The category score of the measurement time obtained by the multivariate analysis in the fourth procedure of the present invention represents the contribution of the measurement time to the second evaluation index. In other words, a change in the second evaluation index caused only by a change in the measurement time can be extracted from the category score of the measurement time. In the fourth procedure, a multivariate analysis is performed for each vibration acceleration component in a different frequency band. Therefore, in the fifth procedure, a predetermined trajectory of the trajectory is determined based on the category score of the measurement timing obtained for each vibration acceleration component in the different frequency band. It is possible to determine the type of state change at the point of. Specifically, in the fifth procedure, for example, the magnitude of the category score at the measurement time corresponding to each frequency band is compared, and the type of the state change corresponding to the frequency band in which the largest category score is obtained is determined as the determination result. It is possible to
As described above, according to the present invention, it is possible to easily and inexpensively detect a change in the state of a track without being affected by a change in the traveling speed of a railway vehicle, and to determine the type of change in the state of a track. is there.

  The method described in Patent Literature 1 is based on the premise that all vehicle characteristics are equivalent (paragraph 0023 and the like in Patent Literature 1). However, even in the case of vehicles of the same type, there is a slight difference in characteristics between trains, and the vibration acceleration, and thus the first evaluation index and the second evaluation index, do not change even if the state of the track does not change. , May change depending on the knitting. For this reason, when the data used for the multivariate analysis includes the traveling speed and the vibration acceleration measured for a plurality of knittings, the change in the first evaluation index and the second evaluation index caused only by the change in the measurement time is accurately detected. To extract, it is preferable to perform a multivariate analysis that further includes the composition number as a qualitative explanatory variable.

  That is, preferably, in the first procedure, the traveling speed of the railway vehicle and the vibration acceleration of the vehicle components provided in the railway vehicle are measured for a plurality of formations, and the quality is determined in the second procedure and the fourth procedure. A multivariate analysis further including a composition number as a general explanatory variable is performed.

  According to the preferred method described above, even when the data used for the multivariate analysis in the second and fourth procedures includes the traveling speed and the vibration acceleration measured for a plurality of knitting, the first evaluation index and the second The influence of the change of the composition on the change of the evaluation index can be separated, and the change of the first evaluation index and the change of the second evaluation index caused only by the change of the measurement time can be accurately extracted. For this reason, it is possible to accurately evaluate the state of the trajectory (it is possible to accurately detect a change in the state of the trajectory and to accurately determine the type of change in the state of the trajectory).

  Further, even within the same train, the magnitude of the vibration acceleration, and thus the first evaluation index and the second evaluation index, may change depending on the position of the vehicle component whose vibration acceleration is measured. For example, in general, vehicle components located rearward in the traveling direction tend to have higher vibration acceleration than vehicle components located forward in the traveling direction of a railway vehicle. For this reason, in order to more accurately extract the changes in the first evaluation index and the second evaluation index caused only by the change in the measurement time, the position of the vehicle component whose vibration acceleration was measured as a qualitative explanatory variable must be determined. It is preferable to perform a multivariate analysis including further.

  That is, preferably, in the second procedure and the fourth procedure, a multivariate analysis further including a position of the vehicle component in the train in which the vibration acceleration is measured as a qualitative explanatory variable is performed.

According to the preferred method described above, it is possible to separate the influence of the change in the position of the vehicle component in the train in which the vibration acceleration is measured with respect to the change in the first evaluation index and the second evaluation index. The changes in the first evaluation index and the second evaluation index thus obtained can be more accurately extracted. For this reason, the state of the orbit can be more accurately evaluated.
In the preferred method described above, the car number can be exemplified as the qualitative explanatory variable "the position of the vehicle component in the train at which the vibration acceleration was measured". Further, a combination of the car number and the front or rear position of the vehicle body (for example, the front position of the third car, etc.) can be exemplified.

As described above, the “first evaluation index represented by using the vibration acceleration of the vehicle component” in the present invention can include the ride comfort level calculated using the vibration acceleration as described above. Other vehicle components related to the state of the track, such as the root mean square (RMS) of the vibration acceleration, the maximum value of the vibration acceleration, the amplitude of the vibration acceleration, and the number of occurrences of the vibration acceleration having an amplitude equal to or greater than a predetermined value. There is no particular limitation on the first evaluation index of the present invention as long as the evaluation index is expressed using the vibration acceleration of.
Further, the “second evaluation index represented using the vibration acceleration component” in the present invention can be exemplified by the root mean square (RMS) of the vibration acceleration component. In addition, the ride comfort level calculated using the vibration acceleration component, the maximum value of the vibration acceleration component, the magnitude of the amplitude of the vibration acceleration component, the number of occurrences of the vibration acceleration component having an amplitude equal to or larger than a predetermined value, and the The second evaluation index of the present invention is not particularly limited as long as it is an evaluation index that is expressed using a vibration acceleration component of a vehicle component related to the state of the track, such as a power spectrum density.
Although there is a difference whether the parameter used for calculation is a vibration acceleration or an extracted vibration acceleration component, the same evaluation index is used as the first evaluation index and the second evaluation index (for example, both are used) It is possible to use a ride comfort level or RMS, or to use a different evaluation index (for example, use the ride comfort level as the first evaluation index and use the RMS as the second evaluation index).

