RU2291066C2 - Method of optical monitoring of wear of contact wires - Google Patents

Abstract

FIELD: railway transport.

SUBSTANCE: method comprises surveying contact wires by means of a scanner of high spatial resolution made of two CCD-gages shifted one with respect to the other by 0.5 pixel and connected with the computer and GPS-receiver, determining extent of wear of each wire by means of a computer program for processing the results of survey, and preparing a map of sites with marks on the extent of wire wear.

EFFECT: enhanced reliability.

4 dwg

Description

The invention relates to remote measuring systems and is intended to determine the degree of wear of the contact wires of the power supply of railway transport as part of specialized laboratory wagons.

A device for measuring the width of the area of wear of the contact wire, consisting of several television cameras with overlapping fields of view located on the bracket collector ski pantograph. The cameras are installed in one row parallel to the pantograph runner below the level of a linear illuminator, which, in turn, is mounted on the same bracket parallel to the current collector ski, but closer to the contact wire [1]. The object of observation of each camera during the movement of the laboratory car is a luminous strip on the contact wire. The width of the wear pad is determined by the number of cells in the camera having a higher brightness relative to neighboring elements.

The binary code corresponding to the conditional size of the luminous strip enters the information processing unit based on the PC. In the processing unit, this code is converted into a physical quantity (mm), based on which the cross section and the amount of wear of the contact wire are calculated. The measurement results in the form of a set of numerical values or graphs are displayed on a printing device.

The disadvantages of this device are:

a) Low measuring accuracy in the longitudinal direction when the laboratory car moves at medium and high speeds. So, at a frame-scan frequency of a camera of 50 Hz and a speed of 60 km / h, measurements are performed in increments of 0.3 m. This does not make it possible to identify all defects and consider such control complete.

b) The filming system has low noise immunity, as it is not protected from the effects of precipitation and dust.

c) The device does not provide the location of the shooting materials,

which greatly complicates the practical use of measurement results.

Another type of device for determining the degree of wear of the contact wire is considered in [2], in which the measurement of the width of the sliding area is carried out using two or more optoelectronic heads. Each head includes one line of charge-coupled receivers (CCD), which records radiation reflected from a linear illuminator. Contact wire is shot simultaneously and independently by several optoelectronic heads with a higher scanning frequency.

In this device, as well as in the device [1], codes are input to the input of the processing computer from each CCD line, specifying the conditional width of the wire wear pad. Codes are translated into physical quantities, which are then averaged over the number of CCD arrays. Moreover, the resolution of the device in the transverse direction is less than [1] and is determined by the number of photosensitive elements in one line, which, as a rule, is equal to 1024.

It should be noted that this device also lacks a mechanism for linking the measurement results to the terrain. In addition, the presence of several independent optoelectronic heads leads to a more complicated and expensive equipment, since it is necessary to ensure photometric alignment of CCD line parameters, as well as geometric alignment of the field of view of all heads. Otherwise, the operation of averaging measurements will not be performed correctly and will not give an increase in accuracy.

The main disadvantage of the considered devices is that the mechanism for measuring the width of the slip strip provided in them does not allow to identify the places of local thinning of the wire in which it breaks. These sections are characterized by a partial burn-out of the wire caused by an electric arc in places where the pantograph has no contact with the current collector, do not have a shiny and correspondingly brighter surface, and are not registered by the devices. The lack of the possibility of post-session analysis of captured images and clarification of the measurements made this disadvantage for the mentioned devices is fundamental.

The aim of the invention is to remedy these shortcomings by improving the accuracy, reliability of measurements and automation of the process of their georeferencing based on computer processing and analysis of the scanner image of contact wires taken during the movement of the laboratory car.

The technical result of the invention is to ensure trouble-free operation of railway transport by obtaining objective information about possible places of a breakdown of a current-supply contact network and taking timely preventive repair measures to replace defective sections of a contact wire.

The stated goal is achieved by the fact that for scanner video shooting of the contact network (1), the laboratory car is equipped with a shooting complex, which includes: a GPS receiver (4), a scanner (3), an analog-to-digital converter board and inputting information into a computer (5) and processing computer (6), Fig. 1. The scanner includes two CCD arrays, a translucent mirror (c), which separates the light flux by CCD arrays, and a lens with a hood (d) that protects the lens from precipitation and extraneous radiation. A scanner installed in the opposite direction to the laboratory carriage registers the light flux reflected from the contact wires generated by the emitter (2). In order to minimize large-scale distortions caused by a change in the height of the contact network, the survey is performed at an angle of 3.3 m. In this case, the spatial resolution in the transverse direction from one ruler will be 0.6 mm with an average height of sag of the contact wire.

Each CCD array contains 1,024 elements with a photosensitive receiver measuring 13 x 13 microns. To increase the spatial resolution by post-session computer processing of materials, the CCD array images are shifted by ^l / ₂ elements per line. Reading information from the rulers occurs simultaneously at a speed of 700 p./s. During this time, at a speed of a laboratory car of 50 km / h, the step of polling the wire will be 20 mm.

The image formed by the CCD arrays is transmitted via the interface card to a personal computer. At the same time, the current geographical ones are entered in the service part of the line; the coordinates of the car and the distance from the starting point of the survey, coming from the GPS-receiver. Computer (6) has a configuration, for example, no worse than the following:

- Pentium-IV processor with a capacity of 2500 MHz;

- RAM 2 GB;

- monitor 17 ";

- color inkjet printer;

- 200 GB hard drive, keyboard, mouse;

- OS Windows 2000 / XP.

