Method for Detecting Strip Edges

Abstract

A method of detecting strip edges in a cold rolling mill and/or strip-forming installation in which the strip edges (13, 14) are continuously detected in a contactless manner with two optical recognition systems, the right strip edge (14) and the left strip edge (13) are detected along a strip length of more than 800 mm with an image recording system (4, 5), the image data are communicated to a computer (11) and are processed there, and a so determined strip edge shape is taken into account during a regulation process.

Description

The invention relates to a method of detecting strip edges in a cold rolling mill and/or strip-forming installation according to which the strip edges are continuously detected in a contactless manner with two optical recognition systems.

DE 38 37 101 A1 discloses a method of controlling a strip path during rolling, in particular, finish-rolling of luminescent hot strips in a rolling mill train, which is formed of several, arranged one after another rolling mill stands, dependent on the position of both strip edges. The method includes combination of the following features:

for detecting the position of the strip, an optical recognition device which functions in a contactless manner, which is arranged centrally above the strip downstream of at least one rolling mill stand in the direction of running of the strip, and which simultaneously detects both strip edges with a measurement line that extends transverse to the strip running direction and passes through both edges.

With the measurement line, the distances of the strip edges from a reference point which corresponds to the center of the rolling mill train, are determined.

By comparing the measured distances of the strip edges from the reference point, an adjustment signal for an adjustment device for adjusting the roll gap of the rolling mill stand is generated.

B adjusting the roll gap, functioning of the rolling mill stand, which is located immediately upstream of the recognition device in the strip running direction, is corrected.

As a recognition device, a line camera is used.

JP 6 147 856 discloses an apparatus in which the strip edges of a rolling stock are detected, in contactless manner, with two cameras, and a conclusion about the strip center running is drawn by evaluation in a processor.

JP 63 225 also discloses detection of strip edges with two cameras.

In JP 80 54 217 A2, surface unevenness of a rolling stock is determined with several cameras.

With known recognition systems, only a point on the right strip edge and a point on the left strip edge are, respectively, detected with the optical recognition device. Such an arrangement does not permit to detect, e.g., a strip shape.

Proceeding from this state-of-the-art, the object of the invention is to improve the detection of the strip position and to automatically correct the strip position.

This object is achieved according to the invention with a method of detecting strip edges in a cold rolling mill and/or strip-forming installation according to which the strip edges are continuously detected in a contactless manner with two optical recognition systems, in which the right strip edge and the left strip edge are detected along a strip length of more than 800 mm with an image recording system, the image data are communicated to a computer and are processed there, and a so determined strip edge shape is taken into account during a regulation process.

Further embodiments of the method follow from subclaims.

In a first embodiment, the image recording device is a digital camera, i.e., a known and proven device is used.

The increase of the edge points from one point per strip edge to, e.g., twelve point per strip edge permits to obtain a more precise picture of the strip path and, thereby, strip position correction values, which permits to noticeably better regulate the roll angular position.

From optically recorded measurement values representing the course of the strip edges, data points which characterize the strip edge angular deviation and position, are determined. Based on these data points, information regarding deviation with respect to the strip center and the strip shape is determined and is used for an active control of, e.g., the adjustment of a pivotal position of the rolls.

In order to detect the strip path during insertion in the closed roll gap in cold rolling plants, an optical recognition system is used. The recognition system has, on one hand, the advantages of a contactless measurement and, on the other hand, of a simple expansion, without large changes in hardware. By a continuous image recordation with two recognition systems, each of which detects, respectively, a strip edge, a continuous recording in the monitored region is possible (quality reliability).

After conversion of image data into the necessary data point, the angular deviation and the position of the strip edges are determined.

The information regarding the strip position is used for controlling an active change of the pivotal position. By changing the pivotal position, the strip position is adjusted.

The entire calculation of the strip position and generation of control variables take place in a computer located outside of rolling mill and/or strip-forming system. With small dimensions of the used optical recognition systems, an expensive modification of the rolling mill and/or strip-forming system is eliminated.

At critical environmental conditions such as presence of a lubricant mist, the optical recognition system can be installed inside the rolling mill and/or strip-forming system. The installation takes place in a protected position; the distance from the strip can be arbitrarily selected. Only indirect illumination is necessary.