  According to another aspect of the present invention, there is provided an apparatus for evaluating a state of a track on which a railway vehicle travels, the traveling speed of the railway vehicle being measured at a plurality of measurement times at a predetermined point on the track. And a vibration acceleration of a vehicle component that is any one of a vehicle body, a bogie, and an axle box included in the railway vehicle is input together with a measurement time, and a first acceleration is represented using the input vibration acceleration of the vehicle component. An analysis unit that performs a multivariate analysis using the evaluation index as a target variable, the input traveling speed of the railway vehicle as a quantitative explanatory variable, and at least the measurement time as a qualitative explanatory variable, An evaluation unit that detects a state change at the predetermined point on the trajectory based on the category score of the measurement time obtained by the variable analysis, and the analysis unit includes: When a state change at the predetermined point on the track is detected, a plurality of vibration acceleration components in different frequency bands are extracted for the input vibration acceleration of the vehicle component, and the extracted vibration acceleration components are different from each other. For each vibration acceleration component in the frequency band, a second evaluation index represented using the vibration acceleration component is used as an objective variable, the input traveling speed of the railway vehicle is used as a quantitative explanatory variable, and at least the measurement timing is used. Perform a multivariate analysis as a qualitative explanatory variable, the evaluation unit, based on the category score of the measurement time obtained for each vibration acceleration component of the different frequency band extracted by the multivariate analysis, based on the The present invention is also provided as an orbital state evaluation device for determining a type of a state change at the predetermined point.

  Advantageous Effects of Invention According to the present invention, it is possible to easily and inexpensively detect a change in track state and to determine the type of track state change without being affected by a change in traveling speed of a railway vehicle.

It is a mimetic diagram showing the schematic structure of the track condition evaluation device concerning one embodiment of the present invention. It is a figure which shows an example of the relationship between the driving | running | working speed of a railroad vehicle (average driving | running speed in an area), and a riding comfort level (average driving in an area). FIG. 3 is a diagram in which the ride comfort levels shown in FIG. 2 are arranged in chronological order according to the traveling speed of the vehicle and the measurement time (year and month) of the vibration acceleration of the vehicle body. Using the same data as shown in FIG. 2, multivariate analysis is performed in the second procedure of the evaluation method according to one embodiment of the present invention, and the measurement time (year and month) is plotted on the horizontal axis and the measurement time is plotted on the vertical axis. It is the figure which plotted the category score. 7 shows an example of a power spectrum of a vibration acceleration of a vehicle body that differs according to a type of a track state change. It is the figure which performed multivariate analysis in the 4th procedure of the evaluation method concerning one embodiment of the present invention, and plotted the measurement time (year and month) on the horizontal axis and the category score β3 (n3) on the vertical axis.

  Hereinafter, a track state evaluation device and an evaluation method according to an embodiment of the present invention will be described with reference to the accompanying drawings as appropriate. In the present embodiment, a case where the vehicle component whose vibration acceleration is to be measured is a vehicle body will be described as an example. However, the present invention can be similarly implemented when the vehicle component is a bogie or an axle box.

FIG. 1 is a schematic diagram illustrating a schematic configuration of a track state evaluation device according to an embodiment of the present invention. As shown in FIG. 1, a track state evaluation device (hereinafter, simply referred to as “evaluation device”) 100 according to the present embodiment evaluates a state of a track 7 on which a railroad vehicle 1 travels in a direction of an arrow. The apparatus includes an analysis unit 10 and an evaluation unit 20. The evaluation device 100 according to the present embodiment includes, for example, a personal computer in which a program having a function as the analysis unit 10 and the evaluation unit 20 (a program corresponding to processing contents executed by the analysis unit 10 and the evaluation unit 20 described later) is installed. Consists of a computer. Further, the evaluation device 100 according to the present embodiment is installed not in the railcar 1 but in another place.
The track state evaluation method according to the present embodiment (hereinafter, simply referred to as “evaluation method” as appropriate) includes first to fifth procedures, and is executed using the evaluation device 100 and a later-described anti-sway control device. . Specifically, the first procedure is executed by the vibration prevention control device, the second procedure and the fourth procedure are executed by the analysis unit 10 of the evaluation apparatus 100, and the third procedure and the fifth procedure are executed by the evaluation unit 20 of the evaluation apparatus 100. The procedure is performed.

