JP3126288B2 - Method of measuring rail rail shape - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a railroad rail manufacturing line for measuring all the individual irregularities and the amount of bending in the longitudinal direction of the upper surface of a rail being conveyed, and for determining the conformity of the shape. About the method.

[0002]

2. Description of the Related Art In railroad rails, if the rails have irregularities on the upper surface of the rails, fluctuations in wheel load occur during running of the vehicle and the vehicle vibrates, which adversely affects ride comfort. The effect increases as the vehicle speed increases.The wavelength of each wave on the rail top surface is 2000 mm or more and the wave height is 0.2 mm or less, and the minute unevenness is 200 mm or more and the wave height 0.13 mm or less. It is required to suppress. If the end bends further,
Since straightening work occurs at the connection between the rails when laying the rails, it depends on the type of steel. The method of measuring the amount of unevenness in the length direction of the upper surface of the rail has conventionally been measured manually by a straight edge and a taper gauge. In addition, automatic measurement by a machine is disclosed in
Japanese Patent Application Laid-Open No. 196457/1992 discloses "Method and Apparatus for Measuring the Top of the Rail Rail Top". In this method, as shown in FIG. 7, the three distance meters 44a, 44b, and 44c are respectively set to L.
a / 2 intervals, and the uneven bend on the upper surface of the rail is S
Assuming that the in-wave and the period are λ, the three-dimensional rangefinder outputs ha, hb, and hc are used to calculate the uneven amount of curvature Δh in the rail top surface length direction.

[0003]

However, in the method disclosed in Japanese Patent Application Laid-Open No. 5-196457, the unevenness existing on the rail upper surface is assumed to be a sine wave. Can not be a sine wave. Therefore, there is a problem that the calculated amplitude and cycle lack reliability. In addition,
To measure the wavelength and amplitude for the entire length of the
Since it is not possible to evaluate individual irregularities existing on the rail upper surface, there is a problem that required quality assurance cannot be met.

The present invention has been made to solve such a problem, and has a function of calculating a wavelength and a wave height with respect to each of all the uneven waves, and further adding a function of calculating a bending amount of an end portion. It is intended to provide a method that has been implemented.

[0005]

The inventors [SUMMARY OF] is railroad Les - Le top height該Re - successively measured in the longitudinal direction of the Le, measured
From the constant data, set the profile data at the end of the rail to 0.
To perform parallel movement, and then the rail end
Fix the profile data and set the end reference length.
Move the profile data to 0 and move the data
- the maximum bending amount of positive relative to 0 axis data were calculated, from the result of that, Le - and performing quality acceptance judgment Le.

One embodiment is characterized in that the end reference length can be arbitrarily changed for each product type.

[0007]

The profile (measured value distribution) in the longitudinal direction of the rail upper surface is obtained by using two distance meters 1a and 1b as shown in FIG.
It is arranged at intervals L in the longitudinal direction x of the rail 2, and the height y of the upper surface of the rail 2 is measured at regular intervals and the measured value of a speedometer for measuring the rail transport speed is used. Can be measured. At this time, assuming that the position in the longitudinal direction is X and the profile is F (X) as shown in FIG. 2, since F (X) contains a lot of noise, moving average processing is performed to solve it. . As a result, a smooth concave-convex curve can be obtained, and preprocessing for the subsequent extremum extraction can be performed. Assuming that the profile data on which the moving average processing has been performed is G (X), a first derivative is applied to G (X) in order to extract an extreme value. Since the first derivative G (X) ′ always changes from plus to minus and minus to plus (including 0) in the x direction, the position where the sign is reversed becomes the extreme value of the unevenness. When the minimum value and the maximum value are determined from the extreme values and the respective values are set as shown in FIG. 3, the wavelength and the wave height of each unevenness can be calculated. By comparing the value with a preset reference value, it is possible to determine whether or not the rail is uneven. On the other hand, regarding the amount of bending at the rail end,
It is calculated from the moving average processed profile data G (X). As shown in FIG. 4, first, rail end data G (0)
Is applied to all the profile data so that is set to zero. Next, in order to make G (T) of the distance T in the longitudinal direction at which the bend is determined equal to 0, an equi-ratio moving process is performed in the vertical direction y based on G (0). By comparing the maximum bending amount of the data with a preset reference value, it is possible to determine the quality of the bending at the end of the rail. Further, the length T in the longitudinal direction in which the bending is determined differs for each type of rail, but since the present method can be set arbitrarily for each type, it is possible to determine whether or not the bending amount is appropriate for the type.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described with reference to the drawings. In FIG. 1, reference numerals 1a and 1b denote scanning distance meters used for measuring rail displacement, 2 denotes a transport rail, and 3a and 3b.
Is a transport roll, 4 is a distance meter controller, 5 is an A / D converter, 6 is an arithmetic unit, and 7 is a rotary encoder.
The two scanning rangefinders 1a and 1b are arranged at L intervals on the rail transport line (x-axis). The scanning rangefinders 1a and 1b measure the height (y direction) of the rail based on the light receiving width at the light receiving portion, which is reduced by the shielding of the rail with respect to the scan width of the projector. Each time the rotary encoder 7 generates a set number of pulses, that is, each time the rail 2 moves a predetermined amount,
The measurement signals (height signals) of the scanning rangefinders 1a and 1b are converted into digital data (height data) and given to the computer 6. The computer 6 sequentially writes the given height data into a memory (profile table).

