JP2008057979A - Railway vehicle monitor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To monitor accurately the state of a vehicle constituting member by dividing measured data fluctuation into one caused by an abnormality of the vehicle constituting member and one caused by difference of a measuring condition. <P>SOLUTION: This railway vehicle monitor for monitoring the state of the vehicle constituting member 100 by measuring devices 1a, 1b installed on the ground relative to a vehicle 10 traveling on a rail is provided with a processing means 4 for recording measured data by the measuring devices, determining a frequency distribution on the measured data when a fixed quantity of measured data are accumulated, determining a mean value of the measured data in a desired distribution domain, and using the mean value as the measured data of the vehicle constituting member. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a railway vehicle monitoring apparatus, and more particularly to a railway vehicle monitoring apparatus that monitors the status of vehicle components such as wheels in a vehicle traveling on a rail using a measuring device installed on the ground.

  Conventionally, as can be seen in the following Patent Document 1, the distance to a vehicle wheel traveling on a rail is measured by a non-contact distance sensor as a measuring device installed on the ground, and the wheel shape, particularly the wheel dimensions are determined from the measurement result. Calculation and monitoring of the wear situation of the wheel is performed.

  This measuring device is installed on a track through which vehicles pass regularly and frequently, such as an entry / exit line at a vehicle base. Thereby, the wheel dimension can be automatically measured by passing the vehicle through this apparatus. Therefore, it can be performed more frequently and efficiently than a method in which the measuring device is applied to the wheel and the dimensions are manually measured.

JP 2001-88503 A

  However, in the prior art of Patent Document 1 described above, since the wheel (vehicle component) and the measuring device are in a non-contact state, the wheel sway and ambient temperature due to factors such as the vehicle speed and the tilt of the carriage during measurement. If there is a difference in measurement conditions due to factors such as weather and weather, the measurement data varies around the actual wheel dimensions for each measurement, even though the wheels have the same dimensions.

  When there was a significant change in the measurement data, it was determined that an abnormality that could lead to an accident occurred in the vehicle component, and a detailed inspection at the vehicle base could result in no abnormality.

  The frequent occurrence of large changes in measurement data despite the fact that no abnormality has occurred in the vehicle constituent members not only wastes detailed inspection work but also reduces the reliability of the monitoring device.

  Note that non-contact monitoring is also performed by, for example, monitoring the wear state of the sliding plate by imaging the pantograph sliding plate with a camera.

  Therefore, an object of the present invention is to provide a railway vehicle monitoring apparatus capable of accurately monitoring the status of a vehicle component by dividing the variation of measurement data into that of the vehicle component itself and due to a difference in measurement conditions. There is.

  Another object of the present invention is to provide a railway vehicle monitoring device capable of improving the accuracy of monitoring vehicle components and supporting efficient and safe maintenance management of vehicle components. is there.

  In order to achieve the above object, a feature of the present invention is that in a railway vehicle monitoring apparatus that monitors the status of vehicle components in a vehicle traveling on a rail with a measuring apparatus installed on the ground, measurement data of the measuring apparatus is stored. When a certain amount of measurement data is recorded, a frequency distribution is obtained for the measurement data, an average value is obtained for the measurement data in a desired distribution region, and the average value is used as measurement data for the vehicle component. The processing means is provided.

  Further, the processing means determines whether or not the average value obtained for the measurement data in the desired distribution region is within the desired range. If the average value is outside the desired range, the measurement data within the desired distribution region is obtained. An individual average value is obtained for the measurement data group divided in the order of measurement, and each individual average value is used as measurement data for each vehicle component of the divided measurement data group.

  According to the present invention, a frequency distribution is obtained by statistical processing for a certain amount of accumulated measurement data, and measurement data that varies greatly due to differences in measurement conditions is obtained without using measurement data outside the desired distribution region. Therefore, the average value obtained for the measurement data in the desired distribution area accurately captures the actual dimensions of the vehicle component, and whether the vehicle component is abnormal according to the average value. It is possible to accurately determine whether or not.

  In addition, according to the present invention, a certain amount of measurement data is accumulated for each desired travel distance of the vehicle, so that it is possible to grasp the status of changes in vehicle components over time.

  Hereinafter, an embodiment of a railway vehicle monitoring device that monitors the wear state of wheels that are vehicle constituent members will be described with reference to the accompanying drawings.

