JPS618610A - Apparatus for inspecting steel sheet surface - Google Patents

Abstract

PURPOSE:To decrease over detection and to improve the reliability of automation, by providing an optical flaw detecting device, an electromagnetic flaw detecting device and an arithmetic control part. CONSTITUTION:When a steel sheet 1 attains to the optical flaw detecting device A, a timing control part 6 outputs the signal from an imputted pulse generator 7. Thereat, the control part 5 performs synchronous control while adjusting an ITV2 and a stroboscope 3 to the sheet passing speed. The image pick up signal from the ITV2 is subjected to a primary flaw detection 16 through the parts 5 and 6, and the area information is stored 17. Simultaneously thereto, a detecting part 16 outputs the primary flaw signal to a microcomputer MC15, and the MC15 inputting the signal sends the signal to a servo circuit 14 of the electromagnetic flaw detecting device B, controls a probe 9 following to the position of detected 16 flaw to perform eddy current flaw detection. A flaw detecting part 8 outputs the flaw detection result to the MC 15 and a secondary detecting part 18. The part 8 compares the flaw detection result with the primary flaw detection data read from a frame memory 17 to perform the discrimination, the output is monitored 19, and the flaw is detected.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application This invention relates to a steel plate surface inspection device used in a continuous steel plate conveyance line.

Conventional technology On the exit side of the pickling line in the rolling process, flaw detection of the hot rolled steel strip is required in order to improve the quality of the cold rolled base material.

Conventionally, optical flaw detection methods, electromagnetic flaw detection methods, etc. have been used to detect flaws in hot rolled steel strips, but optical flaw detection methods often cause false detection due to contamination (water, oil, etc.) on the surface of the steel material after pickling. many,
Separate pretreatment was required to improve flaw detection accuracy.

In addition, although electromagnetic flaw detection is superior in terms of detection accuracy and ability to detect flaws on objects to be tested that move at high speed, the measurement range is narrow, so multiple expensive probes must be arranged depending on the width of the object to be tested. I had to.

Furthermore, the penetrating liquid method has a problem in flaw detection speed and is therefore not suitable for flaw detection on a continuous conveyance line.

-For this reason, when the steel plate is conveyed at a high speed and wide, and there is dirt due to pickling, such as in a continuous conveyance line, it is difficult to completely detect surface flaws on the steel plate using conventional methods. However, the current situation is that it ultimately relies on the operator's intuition, visual inspection, and judgment.

Purpose of the Invention Therefore, the present invention improves the above-mentioned disadvantages of the conventional technology in surface inspection of a continuous steel plate conveyance line, can sufficiently cope with the width and conveyance speed of the steel plate, and also eliminates stains caused by pickling. It is an object of the present invention to provide a steel sheet surface inspection device that can accurately identify defects and the like.

Structure of the Invention A steel plate surface inspection apparatus according to the present invention comprises: an imaging device for capturing an image of a steel plate surface; a flatness measuring device for measuring the flatness of a steel plate surface provided on the downstream side of the imaging device in the steel plate conveyance direction; The arithmetic and control unit controls both devices, and the arithmetic and control unit processes the image captured by the imaging device to select a portion with a high probability of being a steel plate surface flaw, and applies a flatness measuring device to this portion. It is unique in that it is constructed so that it can be moved and measured again to distinguish between scratches and dirt.

Example DESCRIPTION OF THE PREFERRED EMBODIMENTS The embodiments of the present invention will be explained below with reference to the illustrated embodiments.

FIG. 1 shows a control block diagram of an embodiment of the steel plate surface inspection apparatus according to the present invention. B, and an arithmetic control section C.

The optical flaw detection device [A] is installed near the priddle roll 23 where the steel sheet 1 has less backlash, as shown in FIG. 2, in a continuous conveyance line for steel sheets from hot rolling to cold rolling.