The computer 6 calculates the amount of irregularities on the upper surface of the rail based on the height data of the memory, and determines the quality of the rail. This processing is shown in S2 to S6 in FIG.

[0010] Measurements by the above rangefinders 1a and 1b (FIG. 5)
According to S1), the memory (profile table) has a height distribution (rail profile) as shown in FIG.
) Is stored. The computer 6 performs a moving average process on these data to remove noise (S2). Obtained moving average processing data G
In order to extract an extremum from (X), G (X) is subjected to first-order differentiation (S3). The contents of the processing are as follows: G ′ (X) = G (X + 1) −G (X) (1) From X in which G ′ (X) changes from plus to minus and from minus to plus, the minimum and maximum values are extracted as shown in FIG. 3 (S4), and the wavelength and wave height of each unevenness are calculated (S5). At this time, if a perpendicular line is drawn down to a straight line connecting the local maximum value to the local minimum value and the coordinates of the intersection point are (X ", Y"), the calculation formula for obtaining it is:

[0011]

(Equation 2)

## EQU1 ## (X ″, calculated by equation (3)
Y ”), the formula for calculating the wavelength and wave height of the unevenness is

[0013]

(Equation 4)

By performing this for all the extreme values, the wavelength and the wave height of each unevenness are calculated. The calculation result is compared with a pass / fail judgment reference value to judge pass / fail of the shape of the rail (S6).

Next, the processing by the computer 6 for calculating the amount of bending at the end of the rail and determining whether or not the rail quality is acceptable will be described with reference to FIG.
The description will be made according to the flowchart shown in FIG. In the case of the bend amount determination, the profile data is subjected to a moving average process in the same manner as described above, so that G (0) becomes 0 as shown in FIG. G (X)
Is added to ΔG (0) (S7). The calculation formula is: H (X) = G (X) + ΔG (0) (6) Then, an equal ratio movement is performed in the vertical direction with reference to H (0) so that H (T) at the determination reference length T becomes 0 (S8). The calculation formula is as follows: I (X) = H (X)-[H (T) / T] .X (7) It should be noted that the rotation must be performed originally, but since the amount of bending with respect to the reference distance T is small,
There is practically no problem with this moving method. Then, the calculated I
By comparing (X) in the range of 0 to T, the maximum bending amount at the reference length is calculated (S9). This process is performed for both ends of the rail, and the quality of the rail is determined online or not by comparing the calculation result with a pass / fail judgment reference value (S10). The reference length T is calculated by the computer 6 for each steel type.
Is entered and set.

[0016]

According to the invention of the present application, the criterion is determined for each steel type.
Is easy to calculate the amount of end bending
In this way, highly reliable on-line bending inspection can be performed, and inspection of all and all types of steel can be easily performed without lowering the efficiency of the production line, thereby improving rail quality.

According to the above embodiment of the present invention, the end base
The quasi-length can be changed arbitrarily for each product type.
It is possible to make a pass / fail judgment of the bent amount.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an outline of a device configuration for implementing the present invention in one aspect.

FIG. 2 (a) is a graph showing an upper surface distribution of the rail 2 measured by the distance meters 1a and 1b shown in FIG. 1,
(B) is a graph showing a height distribution obtained by smoothing the height data of (a) by moving average processing.

FIG. 3 is a graph showing a part of the graph shown in FIG.
4 is a graph showing an enlarged view of the upper surface height axis (vertical axis).

FIG. 4A is a graph showing a height distribution obtained by smoothing height data of a rail end region by a moving average process, and FIG. 4B is a graph showing a height distribution on the graph shown in FIG. A graph obtained by translating the curve in the height direction, and (c) is a graph obtained by subjecting the curve shown in (b) to vertical equi-ratio movement processing using the rail end as a base point.

FIG. 5 is a flowchart showing details of a shape measurement process of the controller 4 and the computer 6 shown in FIG.

FIG. 6 shows a controller 4 and a computer 6 shown in FIG.
7 is a flowchart showing the contents of the setting process for determining the amount of bending at the end of the rail.

FIG. 7 is a graph showing the relationship between the arrangement of measuring instruments and measured values in the conventional measurement of the amount of bending of the upper surface of the rail head.

[Explanation of symbols]

1a, 1b: Scanning laser range finder 2: Rail 3a, 3b: Carrier roll 4: Laser range finder controller 5: A / D converter 6: Computer 7: Rotary -Encoder D: Transport direction 44a, 44b, 44c: Laser distance meter ha, h
b, hc: Rangefinder output Δh: Bending amount A: Bending amplitude B: Offset λ: Bending period

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-5-196457 (JP, A) JP-A-59-30009 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01B 21/00-21/32 B21B 1/08 B21C 51/00 G01B 11/00-11/30 102

Claims (2)

(57) [Claims] 1. A method for sequentially measuring the height of a railroad rail upper surface in the longitudinal direction of the rail, and setting the profile data at the end of the rail to 0 based on the measured data. A parallel movement process is performed. Thereafter, the profile data at the end of the rail is fixed, and the profile data corresponding to the reference length at the end is moved to zero, and the moved data is A rail end bending amount measuring method, comprising calculating a maximum positive / negative bending amount with respect to the 0 axis and determining whether or not the rail quality is acceptable based on the result. 2. A method according to claim 1, wherein the record, characterized in that to change the end reference length arbitrarily by breed - le end bending amount measuring method.