  FIG. 1 shows a schematic view of a sensor installation state and a device configuration as each measuring device in a railway vehicle monitoring device for monitoring wheels according to an embodiment of the present invention.

In FIG. 1, a vehicle 10 has two wheels 100 (eight wheels of one vehicle) on the front, rear, left, and right for each vehicle, and travels on a rail 11 in the direction of arrow “a”. The first distance sensor 1a installed outside the rail 11 and the second distance sensor 1b installed inside the rail 11 for each rail, each distance sensor (measuring device) 1a, 1b to the wheel 100 Measure distance. Moreover, the speed of each wheel 100 is calculated by the speed detectors 5a and 5b from the time that each wheel 100 passes between the speed detectors 5a and 5b (measurement device) and the installation distance between the both speed detectors 5a and 5b.

  The rail 11 is fixed on the sleeper 12, and the distance sensors 1 a and 1 b and the speed detection units 5 a and 5 b are fixed on the sensor installation plate 13. The rail 11 is electrically insulated between the distance sensors 1a and 1b and the speed detectors 5a and 5b. The sleepers 12 and the sensor installation plates 13 are prevented from moving as much as possible on the ground by ballast or the like. When the rail 11 is fixed to the track slab, the distance sensors 1a and 1b and the speed detectors 5a and 5b may be appropriately fixed to the track slab.

  The distance data obtained by the distance sensors 1a and 1b and the passing time and speed of the wheel 100 obtained by the speed detectors 5a and 5b are transmitted to the converter 2. In the conversion unit 2, the time and speed of each wheel 100 obtained by the speed detection units 5a and 5b and the distance data of the analog signals sampled by the distance sensors 1a and 1b are A / D converted into digital signals.

  The storage unit 3 organizes and stores various data obtained from the conversion unit 2 in units of vehicles. Although not shown in FIG. 1, there is a vehicle number transmitter that signals and transmits a vehicle number (vehicle number) at the head of the vehicle 10, and a vehicle number signal that is transmitted by the vehicle number transmitter of the vehicle 10 on the ground. There is a car number receiver to receive.

  When the vehicle number receiver receives a vehicle number signal from the vehicle 10, the storage unit 3 organizes data for each vehicle 10 to be measured by the distance sensors 1a, 1b and the speed detection units 5a, 5b according to the vehicle number. Remember.

  That is, the train organization for each vehicle 10 and the number of wheels 100 are known, and each wheel 100 is sequentially moved to the head by the distance sensors 1a, 1b and the speed detectors 5a, 5b as the vehicle 10 travels on each rail 11. Since it is measured from the vehicle, the storage unit 3 automatically assigns wheel numbers in order of measurement from the first wheel 100 in accordance with the speed measurement for each wheel 100 in the speed detection units 5a and 5b, and automatically for each wheel 100. Memorize the measurement data.

  The processing unit (processing unit) 4 uses the data in the storage unit 3 to calculate wheel dimensions described later. The accumulation storage unit 6 accumulates the wheel dimensions obtained by the processing unit 4 in time series, and creates a database.

  The screen display unit 7 displays the wheel dimensions obtained by the processing unit 4, and a hard copy can be obtained in the output printing unit 8. Further, the processing unit 4 can perform statistical processing, which will be described later, using the wheel dimensions stored in the storage unit 6, display the statistical processing on the screen display unit 7, and obtain a hard copy in the output printing unit 8. It has become.

  Note that the storage unit 3 and the accumulation storage unit 6 may use the same storage device, and may be simplified by using separate storage areas.

  Next, the state of measuring the distance to the wheel 100 with the shape of the wheel 100 and the distance sensors 1a and 1b will be described with reference to FIG.

As shown in FIG. 2, the tread surface 101 of the wheel 100 is formed so that the outer diameter of the outer peripheral surface gradually increases from the outer portion of the wheel toward the inner portion. The flange 102 is provided integrally with an inner portion of the wheel. The inner surface of the wheel 100 is provided with a reference groove 103 (a shape cut perpendicularly on the outer peripheral side) having a right-angled cross section, and the diameter (reference groove diameter) Ws is determined by the reference. The distance in the radial direction from the reference groove 103 to the tread surface 101 is referred to as the tire thickness Wt of the wheel 100.