<First procedure>
In the first procedure of the evaluation method according to the present embodiment, the traveling speed of the railway vehicle 1 and the vibration acceleration of the vehicle body 2 provided in the railway vehicle 1 are measured at predetermined points on the track 7 at a plurality of measurement times.
The railway vehicle 1 to which the evaluation device 100 according to the present embodiment is applied includes a known anti-sway control device. The anti-oscillation control device includes actuators 4A and 4B that are provided before and after the vehicle body 2, respectively, and that press the vehicle body 2 against the carts 3A and 3B in the left-right direction (the direction perpendicular to the plane of FIG. 1). In addition, the anti-sway control device includes acceleration sensors 5A and 5B that are provided before and after the vehicle body 2 (left and right in FIG. 1) and measure the vibration acceleration of the vehicle body 2 in the left-right direction. Further, the vibration prevention control device includes a controller 6 that controls the actuators 4A and 4B based on the vibration acceleration measured by the input acceleration sensors 5A and 5B. Accordingly, the controller 6 receives the front vibration acceleration of the vehicle body 2 measured by the acceleration sensor 5A and the rear vibration acceleration measured by the acceleration sensor 5B.
Further, the traveling speed of the railway vehicle 1 detected by a device (not shown) provided on the vehicle body 2 for brake control or the like of the railway vehicle 1 is also input to the controller 6 of the present embodiment.
Further, the controller 6 of the present embodiment accumulates the input traveling speed of the railway vehicle 1 and corrects the accumulated value by receiving a station signal indicating that the railway vehicle 1 has arrived at the station. , The position (distance) of the railway vehicle 1 is calculated. Alternatively, the controller 6 of the present embodiment is configured to calculate the position of the railway vehicle 1 by receiving a GPS (Global Positioning System) signal from the outside. In other words, the controller 6 of the present embodiment is configured to calculate the position of the point on the trajectory 7 at which the vibration acceleration has been measured.
As described above, the first procedure of the evaluation method according to the present embodiment is executed by the vibration prevention control device.

<Second procedure>
In the second procedure of the evaluation method according to the present embodiment, the traveling speed of the railway vehicle 1 measured in the first procedure is set as a target variable using a first evaluation index expressed using the vibration acceleration of the vehicle body 2 measured in the first procedure. Is a quantitative explanatory variable, and multivariate analysis is performed using at least the measurement time as a qualitative explanatory variable.
The analysis unit 10 included in the evaluation device 100 according to the present embodiment includes the traveling speed of the railway vehicle 1 measured at a plurality of measurement times (for example, the time from the latest one year) at a predetermined point on the track 7 and the railway The lateral acceleration of the vehicle body 2 of the vehicle 1 is input together with the measurement time. In the present embodiment, the case where the vibration acceleration in the left-right direction is measured / input is exemplified. However, the present invention is not limited to this. For example, the acceleration sensors 5A and 5B measure the vibration acceleration in the vertical direction of the vehicle body 2. If possible, it is also possible to input the measured vertical acceleration to the analysis unit 10.

  Specifically, for example, the traveling speed and vibration acceleration of the railway vehicle 1 input to the controller 6 are calculated based on the position of the railway vehicle 1 calculated by the controller 6 (that is, the position of the point on the track 7 at which the vibration acceleration was measured). ) And the measurement time (year and month). In the present embodiment, the traveling speed and the vibration acceleration of the railway vehicle 1 input to the controller 6 are based on the formation number of the formation to which the railway vehicle 1 belongs and the position of the vehicle body 2 where the vibration acceleration is measured (the car number of the vehicle body 2). And the acceleration sensors 5A and 5B (that is, the distinction between the front position and the rear position of the vehicle body 2). Then, these data sequentially stored in the controller 6 over a plurality of measurement times are output to an external storage medium such as a CF card at an appropriate timing. The external storage medium to which the data has been output is manually carried to the place where the evaluation device is installed, and the stored data is input to the analysis unit 10 via the external storage medium. However, the present invention is not limited to this, and the controller 6 and the analyzer 10 are connected by a wireless network, and the data sequentially stored in the controller 6 is transmitted to the analyzer 10 using the wireless network. It is also possible to adopt a configuration that does the following.

  The above-described data is input to the analysis unit 10 of the present embodiment from all the railcars 1 constituting one formation running on the same track 7. Further, the above-mentioned data is input from a plurality of trains running on the same track 7 and having different train numbers (from all the railway vehicles 1 constituting each train).