The process of shooting and processing materials is controlled by the operator of the complex (7), Fig. 1.

A contact network survey is carried out in two stages. At the first stage, the intended area is shot, and at the second, the processing and analysis of the captured images. The speed of movement of the laboratory car and the length of the surveyed area are selected from the conditions for the formation of an image file of 5-7 GB.

The time-lapse shooting process is illustrated in the upper part of FIG. In block 1, the formation of the current line from 2 CCD lines takes place. In block 2, the geographical coordinates of the laboratory car are determined: φ - latitude, λ - longitude and distance from the start point of shooting L. This information is entered into the service part of the image line and the combined data is entered into the computer’s RAM (block 3).

; n - номер столбца,

; N - число пикселей в одной ПЗС-линейке;The control program reads data from the computer's memory and displays a new line on the monitor screen, organizing continuous scrolling of the captured image, and thereby allows you to establish control over the shooting process (block 4). At the same time, the line is read from memory and written to the hard disk (block 5). At the end of the survey, the image B = [b (m, n)] of the entire survey area, where m is the line number, is created from the two CCD arrays on the hard disk

; n is the column number,

; N is the number of pixels in one CCD array;

b ₁ , b ₂ - the brightness of the image elements from the first and second CCD arrays, respectively.

At the second stage, to determine the degree of wear of the contact wire, the following steps are taken for computer processing of image B.

Firstly, the preliminary processing of the image B is performed in order to form a new image B ^{* of} high resolution (block 6). The image forming principle B ^{* is} illustrated in FIG. 3, which shows the relative position of the image elements m - and the lines from the first and second CCD arrays. From the drawing it follows that the samples b ₁ , b ₂ and b ^* are interconnected by the formula:

b (m, 1) = b ^* (m, 1) + b ^* (m, 2), 2b (m, 2) = b ^* (m, 2) + b ^* (m, 3), etc.

The multiplier 2 performs the normalization function so that the radiation energy attributable to one of the outgoing CCD arrays is equal to the energy of the new image B ^* . The formula for determining an arbitrary element B ^{* is} as follows:

Thus, as a result of pre-processing, a new image B ^* is formed in the computer's memory with a spatial resolution 2 times higher than that obtained when shooting with one CCD-ruler and equal to 0.3 mm.

During the second processing step (block 7, FIG. 2), images of two contact wires are highlighted in picture B ^* . Using the threshold algorithm, a brighter section of each wire corresponding to the wear strip is recognized and the width of the wire and the width of the sliding area are calculated. To increase the accuracy of calculations, the measurements obtained are smoothed by the least squares method in a sliding window. The ratio of the measured values of the wire diameter to the size of the wear area is compiled, based on which the residual height of each wire and its degree of wear are calculated.

The portions of the contact wire image on which the light strip is not recognized are additionally analyzed visually on the computer screen 6 (Fig. 1) for the presence of a local burnout of the wire (block 8, Fig. 2).

Based on the geographical coordinates that are part of the image lines that specify the current location of the laboratory car and, accordingly, the scanned line, a cartographic plan of the contact network is formed with colored marks of sections of varying degrees of wear. The plan attached to the terrain is printed on the printer and transmitted to the repair team (block 9, figure 2). Additionally, in the form of a file or its printer copy, a graph of the degree of wire wear is formed depending on the path traveled by the laboratory car. An example of a cartographic plan with symbols (exact measurement values changed) is presented in Fig.4. To carry out repair work, the defective section of the wire on the ground is identified using a GPS-receiver or on the basis of transferring the length of the distance traveled to the reference system adopted by the railway with an accuracy of 2-3 m.

Information sources

1. RF patent No. 2120866. Device for measuring and recording wear of a contact wire.

2. RF patent №2137622. Device for measuring contact wire parameters.

3. A device for determining the wear of the contact wire. World Railways, 2001, No. 9 / H. Nagasawa et al. Quarterly Report of RIRI, 2000, No. 4, p. 117-119.

4. Measuring system for determining the position and wear of the contact wire. Railways of the World, 2003, No. 4 / B.Sames. Elektrische Bahnen, 2001, No. 12, s. 490-495.

Claims (1)

A method for optical control of wear of contact wires of a railway power supply network consists in using a scanner mounted on the roof of a moving laboratory car and consisting of two CCD arrays, shifted in the focal plane relative to each other in the row direction by 0.5 sizes of one photodetector element and oriented transversely to the longitudinal axis of the car, they shoot wires that are 3.3 m away from the scanner lens and illuminated from below by a radiator located on the roofs the car at a certain distance from the scanner, and form a contact wire image in one of the spectral subbands, which is stored in the memory of the computer connected to the scanner, and the current part of the line includes the current geographic coordinates of the car and the distance from the start point of the path coming from the GPS receiver, using computer tools by algebraic operations with brightness codes belonging to different CCD arrays, a new image with spatial resolution in 2 times higher than that obtained when shooting with one CCD-ruler, the image of the first contact wire is sequentially highlighted on the new image, the width of the contact area of the wire is calculated visually or using mathematical methods for processing video data and the degree of wear of the wire is calculated, and areas in which there is a partial thinning of the wire due to its partial burnout, then similar operations are applied to the image of the second contact wire, according to known geo graphic coordinates that specify the location of the car, scanner and the observation section of the contact wires, form a table of the degree of wear of the wires depending on the path traveled by the car, as well as a cartographic image of the contact network, which marks the places where each of the contact wires has an increased degree of wear using a cartographic plan or table, determine on the ground areas of the contact wire to be repaired.

Images