  The outer flange surface 104 on the outside of the wheel 100 is referred to as a flange outer surface, and the inner surface of the wheel 100 including the inner flange surface of the wheel 100 is referred to as a back gauge surface or a wheel inner surface 105.

  The first distance sensor 1a measures and outputs the distance L3 to the flange outer surface 104 of the wheel 100 in time series according to the sampling period. Further, the second distance sensor 1b measures and outputs the distance L4 to the wheel inner surface 105 in time series according to the sampling period in the same manner as the first distance sensor 1a.

  The distance sensors 1a and 1b are set at θ1 and θ2 so that the light emitted from the distance sensors 1a and 1b is not blocked by the rail 11.

The distances L1 and L2 are horizontal component distances orthogonal to the rails 11 of the distances L3 and L4. L0 is an installation distance between the first distance sensor 1a and the second distance sensor 1b fixed and installed on the sensor installation plate 13, d is a thickness of the flange 102 (flange thickness), and Fh. Is the height of the flange 102 (flange height), and Dh is the diameter of the wheel 100 (hereinafter referred to as the wheel diameter). From the installation angle θ1 of the first distance sensor 1a, the installation angle θ2 of the second distance sensor 1b, and the wheel speed detected by the speed detectors 5a and 5b, the thickness of the flange 102 of the wheel 100 is sequentially determined by Expressions 1 to 9 described later. d, the height Fh of the flange 102, and the wheel diameter Dh of the wheel 100 are calculated to obtain the shape of the wheel 100. Hg is the installation height from each distance sensor 1a, 1b to the tread surface 11a of the rail 11, and the installation height Hg of each distance sensor 1a, 1b has the same dimension, but may be different.

  FIG. 3 is a diagram showing the wheel diameter Dh, the reference groove diameter Ws, and the like of the wheel 100 from the flange outer surface 104 side of the wheel 100.

  The vehicle 10 travels in the direction of arrow a, rotates the wheel 100 in the direction of arrow b as shown in FIG. 3, passes between the speed detectors 5a and 5b, and further, the first distance sensor 1a and the second distance sensor 1b. Drive between. The first distance sensor 1a measures the distance L3 to the flange outer surface 104, and the second distance sensor 1b measures the distance L4 to the wheel inner surface 105.

At this time, as shown in FIG. 3A, the first distance sensor 1a is a distance between A and B of the flange outer surface 104 of the wheel 100, a distance between the outer surfaces B and C of the wheel 100, and a distance between C and D of the flange outer surface 104. L3 is measured. The output waveform from the first distance sensor 1a is a waveform as shown in FIG.

  In FIG. 3A, the portion of the flange 102 of the wheel 100 corresponds to a portion between AB and a portion between CD.

  The second distance sensor 1b measures a distance L4 to a portion between EF, FG, and GH on the wheel inner surface 105 shown in FIG. The portion of the flange 102 of the wheel 100 corresponds to a portion between EF and a portion between GH. The output waveform from the second distance sensor is a waveform as shown in FIG.

  The distance from the falling edge to the rising edge of the output waveform of the second distance sensor 1b is Lf, and the distance between two sawtooth waveforms appearing when crossing the reference groove 103 in the output waveform of the second distance sensor 1b ( The wheel reference groove crossing distance) is Ls.

  Next, calculation of the wheel shape (wheel dimension) from the measurement results of both distance sensors 1a and 1b in the processing unit 4 shown in FIG. 1 will be described.

  The output waveforms from the first distance sensor 1a and the second distance sensor 1b are arranged by the conversion unit 2 in units of vehicles. Similarly, the passing time and speed of the wheel 100 obtained by the speed detection units 5 a and 5 b are also transmitted to the conversion unit 2. The conversion unit 2 performs A-D conversion so as to be easily stored in the storage unit 3, and the storage unit 3 stores the measurement result of the knitting unit.

In the processing unit 4, the measurement data stored in the storage unit 3, the installation distance L0 between the first distance sensor 1a and the second distance sensor 1b, the installation angles θ1, θ2, and the installation height Hg of the distance sensors 1a, 2a ( 2), the flange thickness d, the flange height Fh, the wheel diameter Dh, and the like are calculated by the following formulas 1-9.

  That is, as shown in FIG. 2, both distance sensors 1a and 1b are installed on the sensor installation plate 13, and the installation distance L0, installation angles θ1 and θ2, and the installation height Hg are known. , 1b, the flange thickness d of the wheel 100 can be calculated by Formulas 1-3.