The analysis unit 10 uses the first evaluation index expressed using the input vibration acceleration of the vehicle body 2 as a target variable, uses the input traveling speed of the railway vehicle 1 as a quantitative explanatory variable, and uses the input measurement time ( ) Is performed as a qualitative explanatory variable (extended quantification class 1). As a preferable configuration, the analysis unit 10 according to the present embodiment preferably includes, as qualitative explanatory variables, a knitting number and a position of the vehicle body 2 in the knitting whose vibration acceleration is measured (the car number of the vehicle 2 and the front and rear positions of the vehicle body 2). Multivariate analysis is further performed.
In this embodiment, the ride comfort level calculated using the equation 1 described in Patent Document 1 is used as the first evaluation index. Specifically, the analysis unit 10 of the present embodiment calculates the input vibration acceleration of the vehicle body 2 based on the input position of the track 7 (the position of the track 7 at which the vibration acceleration was measured). The average riding comfort level in each evaluation section is calculated for each evaluation section which is a unit for evaluating the state of No. 7, and this is used as a first evaluation index. Similarly, the analysis unit 10 of the present embodiment summarizes the input traveling speed of the railway vehicle 1 for each evaluation section for evaluating the state of the track 7 and calculates the average traveling speed in each evaluation section. Is a quantitative explanatory variable.

That is, the ride comfort level (average ride comfort level) for each evaluation section i for evaluating the state of the track 7 is yi, the constant is β0, the category score of the formation number is β1 (n1), and the knitting in which the vibration acceleration is measured is The category score of the position of the vehicle body 2 (the car number of the vehicle body 2 and the front position / rear position of the vehicle body 2) is β2 (n2), the category score of the measurement time (year and month) is β3 (n3), and each evaluation section i Assuming that the traveling speed (average traveling speed) is xi and the regression coefficient of the traveling speed xi is β4, the analysis unit 10 calculates the actual riding comfort level (the riding comfort level directly calculated from the vibration acceleration of the vehicle body 2), Β0, β1 (n1), β2 (n2), β3 (n3), and β4 are evaluated in such a manner that the sum of squares of the residual with the ride comfort level approximately calculated by the following equation (1) is minimized. For each section i It will be constant.
yi = β0 + β1 (n1) + β2 (n2) + β3 (n3) + β4 · xi (1)

In the above formula (1), n1 is an integer from 1 to s, and s means the number of knitting numbers included in the data used for the multivariate analysis. For example, if the data used for the multivariate analysis includes the traveling speed and the vibration acceleration associated with three knitting numbers, s = 3, and the category scores β1 (1) to β1 (3) is determined.
Further, in the above equation (1), n2 is an integer from 1 to t, and t is the position of the vehicle body 2 (the car number of the vehicle body 2 and the car number of the vehicle body 2) in the knitting measuring the vibration acceleration included in the data used for the multivariate analysis. Means the number of the front and rear positions of the vehicle body 2). For example, if the data used for the multivariate analysis includes the vibration accelerations (measured by the acceleration sensors 5A and 5B) at the front and rear positions of each of the first to sixth cars, t = 6 × 2 = 12, the category scores β2 (1) to β2 (12) corresponding to the combination of each car number and the front or rear position are determined.
Further, in the above equation (1), n3 is an integer from 1 to u, and u means the number of measurement times (years and months) of data used for multivariate analysis. For example, if the data used for the multivariate analysis includes the traveling speed and the vibration acceleration measured from January 2015 to December 2015, u = 12, and the category score corresponding to each measurement time β3 (1) to β3 (12) are determined.

  In the present embodiment, as a preferable configuration, the position of the vehicle body 2 in the knitting whose vibration acceleration is measured is represented by a combination of each car number and the front position or the rear position. Instead, it is also possible to simply express each car number. In this case, for example, even if the data used for the multivariate analysis includes the vibration acceleration measured at the front and rear positions of each of the first to sixth cars, t = 6 in equation (1), and each car number , The category scores β2 (1) to β2 (6) are determined.

Further, as a preferable configuration, the analysis unit 10 of the present embodiment performs a multivariate analysis in which all of the input measurement time, the knitting number, and the position of the vehicle body 2 in the knitting in which the vibration acceleration is measured are used as qualitative explanatory variables. However, the present invention is not limited to this. For example, if the difference in vibration acceleration due to the difference in the position of the vehicle body 2 (difference in car number, etc.) is negligible, multivariate analysis using only the input measurement time and composition number as qualitative explanatory variables It is also possible to do. In this case, the analysis unit 10 sets β0, β1 (n1) such that the sum of squares of the residuals between the actual ride comfort level and the ride comfort level yi approximately calculated by the following equation (2) is minimized. , Β3 (n3), and β4 are determined for each evaluation section i.
yi = β0 + β1 (n1) + β3 (n3) + β4 · xi (2)

Furthermore, for example, when all the data used for the multivariate analysis are the traveling speed and the vibration acceleration associated with the same composition number (the same composition), only the input measurement time is used as a qualitative explanatory variable. It is also possible to perform a variate analysis. In this case, the analysis unit 10 sets β0, β3 (n3) such that the sum of squares of the residuals between the actual ride comfort level and the ride comfort level yi approximately calculated by the following equation (3) is minimized. , Β4 are determined for each evaluation section i.
yi = β0 + β3 (n3) + β4 · xi (3)
As described above, the second procedure of the evaluation method according to the present embodiment is executed by the analysis unit 10.

