RU2652673C1 - Method of identification of switch points and position of rail tongue - Google Patents

Abstract

FIELD: transportation.

SUBSTANCE: invention relates to railway transport, namely to methods and devices for identification of railroad track elements, in particular switch points, and can be used in computerized NDT-cars and track-measuring diagnostic wagons, motor vehicles performing non-destructive testing and track diagnostics using available mobile means. Method of identification of switches and position of switch points is that flaw-defective device are moved along the track, probing it, results of probing are saved in defectogram, evaluated and actual position of switch points and current movement direction of flaw-defective device are determined. When probing, rail joints are detected, when estimating relative distances between joints are calculated, obtained values are compared with previously known matrix of distance patterns between joints, switch type is determined, as well as actual position of switches and current movement direction of flaw-defective device.

EFFECT: as a result, it becomes possible to reliably identify type of turnout, position of switch points and current movement direction of the diagnostic tool by indirect signs (rail joints).

1 cl, 3 dwg, 3 tbl

Description

The method for determining the turnouts and the position of the turnouts wagons relates to railway transport, and in particular to methods and devices for identifying elements of a railway track, in particular turnouts, and can be used in computerized defectoscopic and track-measuring diagnostic cars, car trips, etc., performing non-destructive testing and diagnostics of the rail track using existing mobile tools.

A known mechanical method for determining the position of wits of turnouts [1], which consists in moving the flaw detector along the rail track, probing it, measuring the vibration accelerations created in the flaw detector when it is moving along the rail switch, saving the sounding results in the defectogram, evaluating them and determine the actual position of the wits of the switch and the current direction of movement of the flaw detector.

The disadvantage of this method is its complexity, associated with the need to install means of autonomous navigation measurements and their processing, to assess the speed of lateral movement. With the direct movement of the vehicle, this method does not detect the turnout.

A known optical method for determining the position of switch wagons [2], which consists in moving the flaw detector along the track, probing it with video surveillance tools, storing the sounding results in the defectogram, evaluating them and determining the actual position of the switch wagons and the current direction of movement of the flaw detector , based on a comparison of video images of turnouts with their known types.

The disadvantage of this method is associated with the difficulties of obtaining such images in different conditions (snow, darkness, etc.) and the complexity of comparing video images.

Closest to the claimed one is a method for determining the position of switch points wagons, which means that the flaw detector is moved along the rail track, probing it, the sounding results are stored in the defectogram, they are evaluated and the actual position of the switch points is determined and the current direction of movement of the flaw detector. In this case, eddy current (VT) [3], [4], ultrasonic (ultrasound) [5, p. 241-271] or magnetodynamic (MD) [5 p. 251] flaw detectors can be used.

In any case, the methods [3-5] are based on an attempt to directly detect the elements of the rail translation with a flaw detector. The disadvantages of the known methods [3-5] is the dependence of their results on many factors (BT methods - on the clearance and contamination of the rail surface, ultrasound - on acoustic contact). In addition, the signals from the objects of the rail track (joints, gaps) in the known methods are considered separately from each other without regard to their relative position. The issues of determining the direction of movement of the diagnostic tool for fixed flaw signals in the known methods are not considered.

The problem solved by the claimed method is the detection and determination of the turnout and the position of the rail wag by the indirect method - according to the positions of the rail joints.

To solve this problem, in the method for determining the turnouts and the position of the switches of the turnouts, which consists in moving the flaw detector along the rail track, probing it, save the sounding results in the defectogram, evaluate them and determine the actual position of the switches of the turn signal and the current direction of movement of the flaw detector means, when probing, they detect rail joints, when evaluating, calculate the relative distances between the joints, compare the obtained values The lines with a previously known matrix of distance patterns between the joints determine the type of turnout switch, as well as the actual position of the switch wits and the current direction of movement of the flaw detector.

The technical result of using the proposed method is to increase the reliability (reliability) of determining the position of the wits of railroad switches.

The invention is illustrated by graphic materials.