Further, the distance Lf from the fall of the waveform to the rise in the output waveform of the second distance sensor 1b shown in FIG. 3B is obtained from the elapsed time and the wheel speed, and the wheel reference groove crossing distance Ls is represented by the waveform (b). It is obtained from the elapsed time from the F point to the G point indicating the wheel reference groove 103 on the upper side and the wheel speed, the known reference groove diameter Ws of the wheel 100, the distance L4 measured by the second distance sensor 1b, the installation angle θ2, and its From the installation height Hg, the wheel diameter Dh and the wheel flange height Fh can be calculated by Formulas 4-9.

  In the following formula, Rf is a flange radius from the axial center of the wheel 100 to the outer periphery of the flange 104, and Wt is a tire thickness of the wheel 100.

  Next, a description will be given of the wheel size fluctuation calculated by the above formulas 1 to 9 and the monitoring performed by the processing unit 4.

  The measurement of the wheel dimensions in the wheel 100 is periodically performed based on the travel distance. FIG. 4 shows the distribution of wheel dimensions obtained by accumulating the wheel dimensions, with the abscissa representing the travel distance and the ordinate representing, for example, the tire thickness.

  When the area Z0 in which the change of the actual wheel dimension is negligible due to wear is enlarged as shown in FIG. 5, the dimension varies in this area Z0.

  This variation is a measurement error due to the influence of measurement conditions caused by the fluctuation of the vehicle 10 due to the vehicle speed and the inclination of the carriage, the ambient temperature, the weather, etc. for each measurement. When the wheel dimensions obtained from the measurement data are compared with the actual wheel dimensions, the measurement errors due to the above-mentioned influences are included.

  As means for solving this, it is necessary to eliminate the influence of the dynamic characteristics of the vehicle 10 and the influence of the environmental conditions. However, in order to realize this, this monitoring device is installed because of the problem of the installation location where this monitoring device is installed in a straight section where dynamic characteristics are unlikely to occur, and the need to install structures that prevent direct sunlight, wind and rain, etc. In doing so, there is a problem of conflicting with vehicle limits (not shown) and construction limits (not shown) depending on the location.

  Therefore, in the following, taking advantage of the high measurement frequency that is the feature of this monitoring device, the accumulated data is used to eliminate the influence of measurement errors and to grasp the dimensional variation that occurs in the wheels, and to accurately determine the wheels that are vehicle components. An explanation will be given of monitoring the wear state of the steel.

  As shown in an example in FIG. 6, the wheel dimensions of the wheel 100 that are periodically executed and calculated are stored and stored in the storage unit 6 as various data as a database 6a based on the travel distance.

  In the example of FIG. 6, the measurement is performed every mileage 2000 km, and the data about the tire thickness, for example, is accumulated as shown as the dimension (measurement data) A for the eight wheels existing in one vehicle. Show. The dimensions other than the tire thickness are accumulated in the same manner, but the illustration is omitted for simplification of description. The data for the travel distance of 0 km is a design value.

  The measured data in the Z0 area shown in FIG. 4 corresponds to the measured data in the range Z0 in which the actual wheel dimensions in FIG. 4 can be ignored due to wear, and a certain amount of measured data is accumulated in Z0 to Zm and Zn to. By the way, in the measurement data of the individual areas for which statistical processing is performed in the processing unit 4, each of the areas Z0 to Zm, Zn is set in advance according to the travel distance.

  The wheel wear situation monitoring process will be described in accordance with the flowcharts of FIGS.

  In step (hereinafter abbreviated as S) 1 in FIG. 7, as described with reference to FIG. 6 according to the traveling of the vehicle 10, measurement data is recorded sequentially from the first wheel 100 for each vehicle 10 according to the obtained vehicle number. It memorizes as database 6a. In addition, the dimensions calculated in Equations 1 to 9 are also calculated, and the calculated data is also stored sequentially. In addition, data obtained by calculation in the following steps is also stored in the storage unit 6 every time it is calculated.

  When the data storage is completed for each vehicle 10, the process proceeds to S2, and the processing unit 4 determines whether the vehicle 10 has traveled the specified distance, and whether the storage area corresponding to the area Z0 in FIG. 6 is filled with tire thickness data. Judge with. If the data is not satisfied, the process returns to S1, continues the measurement, and stores and records the data.