<Third procedure>
In the third procedure of the evaluation method according to the present embodiment, a state change at a predetermined point on the trajectory 7 is detected based on the category score β3 (n3) at the measurement time obtained by the multivariate analysis in the second procedure.
Specifically, the evaluation unit 20 detects a state change in each evaluation section i of the trajectory 7 based on the category score β3 (n3) at the measurement time obtained by the multivariate analysis in the analysis unit 10. The category score β3 (n3) of the measurement time obtained by the multivariate analysis in the second procedure represents the contribution of the measurement time to the ride comfort level yi. In other words, the change in the riding comfort level yi caused only by the change in the measurement time can be extracted from the category score β3 (n3) of the measurement time. Accordingly, it is possible to detect a state change in each evaluation section i of the trajectory 7 based on the category score β3 (n3) at the measurement time.
As described above, the third procedure of the evaluation method according to the present embodiment is executed by the evaluation unit 20.

Hereinafter, the second and third procedures will be described more specifically.
The present inventors investigated the relationship between the traveling speed of the railway vehicle 1 and the riding comfort level. Specifically, the average traveling speed in the evaluation section of the railway vehicle 1 traveling in the predetermined evaluation section of the track 7 is measured over a predetermined period, and the average riding comfort level in the evaluation section is measured. A test was conducted. The evaluation section of the track 7 was cleaned during the test period. Further, it was confirmed that the evaluation section of the track 7 was deteriorated after the care was performed during the test period.

FIG. 2 is a diagram illustrating a relationship between the traveling speed of the railway vehicle 1 (average traveling speed in the evaluation section) and the riding comfort level (average riding comfort level in the evaluation section) measured by the test.
As shown in FIG. 2, it was found that as the present inventors assumed, as the traveling speed increased, the riding comfort level also tended to increase. Therefore, as described above, it is considered that the method described in Patent Literature 1 that does not consider the traveling speed of the railway vehicle 1 cannot detect a change in the track state unless a large change in the track state occurs.

FIG. 3 is a diagram in which the ride comfort levels shown in FIG. 2 are arranged in chronological order according to the traveling speed of the railway vehicle 1 and the measurement time (year and month) of the vibration acceleration of the vehicle body 2.
As can be seen from FIG. 3, it is difficult to detect a change in the track state (maintenance of the track 7, deterioration of the track 7) by a simple change in ride comfort level (time-series change). This is because, as shown in FIG. 2, the change in the riding comfort level is affected not only by the change in the state of the track 7 but also by the traveling speed of the railway vehicle 1. This is because it is not possible to extract only the state change of the track 7 if the data including the riding comfort level calculated in step (1) is used as it is.

FIG. 4 shows a multivariate analysis represented by the equation (1) in the second procedure of the evaluation method according to the present embodiment, using the same data as those shown in FIG. And a vertical axis plotting a category score β3 (n3).
As shown in FIG. 4, it can be seen that the category score β3 (n3) is a negative value having a large absolute value in the year and month when the evaluation section of the orbit 7 was cleaned. That is, it is possible to recognize that the value of the riding comfort level yi in the evaluation section of the track 7 has been reduced (the riding comfort has been improved) by performing care on the evaluation section of the track 7 in this date. For this reason, in the third procedure of the evaluation method according to the present embodiment, for example, if the category score β3 (n3) at the measurement time becomes a negative value equal to or less than a predetermined threshold value, the evaluation of the trajectory 7 is performed. It is possible to automatically evaluate that the state of the section is good. In other words, it is possible to automatically detect that the state of the track 7 has changed (improved). Such evaluation (detection) cannot be performed based on the results shown in FIG.

  Further, as shown in FIG. 4, it can be seen that the category score β3 (n3) is a positive value having a large absolute value in the date when the evaluation section of the trajectory 7 has deteriorated. That is, it is possible to recognize that the value of the riding comfort level yi in the evaluation section of the track 7 is increasing (the riding comfort is decreasing) due to the deterioration of the evaluation section of the track 7 in this date. For this reason, in the third procedure of the evaluation method according to the present embodiment, for example, if the category score β3 (n3) at the measurement time becomes a positive value equal to or more than a predetermined threshold value, the evaluation of the trajectory 7 is performed. It is possible to automatically evaluate that the state of the section has deteriorated. In other words, the change (deterioration) of the state of the track 7 can be automatically detected. Such an evaluation cannot be performed based on the results shown in FIG.

The result of the automatic evaluation (detection) as described above by the evaluation unit 20 or the category score β3 (n3) at the measurement time as shown in FIG. 4 is displayed on a monitor (not shown) provided in the evaluation device 100. ) Is displayed, and the result of visual observation by a human can be expected to be effectively used as follows.
(1) When the category score β3 (n3) continues to be a positive value equal to or greater than a predetermined threshold value (that is, when the state of the evaluation section of the track 7 is deteriorated and the riding comfort is continuously reduced). , Raise the priority of care for the evaluation section.
(2) By comparing the value of the category score β3 (n3) before and after the maintenance of the evaluation section of the track 7 (that is, by evaluating the improvement level of the riding comfort level before and after the maintenance), Quantify the effect of care.
(3) By evaluating the rate of change of the value of the category score β3 (n3) with time after the maintenance of the evaluation section of the track 7, the period until the riding comfort level becomes equal to that before the maintenance is predicted, and the prediction period In the meantime, review the plan so that care will be performed again in the evaluation section.