FIG. 1 - the position of the rail joints at the turnout switch of project 2768 (more precisely, “Turnout switch of type P65 grade 1/11 of project 2768.00.000” according to [7. Product catalog. Translations for trunk tracks. Murom turnout plant. 2013].

FIG. 2 is a schematic diagram of the turnout project 2768 and defectograms of each thread of the path, measured by MD, and below the ultrasound channels of the flaw detector complex. The joints of the rail track on the turnout diagram on the defectogram are marked with the same numbers.

FIG. 3 magnetic flaw detector on rails, where

1. Rail.

2. Wheels of the flaw detector.

3. The magnetic field excitation coils in the rails mounted on the axles of the wheelsets.

4. The magnetic field.

5. The magnetic field sensor.

6. Processing unit.

7. Rail linings of the rail joint.

8. The odometer.

There are various ways of navigating rail vehicles [6]: including using global navigation systems GPS and GLONAS, odometers, etc. None of these methods can be considered satisfactory because of an obvious contradiction: large displacement distances (tens to hundreds of kilometers) and high accuracy requirements for detecting defects (centimeters or less). As a result, there is a need for periodic binding of vehicle coordinates to the structural elements of the rail: picket posts, lighthouses, rail joints, turnouts, etc. In addition, on turnouts it is necessary to track the direction of movement of the vehicle. With apparent simplicity, the task is not obvious. For example, the accelerometer does not detect a turnout when moving in a straight branch. The methods [3-5] are aimed at solving these problems, however, they are distinguished by low accuracy associated with the difficulty of accurately detecting the turnout elements by the indicated flaw detection tools.

Elements of the upper structure of the rail track are made of ferromagnetic materials (primarily rail steel) and have joints with bolted joints or welded. These joints with different reliability are detected by VT, ultrasonic and MD flaw detection tools. This allows you to determine the coordinates of the above path elements on a defectogram in manual or automatic mode. The accuracy of determining the coordinates is determined primarily by the characteristics of the odometer installed on the flaw detector car and is about 10 cm.

Each railroad switch has certain geometrical dimensions described in standard designs [7] http://murom-switch.ru/page/10 Product catalog of the Murom switch plant. In particular, the projects describe the distance between rail joints. An analysis of the projects shows that most of them involve the implementation of one of three schemes for placing rail segments in a railroad switch. Scheme 1: projects 2750, 2830, 2882, 2751; scheme 2: projects 2768, 2764, 2759, 2802, 2851; Scheme 3: Projects 2769, 2766. These three schemes cover more than 98% of all turnouts used on the Russian Railways road network. In some rare cases, several more typical schemes can be added. But this does not change the essence of the proposed method, it simply increases the number of templates - matrices with which the sequence of diagnostic signals is compared.

Taking into account the fact that the car can move relative to the same turnout in one of four ways (woolly on the left, woolly on the right, woolly on the left, woolly on the right), defectograms may contain 3 × 4 = 12 typical images when the arrows pass.

Detected rail joints are checked for compliance with the expected images of turnouts. To do this, take a set of adjacent butt elements falling into the zone, the length of which is determined by the expected length of the turnout, and calculate the cross-correlation coefficient K of the set of coordinates of the detected joints with 12 coordinates of typical arrow patterns:

Where

- координаты обнаруженных на дефектограмме стыков;

- coordinates of joints detected on the defectogram;

- координаты, взятые из шаблона;

- coordinates taken from the template;

T - tolerance coefficient, in this example, taken equal to 500 mm (custom parameter);

r _n - rail number from a real set of joints;

R _n - rail numbers from a template set of joints;

- символ Кронекера (равен 0, если r_n и R_n не совпадают и 1, если совпадают).

- Kronecker symbol (equal to 0 if r _n and R _n do not match and 1 if they match).

If the cross-correlation coefficient K of the set of coordinates of the detected joints with any of the patterns exceeds a threshold value (for example, 0.92), a decision is made on the availability of turnout. In addition, in this way you can determine

1. Arrow diagram - to which of the 3 standard, template diagrams is the detected turnout.

2. How exactly is the arrow passed:

- woolly left or right;

- anti-wool left or right.