  If it is determined in S2 that the storage area corresponding to the area Z0 in FIG. 6 is filled with the tire thickness data, the process proceeds to S3, where the tire thickness data in the storage area corresponding to the area Z0 is extracted, and in S4 The average value is calculated.

  Thereafter, in S5, a deviation ΔX of the tire thickness (data) of all the wheels 100 in the storage area corresponding to the area Z0 from the average value of the tire thickness obtained in S4 is calculated, and in S6, the deviation ΔX is illustrated. As shown in FIG. 10, a histogram is created for the frequency distribution of the deviation ΔX.

  As shown in FIG. 4, the distribution of measurement data shows a tendency for the tire thickness to gradually decrease due to wear as the mileage increases, but the measurement data varies, and much of the adjacent data is vertically It has fluctuated.

  The measurement itself fails due to the influence of measurement conditions and the like, and an unnecessary component that causes the size calculation result to protrude is included. This projecting piece component is a component that affects the tendency in calculating an accurate wheel size and analyzing the wear tendency.

  In order to exclude this projecting piece component, in S7, in the histogram of FIG. 10, the data of the total r% (r: exclusion range setting value) of which the deviation ΔX is extremely large from the maximum side and the minimum side from the center of the histogram. Are excluded as abnormal data.

  Note that if the exclusion range set value r is excessively increased, even components that show fluctuations in actual dimensions may be removed. Therefore, the exclusion range set value r is set optimally by analyzing the frequency of measurement failure of the measuring instrument. .

  Then, an average value An is calculated using the tire thickness data left after the removal in S8. Therefore, by removing and averaging abnormal data having an extremely large deviation ΔX within a continuously measured range, a calculated value approximated to the actual size that can avoid compromising the grasp of the wear tendency is obtained. be able to.

  In other words, errors that occur in repeated measurement are evaluated with repeatability. The repeatability indicates that the lower the numerical value, the smaller the error and the more stable measurement is possible.

  The repeatability is a standard deviation σ which is a numerical value indicating the degree of dispersion of measurement data (statistical values), and the standard deviation σ is obtained by Equation 10. A is measurement data, An is an average value of measurement data, and n is the number of measurements.

  When the distribution is a normal distribution, if the standard deviation σ is doubled, the measured data sampled within the range is statistically included with a probability of 95.4%, and the repetition accuracy generally used is 2σ. Also in this embodiment, the evaluation is made with 2σ.

  For measurement data that is out of the range, the entire r% (r: exclusion range set value) data with an extremely large deviation ΔX is excluded as abnormal data, and the accumulated data is used by taking advantage of the high measurement frequency. It is possible to grasp the dimensional variation occurring in the wheel by eliminating the influence of the measurement error.

  In subsequent S9, it is determined whether the average value An obtained in S8 is obtained for the first time.

  Since the average value An in the area Z0 in FIG. 6 is the first basis, the process returns to S1, continues to store and record data, and repeats the processes up to S9.

  If it is determined in S9 that it is not the first average value An, the process proceeds to S10 in FIG. 8, and a difference ΔS between the average value An obtained this time in S8 and the average value An-1 obtained last time in S8 is calculated.

In S11, it is determined whether the difference ΔS is smaller than the desired value M. The value M is a threshold value for judging whether the wheel 100 is worn and the tire thickness is impaired.

  If the difference ΔS is smaller than the desired value M (ΔS <M), it means that the wheel 100 has not been worn, so the process proceeds to S12, and it is determined whether data recording (storage) is to be continued. If data recording (memory) is to be continued (no stop command is entered), the process returns to S1 in FIG. 7, data recording (memory) is continued, and the processes up to S12 described above are repeated. The tire thickness thus obtained is an average value sequentially from the area Z0 as shown in FIG. 11 (the horizontal display in FIG. 11 shows the average value An. The dots are the individual tire thicknesses excluding the projecting piece components. Data is shown.)

  When the above processing is repeated and it is determined in S11 for the region Zm in FIG. 6 that the difference ΔS is larger than the desired value M (ΔS> M), there is a possibility that the wheel 100 has progressed and the dimensions have changed. Assume that there is. In this case, the dimensional variation, that is, the wear tendency may be averaged by averaging even though the dimensional variation occurs in the wheel 100.