  As described above, according to the evaluation device 100 and the evaluation method according to the present embodiment, by executing the first to third procedures, without being affected by the change in the traveling speed of the railway vehicle 1, Further, it is possible to easily and inexpensively detect a change in the state of the track 7 without being affected by a difference in knitting or a difference in the position of the vehicle body 2 in which the vibration acceleration is measured.

<Fourth procedure>
The fourth procedure of the evaluation method according to the present embodiment is executed when a state change at a predetermined point on the trajectory 7 is detected in the third procedure. In the present embodiment, the fourth procedure is executed when it is detected that the state at a predetermined point on the track 7 has deteriorated. In the fourth procedure, a plurality of vibration acceleration components in different frequency bands are extracted from the vibration acceleration of the vehicle body 2 measured in the first procedure, and the vibration acceleration components are extracted for each of the extracted vibration acceleration components in the different frequency bands. The multivariate analysis is performed using the second evaluation index expressed as an objective variable, the traveling speed of the railway vehicle 1 measured in the first procedure as a quantitative explanatory variable, and at least the measurement time as a qualitative explanatory variable.

Specifically, the analysis unit 10 first extracts a plurality of vibration acceleration components in different frequency bands from the input vibration acceleration of the vehicle body 2.
FIG. 5 shows an example of the power spectrum of the vibration acceleration of the vehicle body 2 which differs depending on the type of the state change of the track 7. FIG. 5A shows an example of a power spectrum obtained when the ballast of the track 7 is loosened, and FIG. 5B shows a power spectrum obtained when the joint of the rails constituting the track 7 is displaced. 4 shows an example of a power spectrum. In FIG. 5, the power spectrum before the state change of the track 7 is indicated by a broken line, and the power spectrum after the state change of the track 7 is indicated by a solid line.
In the example shown in FIG. 5A, the power spectrum density of the vibration acceleration is increased in the frequency band of about 1.1 to 2.5 Hz due to the loosening of the ballast of the track 7. On the other hand, in the example shown in FIG. 5B, the power spectrum density of the vibration acceleration is increased in the frequency band of about 0.4 to 0.9 Hz due to the occurrence of the displacement of the joint of the rails constituting the track 7. . In other words, it can be said that the frequency band of the vibration acceleration generated in the vehicle body 2 due to the state change of the track 7 varies depending on the type of the state change of the track 7. According to the study by the present inventors, the frequency band of the vibration acceleration generated due to the state change of the track 7 is often a unique frequency band for each type of state change of the track 7. Accordingly, it can be expected that the type of the state change of the track 7 can be determined based on the difference between the frequency bands of the vibration acceleration.

The analysis unit 10 includes a plurality of filters (a band-pass filter, a high-pass filter, or a low-pass filter). The same number of filters as the number of types of state change of the trajectory 7 determined in the fifth procedure described later are provided, and each filter corresponds to a state change of the trajectory 7 to be determined. For example, the three types of state change of the track 7 to be determined are deterioration of the surface properties of the rails forming the track 7, looseness of the ballast of the track 7, and displacement of the seam of the rails forming the track 7. Then, three filters corresponding to each type are provided. The frequency bands (pass frequency bands) of the vibration acceleration passed by each filter are set to different frequency bands according to the type of state change of the track 7 to be determined. When the type of the state change of the track 7 to be determined is the above-mentioned three types, the filter corresponding to the deterioration of the surface texture of the rails constituting the track 7 includes, for example, a pass frequency having a center frequency of 5 Hz and a bandwidth of ± 1.0 Hz. The band is set. A pass frequency band having a center frequency of 1.8 Hz and a bandwidth of ± 0.7 Hz is set in the filter corresponding to the loose ballast of the track 7. Further, a pass frequency band having, for example, a center frequency of 0.6 Hz and a bandwidth of ± 0.2 Hz is set in the filter corresponding to the displacement of the joint of the rails constituting the track 7. As for the pass frequency band of each filter, the power spectrum of the vibration acceleration as shown in FIG. 5 is investigated in advance for each type of state change of the trajectory 7, and the frequency band where the power spectrum density greatly changes before and after the state change is specified. Then, it may be set in advance based on the specified frequency band.
When the vibration acceleration of the vehicle body 2 input to the analysis unit 10 passes through each filter included in the analysis unit 10, a plurality of vibration acceleration components in different frequency bands are extracted.