Furthermore, with the optical recognition system, there exists a possibility to obtain information about the strip edge condition. E.g., the strip edge fissures can be metrologically detected and evaluated. The advantage consists in reduction of damages resulting from the strip fissures.

In addition, with an optical recognition system, large surface defects, e.g., markings on the working roll that are transferred to the strip surface, can be recognized and correspondingly detected.

The optical recognition systems, as a result of their stationary installation, are suitable for all strip widths of the rolling mill, strip-forming system.

With the above-described system, also, the existing plant can be equipped.

Embodiments of the invention will now be described in detail below with reference to very schematic drawings. The drawings show:

FIG. 1 a plan view of a recognition system according to the present invention;

FIG. 2 a plan view illustrating differences between a known recognition system and the recognition system according to the present invention; and

FIG. 3 a plan view of a recognition system for a particular situation of strip edges.

FIG. 1 illustrates the principle of a recognition system 1. A rolled strip 3, which exits or enters a roll gap 2, is detected by two cameras 4, 5 arranged above and below the strip. The field of vision 6, 7 of the cameras 4, 5 embraces a region that does not include the strip center 8. The strip center can be, however, taken into consideration with installation of additional cameras.

The digital video signals 9, 10, which are obtained by the cameras 4, 5, are read into a computer 11. An evaluation software determines a plurality of edge points 12 of strip edges 13, 14. The position of the strip points 12 and their shape (see FIG. 2) are evaluated in the computer 11. If differences with a predetermined edge shape is detected, then e.g., the roll gap is adjusted by a signal for pivoting the rolls.

The action can take place before and/or after the following rolling mill stand.

FIG. 2 shows, in a simplified way, the differences between the strip edge detection only for one detection point 12 according to the state of the art (left column) and the strip edge detection according to the present invention for a plurality of edge points 12 (right column). While with the only one edge point 12, in effect, the distance and the position of both these edge points can be determined/detected, no evidence of the shape of the rolled strip 3 can be obtained. It is another matter with a strip edge detection with a plurality of edge points 12. With, shown in the example, twelve edge points 12 on the right strip edge 14 and on the left strip edge 13, it can be determined whether the rolled strip runs along a center (a), is pivoted to the left (b), forms an arch (c), or has a wavy shape (d).

At a particular condition, as shown in FIG. 3, it can be determined whether the rolled strip 3 has fissures 15 on the left strip 13 and/or the right strip edge 14.

LIST OF REFERENCE NUMERALS

• • 1. Recognition system
  • 2. Roll gap
  • 3. Rolled strip
  • 4. Camera
  • 5. Camera
  • 6. Field of vision
  • 7. Field of vision
  • 8. Strip middle
  • 9. Video signal
  • 10. Video signal
  • 11. Computer
  • 12. Edge point
  • 13. Left strip edge
  • 14. Right strip edge
  • 15. Fissure

Claims

1-9. (canceled)

10. A method of detecting strip edges in a cold rolling mill according to which the strip edges (13, 14) are continuously detected in a contactless manner with two optical recognition systems,

characterized in that

the right strip edge (14) and the left strip edge (13) are simultaneously detected along a strip length of more than 800 mm with an image recording system (4, 5) as image data, the image data are communicated to a computer (11) and are processed there, and so a strip edge shape is determined,

wherein at least eight edge points (12) are taken into consideration, and

image data, which are determined in a computer (11) are used for regulating adjustment of a roll gap.

11. A method according to claim 10,

characterized in that

position and angular inclination of edge points (12) are determined by reduction of image data into necessary data points in a computer (11).

12. A method according to claim 10,

characterized in that

the recognition system (1) and the computer (11) are fitted on existing installations.

13. A method according to claim 10,

characterized in that

an entire strip shape in a monitored region is continuously recorded.

14. A method according to claim 10,

characterized in that

in accordance with an image of the strip shape, strip position correction values are set by an angular position of rolls.

15. A method according to claim 10,

characterized in that

data, which are generated by the computer (11) are additionally used for detection, if necessary, elimination of strip edge fissures and/or strip surface defects.