  Therefore, the process proceeds to S13, and the area to be averaged is divided into two and examined in detail to follow the wear tendency.

  However, when the area to be averaged is divided, the component (number of measurement data) in the area after the division is reduced. If averaging is performed in this area, the accuracy of the average is lowered. That is, with respect to the effect of improving the accuracy of the repetition accuracy with respect to the averaged value, if the accuracy when averaging the n pieces of measurement data having the repetition accuracy σ obtained by Equation 10 is δ, the accuracy is improved. δ can be obtained by Equation 11.

  The average accuracy δ is said to be improved as the numerical value is smaller, and according to Equation 11, it can be seen that the larger the number n of measurement data, the better.

  For this reason, when the area to be averaged is divided, a threshold value is introduced for the number of measurement data after division, and the minimum average number N is set as the threshold value. Thereby, the minimum measurement accuracy after the averaging process can be ensured.

  Although the number of divisions is two in S13, the number may be divided into an arbitrary number as long as the minimum average number N can be maintained.

  In S14, it is confirmed whether or not the minimum average number N can be maintained. When the number of data in each measurement data group divided in S14 is smaller than the minimum average number N, the process returns to S1 through S12 and waits for the number of data in each measurement data group to be larger than the minimum average number N.

  When the number of data in each measurement data group is larger than the minimum number N to be averaged, the process proceeds to S15 in FIG. 9, and the average value is calculated for the measurement data in each measurement data group in the same manner as S4 in FIG.

  Subsequently, in S16, the deviation ΔY of the measurement data of each measurement data group from the average value obtained in S15 is obtained in the same manner as S5 in FIG. 7, and the histogram of each deviation ΔY as shown in FIG. 10 is obtained in S17, similar to S6 in FIG. Create Then, in S18, the abnormal data is excluded from the histogram created in S17 as in S7 of FIG. 7, and in S19, the average value Bna, for the measurement data of each measurement data group remaining in the abnormal data exclusion as in S8 of FIG. Each Bnb is calculated.

  Subsequently, in S20, differences ΔTna and ΔTnb from the previous average value An-1 are calculated for each measurement data group as in S10 of FIG.

  Then, as in S11 of FIG. 8, it is determined in S21 whether the difference ΔTna from the previous average value An−1 in one measurement data group is smaller than the desired value M. The value M is a threshold value for determining whether or not the tire thickness is reduced due to wear of the wheel 100 as described above.

  When the difference ΔTna is larger than the desired value M, the process proceeds to S22, and an abnormal display indicating that the wear has increased more than ever is provided in one measurement data group.

  As this abnormality display, the average value in the database of FIG. 6 is displayed by color coding, the average value of the tire thickness of FIG. 11 is displayed by color coding, or the vertical line of the alternate long and short dash line indicating the division of the measurement data group in FIG. Wearing the display of the screen display unit 7 in FIG. 1 and the output of the output printing unit 8 has been increased more than ever by setting to display the vertical lines of the dotted lines indicating the lines and divisions by color coding. The area can be easily understood by displaying by color coding.

If the difference ΔTna is smaller than the desired value M, the process proceeds to S23 without proceeding to the abnormality display of S22, and the difference ΔTnb with respect to the previous average value An-1 is larger than the desired value M for the remaining measurement data group. Judge whether it is small.

  If the difference ΔTnb is larger than the desired value M, an abnormal display process is performed in S24 as in S22. If the difference ΔTnb is smaller than the desired value M, the process returns to S1 in FIG. 7 via S12 in FIG. Continue recording and memory.

  In S11, it is determined that the difference ΔS is larger than the desired value M (ΔS> M), the process proceeds to S13, the data division process is performed, and the process proceeds from S11 to S13 also in each area after the next area Zn. In such a case, it is considered that the wheel 100 is in a period of wheel replacement because the wear progresses so much that it is preferable to issue a warning to enter an early maintenance inspection.

  The above process is continued, and if the instruction to stop data recording (storage) is entered in S12 of FIG. 8, the process is terminated.

  In the prior art, measurement data was handled in the presence of variations due to fluctuations in measurement conditions, etc., even though there were no dimensional fluctuations due to wear on the actual wheels. By eliminating variations in measurement data due to fluctuations and improving accuracy by averaging, it is possible to obtain measurement data that approximates a stable actual dimension obtained by eliminating fluctuation factors on the measurement side. Furthermore, in order to minimize the loss of wear tendency due to averaging, by dividing the averaging area, the wear tendency can be accurately obtained, monitored with high-precision dimensions, and accurately applied to wheels with abnormal wear. Can deal with.