Consider an example. In FIG. Figure 1 shows the position of the rail joints of the turnout switch of scheme 2 - project 2768 (the square dots indicate the positions of the turnout objects - bolt joints and the clearance (gap) near the cross, giving a clear signal (response) to the MD control method; the coordinates of these objects in mm are shown on the horizontal axis )

Consider the arrow pattern in the form of distances X _i between the joints. The template corresponds to the arrow of the project 2768, Table 1, FIG. 2, walked anti-woolly to the left. In this case, the length of the turnout of the family is ~ 34 m.

Let the joints with the following absolute coordinates x _i be found during the passage along a certain length of 34 m, Table 2:

After normalization - bringing the first joint to zero, we obtain the following measurement results x _i , Table 3:

- это первый столбец набор данных Таблицы 1,

- это первый столбец набор данных Таблицы 3; слагаемое в формуле учитывается, только если рельсы совпадают:

- это второй столбец набор данных (1),

- это второй столбец набор данных (3). Получаем коэффициент взаимной корреляции около 0.99, поскольку координаты в наборах х и X очень близки.We substitute the two available sets of coordinates from Tables 1 and 3 into formula (1) to calculate the correlation coefficient: N = M = 9,

is the first column of the dataset of Table 1,

- This is the first column of the data set of Table 3; the term in the formula is taken into account only if the rails coincide:

is the second column of the data set (1),

is the second column of the data set (3). We get a cross-correlation coefficient of about 0.99, since the coordinates in the x and X sets are very close.

The analysis of MD, US and VT of flaw detection methods suggests that the most reliable method for detecting rail joints is the MD method, which provides good detecting ability and repeatability of measurement results. A comparison of MD and ultrasound methods is presented in FIG. 2, where it is clear that the ultrasonic method gives less reliable results, and, in addition, substantially depends on the quality of contact of electro-acoustic transducers with the rail surface.

Consider the possibility of implementing the proposed method when using MD flaw detector.

For the operation of the proposed method, FIG. 3, two known wheelsets 2 are installed on both strands of rails 1, on the axes of which are mounted coils 3 — magnetic field excitations in the rails 1 of the MD flaw detector. In a simpler case, known U-shaped magnets can be used. Between the points of entry of the magnetic field 4, receiving devices 5 are installed — induction coils (or Hall sensors) that perceive inhomogeneities of the magnetic field in the rail 1. Bolted 7 joints of the rails 1 are considered as inhomogeneities. When moving the flaw detector in processing unit 6 containing a computer with corresponding controllers detect rail joints and tie their coordinates to the rail using odometer 8 with absolute coordinates x _i . The obtained coordinates are normalized and compared with the matrix of arrow patterns according to formula (1), as shown above. The type of turnout is determined, the actual position of the wits of the turnout and the current direction of movement of the flaw detector.

Thus, the inventive method can be implemented and allows you to reliably determine the type of turnout switch, the position of the switches of the turnout switch and the current direction of movement of the flaw detector indirectly.

Information sources

1. Patent RU 2349480.

2. Patent EP 1747422.

3. Patent RU 44624.

4. Patent RU 48506.

5. Markov A.A., Kuznetsova E.A. Defectoscopy of rails. The formation and analysis of signals. Prince 2. Decoding of defectograms. - SPb .: Ultra Print. 2014-332 pp.

6. Patent RU 150721.

7.http: //murom-switch.ru/page/10. Product catalog Murom switch plant

Claims (1)

The method for determining the turnouts and the position of the switches of the turnouts, which consists in moving the flaw detector along the rail track, probing it, storing the sounding results in the defectogram, evaluating them and determining the actual position of the switch wits and the current direction of movement of the flaw detector, characterized in that when sounding, rail joints are detected, when evaluating, the relative distances between the joints are calculated, the obtained values are compared with the matrix of distance patterns between the joints is known in advance, the type of turnout is determined, as well as the actual position of the switches of the turnout and the current direction of movement of the flaw detector.

Images