Note that the frequency band of the vibration acceleration generated due to the state change of the track 7 may be different depending on the point of the track 7 where the vibration acceleration is measured, even if the type of the state change of the track 7 is the same. is there. For example, even if the type of the state change of the track 7 is the same at the displacement of the rail joint, if the point of the track 7 is different such as a bridge, a railroad crossing, a curved section, etc., the frequency band of the generated vibration acceleration may be different. is there. For this reason, for example, a plurality of filters are provided corresponding to each point of the track 7 (for each evaluation section i), and even if the filters correspond to the same state change type, the pass frequency of the filter corresponding to each point is provided. Preferably, the bands can be set to different frequency bands.
Further, the type of the state change of the orbit 7 that is generated may be different depending on the location of the orbit 7. For this reason, it is preferable that the number of filters provided for each point on the track 7 can be changed.

Next, the analysis unit 10 is expressed by using the extracted vibration acceleration components for each of the extracted vibration acceleration components in the different frequency bands (for each vibration acceleration component that has passed through each of the plurality of filters included in the analysis unit 10). Multivariate analysis (extended type) including the second evaluation index as an objective variable, the input traveling speed of the railway vehicle 1 as a quantitative explanatory variable, and at least the input measurement time (year and month) as a qualitative explanatory variable Quantification 1) is performed. As a preferred configuration, the analysis unit 10 of the present embodiment preferably includes, as in the second procedure, the formation number and the position of the car body 2 (car number and car body number of the car body 2) in the formation where vibration acceleration is measured, as qualitative explanatory variables. Multivariate analysis is further performed.
In the present embodiment, the root mean square (RMS) of the vibration acceleration component is used as the second evaluation index. Specifically, the analysis unit 10 of the present embodiment converts the extracted vibration acceleration component into the state of the trajectory 7 based on the input position of the trajectory 7 (the position of the trajectory 7 at which the vibration acceleration was measured). Are calculated for each evaluation section i, which is a unit for evaluating, and the average RMS (mean value of root mean square) in each evaluation section i is calculated, and this is used as a second evaluation index. Similarly, the analysis unit 10 of the present embodiment summarizes the input traveling speeds of the railway vehicle 1 for each evaluation section i for evaluating the state of the track 7, and calculates the average traveling speed in each evaluation section i. , This is a quantitative explanatory variable.

That is, assuming that the RMS (average RMS) for each evaluation section i for evaluating the state of the track 7 is yi, the analysis unit 10 determines the actual RMS (extracted vibration acceleration of the vehicle body 2) as in the second procedure. The residual between the RMS calculated directly by the components) and the RMS approximately calculated by the above equation (1) (however, it can also be approximately calculated by the above equation (2) or (3)) Β0, β1 (n1), β2 (n2), β3 (n3), and β4 are determined for each evaluation section i so that the sum of the squares becomes minimum.
The multivariate analysis of the fourth procedure is performed for each extracted vibration acceleration component (for each vibration acceleration component that has passed through each of the plurality of filters). That is, β0, β1 (n1), β2 (n2), β3 (n3), and β4 are determined for each extracted vibration acceleration component.
As described above, the fourth procedure of the evaluation method according to the present embodiment is executed by the analysis unit 10.

<Fifth procedure>
In the fifth procedure of the evaluation method according to the present embodiment, the trajectory 7 is determined based on the category score β3 (n3) of the measurement timing obtained for each of the vibration acceleration components in the different frequency bands extracted by the multivariate analysis of the fourth procedure. The type of state change at a predetermined point is determined.
Specifically, based on the category score β3 (n3) of the measurement time obtained for each vibration acceleration component extracted by the multivariate analysis in the analysis unit 10, the evaluation unit 20 Determine the type of state change. The category score β3 (n3) of the measurement time obtained by the multivariate analysis in the fourth procedure represents the contribution of the measurement time to the second evaluation index (RMS) yi. In other words, the change of the second evaluation index (RMS) yi caused only by the change of the measurement time can be extracted from the category score β3 (n3) of the measurement time. For this reason, in the fifth procedure, it is possible to determine the type of the state change in each evaluation section i of the trajectory 7 based on the category score β3 (n3) of the measurement time obtained for each vibration acceleration component of a different frequency band. is there. Specifically, in the fifth procedure, for example, the magnitude of the category score β3 (n3) at the measurement time corresponding to each frequency band is compared, and the fifth category corresponds to the frequency band in which the largest category score β3 (n3) is obtained. The type of state change can be used as the determination result.
As described above, the second procedure of the evaluation method according to the present embodiment is executed by the evaluation unit 20.