  Furthermore, by monitoring the mileage and wear tendency over a long period of time, it can be applied to grinding work to restore the wheel dimensions to the designed state, wheel replacement plan, and confirmation of wheel dimensions to be set in the safety device, etc. It becomes possible to do.

  Note that the projecting piece component, which is a measurement value that has failed due to the influence of the measurement conditions excluded in the statistical process, can be used by calculating the number of the components and reporting the failure of the distance sensor or the abnormality of the wheel. .

  In the above embodiment, measuring the shape of the wheel has been described. However, the present invention captures a pantograph found in Japanese Patent Application Laid-Open No. 2005-337714, Japanese Patent Application Laid-Open No. 2006-118900, etc. with a camera, and performs a slab by image processing. It can also be carried out in monitoring the wear state of the.

1 is a schematic diagram of a railway vehicle monitoring apparatus according to an embodiment of the present invention. It is a figure explaining the condition which measures the distance to a wheel with the distance sensor of the railway vehicle monitoring apparatus shown in FIG. It is the figure which showed the wheel from the flange outer surface side. It is the figure which distributed and showed the tire thickness of the wheel on the basis of the travel distance. It is the figure which expanded and showed the area | region where the change by the dimension of the real wheel in FIG. 4 can be disregarded. It is the figure which showed an example of the database in the storage part of the railway vehicle monitoring apparatus shown in FIG. It is a flowchart of the monitoring process of the wheel wear condition by the rail vehicle monitoring apparatus shown in FIG. It is a flowchart of the monitoring process of the wheel wear condition by the rail vehicle monitoring apparatus shown in FIG. It is a flowchart of the monitoring process of the wheel wear condition by the rail vehicle monitoring apparatus shown in FIG. It is the figure which showed the frequency distribution of deviation with the average value about the tire thickness of all the wheels obtained by measurement by the railcar monitoring apparatus shown in FIG. 1 as a histogram. It is the figure which showed what remained after removing the abnormal data in the tire thickness distribution of the wheel in FIG. 4 with the average value.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1a ... 1st distance sensor (measurement apparatus), 1b ... 2nd distance sensor (measurement apparatus), 3 ... Memory | storage part, 4 ... Processing part (processing means), 5a, 5b ... Speed detection part, 6 ... Accumulation memory | storage part, 7 ... Screen display unit, 8 ... Output printing unit, 10 ... Vehicle, 11 ... Rail, 100 ... Wheel.

Claims (4)

In a railway vehicle monitoring device that monitors the status of vehicle components in a vehicle traveling on a rail with a measuring device installed on the ground,
When measurement data of the measurement device is recorded and a certain amount of measurement data is accumulated, a frequency distribution is obtained for the measurement data, an average value is obtained for the measurement data in a desired distribution region, and the vehicle is obtained using the average value. A railway vehicle monitoring device comprising processing means for measuring data of a constituent member.   In the railway vehicle monitoring apparatus according to claim 1, the processing means determines whether or not the average value obtained for the measurement data in the desired distribution region is within a desired range, and if it is outside the desired range, The measurement data in the desired distribution region is divided in the order of measurement, and individual average values are obtained for the divided measurement data groups, and the measurement data of each vehicle component of the measurement data group obtained by dividing each individual average value and A railway vehicle monitoring device characterized by that.   The railway vehicle monitoring device according to claim 1, wherein the measurement device measures the distance to the outer flange surface of the wheel, which is a vehicle constituent member, installed on the ground outside the rail, and outputs the measurement result. A first distance sensor and a second distance sensor that is installed on the ground inside the rail and that measures the distance to the back surface inside the wheel, which is a vehicle component, in a non-contact manner and outputs a measurement result; The processing means obtains the wheel shape data by calculation from the respective measurement results of the first and second distance sensors and the distance data related to the installation of both distance sensors, and the wheel shape data obtained by the calculation. A railway vehicle monitoring device characterized in that a frequency distribution for measurement data is obtained. 2. The railway vehicle monitoring apparatus according to claim 1, wherein the processing means accumulates a predetermined amount of measurement data for each desired travel distance of the vehicle. .

Images