FIG. 6 shows a multivariate analysis represented by Expression (1) in the fourth procedure of the evaluation method according to the present embodiment (where the left side of Expression (1) is the second evaluation index (RMS)), and the horizontal axis is FIG. 7 is a diagram in which the measurement time (year and month) is plotted, and the vertical axis represents the category score β3 (n3). FIG. 6A shows that the signal passes through a filter (a filter in which a pass frequency band having a center frequency of 0.6 Hz and a bandwidth of ± 0.2 Hz is set) corresponding to the displacement of the joint of the rails constituting the track 7. 5 shows an example of a result obtained by a multivariate analysis using the vibration acceleration component extracted in step (1). FIG. 6B is extracted by passing through a filter corresponding to the loosening of the ballast of the track 7 (a filter having a center frequency of 1.8 Hz and a bandwidth set to a pass frequency band of ± 0.7 Hz). 4 shows an example of a result obtained by a multivariate analysis using a vibration acceleration component.
In the example shown in FIG. 6, the category score β3 (n3) shown in FIG. 6A is shown in FIG. 6B in the date when the evaluation section of the trajectory 7 detected in the third procedure has deteriorated. It is larger than the category score β3 (n3). Accordingly, it is possible to determine that the type of deterioration of the evaluation section of the track 7 is a displacement of a joint of the rails constituting the track 7.

DESCRIPTION OF SYMBOLS 1 ... Railway vehicle 2 ... Body 3A, 3B ... Bogie 5A, 5B ... Acceleration sensor 10 ... Analysis part 20 ... Evaluation part 100 ... Evaluation device

Claims (6)

A method for evaluating a state of a track on which a railroad vehicle travels,
At a predetermined point on the track, a first procedure of measuring a traveling speed of the railway vehicle and a vibration acceleration of a vehicle component that is any one of a vehicle body, a bogie, and an axle box included in the railway vehicle at a plurality of measurement times. When,
A first evaluation index represented by using the vibration acceleration of the vehicle component measured in the first procedure as an objective variable, and a running speed of the railway vehicle measured in the first procedure as a quantitative explanatory variable, at least A second procedure of performing a multivariate analysis using the measurement time as a qualitative explanatory variable;
A third step of detecting a state change at the predetermined point on the trajectory based on the category score of the measurement time obtained by the multivariate analysis of the second step;
When detecting a state change at the predetermined point on the trajectory in the third procedure, a plurality of vibration acceleration components of different frequency bands are extracted for the vibration acceleration of the vehicle component measured in the first procedure, For each of the extracted vibration acceleration components in the different frequency bands, a second evaluation index represented using the vibration acceleration component is used as an objective variable, and the traveling speed of the railway vehicle measured in the first procedure is quantitatively described. A fourth step of performing a multivariate analysis as a variable and at least the measurement time as a qualitative explanatory variable;
A type of state change at the predetermined point on the trajectory is determined based on a category score of the measurement time obtained for each of the vibration acceleration components of the different frequency bands extracted by the multivariate analysis of the fourth procedure. A fifth procedure,
A trajectory state evaluation method comprising: In the first procedure, a running speed of the railway vehicle and a vibration acceleration of the vehicle component included in the railway vehicle are measured for a plurality of formations,
The method according to claim 1, wherein in the second step and the fourth step, a multivariate analysis further including a formation number as a qualitative explanatory variable is performed.   3. The multi-variable analysis further including a position of the vehicle component in the train in which the vibration acceleration is measured as a qualitative explanatory variable in the second and fourth procedures. 4. Orbit condition evaluation method.   The track state evaluation method according to claim 1, wherein the first evaluation index is a ride comfort level calculated using the vibration acceleration.   5. The method according to claim 1, wherein the second evaluation index is a root mean square of the vibration acceleration component. 6. An apparatus for evaluating a state of a track on which a railway vehicle travels,
The running speed of the railway vehicle measured at a plurality of measurement times at a predetermined point on the track and the vehicle body provided by the railway vehicle, the vibration acceleration of a vehicle component that is one of the bogie and the axle box is measured together with the measurement time. The first evaluation index, which is input and expressed using the input vibration acceleration of the vehicle component, is used as an objective variable, and the input traveling speed of the railway vehicle is used as a quantitative explanatory variable, and at least the measurement timing An analysis unit that performs a multivariate analysis using as a qualitative explanatory variable,
An evaluation unit that detects a state change at the predetermined point on the trajectory based on the category score of the measurement time obtained by the multivariate analysis in the analysis unit,
The analysis unit extracts a vibration acceleration component of a plurality of different frequency bands with respect to the input vibration acceleration of the vehicle component when a state change at the predetermined point on the track is detected by the evaluation unit. Then, for each of the extracted vibration acceleration components in the different frequency bands, a second evaluation index expressed using the vibration acceleration component is used as an objective variable, and the input traveling speed of the railway vehicle is a quantitative explanatory variable. And perform a multivariate analysis with at least the measurement time as a qualitative explanatory variable,
The evaluation unit determines a type of a state change at the predetermined point on the trajectory based on a category score of the measurement time obtained for each of the vibration acceleration components in the different frequency bands extracted by the multivariate analysis. Do
A track state evaluation device characterized by the above-mentioned.

Images