This optical flaw detection device A is equipped with three ITVs (CCD cameras) 2 arranged side by side in the width direction of the steel plate 1 and three strobes 3 that irradiate strobe light onto the imaging surface of the steel plate 1 of each ITY2. . These three strobes; two strobes 3 are placed on one side of the steel plate in the width direction, and the remaining strobe 3 is placed on the other side of the steel plate in the force direction. ing. A light shielding plate 4 is provided to prevent mutual interference of light between the strobes that are in a facing relationship.

The ITV 2 and the robot 3 are connected to a synchronization control unit 5), which controls the strobe light emission timing in accordance with the threading speed of the steel plate so that the ITV fields of view do not overlap. . A timing control section 6 is connected to the next stage of the synchronization control section 5, and the imaging signal from each ITV 2 is inputted from the synchronization control section 5 to the timing control section 6. The timing control section 6 receives the tracking number i from the pulse generator 7 that performs sheet tracking, controls the timing of imaging and image processing of the input imaging signal, and sends it to the arithmetic control section C side. Output. Further, a synchronization signal is outputted from the timing control section 6 to the synchronization control section 5, and synchronization control between the ITV 2 and the strobe 3 is performed in accordance with the threading speed.

Here, the relationship between the ITV 2 and the strobe 3 of the optical flaw detection device A will be explained in detail according to the layout diagram shown in the TLCA diagram.
Strobe 3 uses ITV to compensate for the lack of resolution of ITV2.
2. The camera is configured to irradiate strobe light obliquely onto the camera range AB. Also, the firing angle α and firing distance of strobe 3! 4 and 5, in order to detect a flaw with a depth of IIuIl on one surface of the steel plate, Step 1)
- By setting the irradiation angle of Bo 3 to less than 200, 1) ztu
It can be generated in a long form with a length of n or more. In addition to the shadows generated in this way, in order to bring the illuminance ratio in the width direction of the surface of the steel plate 1 due to oblique strobe irradiation into the resolution capability range of ITV2, the irradiation distance from the steel plate 1 with the strobe 3tI is expressed as 1000su+ or more. I placed it and got a good image.

In this example, three ITVs 2 and three strobes 3 are used, but if the width of the copper plate 1 is narrow, it is possible to use only one ITV 2, and similarly, if the strobe 3 has a sufficiently large irradiation capacity. It is also possible to use only one unit.

Furthermore, an electromagnetic flaw detection device B, which is a flatness measuring device, is provided downstream of the optical flaw detection device A by a predetermined distance on the conveyance line. This electromagnetic flaw detection device B has two probes 9 disposed so as to be freely displaceable in the width direction of the steel plate l.
and an eddy current flaw detection section 8 that uses the probe 9 as a detection end.

The two probes 9 are screwed into two ball screws 10 that each independently extend outward in the width direction from the center of the steel plate 10 in the width direction. A motor 11 is attached to each outer end of the ball screw 10 in the width direction of the steel plate.
The two probes 9 can be independently displaced in the width direction of the steel plate 1 by the ball screw 10 rotated by the ball screw 10.

A flaw detection position signal from an arithmetic control section C, which will be described later, is input to the motor 11 via a servo circuit 14. Furthermore, a pulse generator 12 and a tacho generator 13 are attached to each of the two motors 11, and these pulse generators
The rotational speed and rotational direction of the motor 11 indicating the displacement amount and displacement direction of each probe 9 are inputted to the servo circuit 14 by the generator 12 and the tacho generator 13 . In this way, the electromagnetic flaw detection device B covers half the area in the steel plate 1 direction by the flaw detection signal from the calculation control unit C.
The probes 9 are servo-controlled to perform eddy current flaw detection on the steel plate 1.

Furthermore, the steel plate surface inspection apparatus of the present invention is provided with an arithmetic control section C that controls the optical flaw detection apparatus A and the electromagnetic flaw detection apparatus B.

This arithmetic control section C is equipped with a 3-channel primary flaw detection section 16, which inputs the image signal from the timing control section 6 of the optical flaw detection device A, and detects flaws at a rate of about 30 times/second in real time. The judgment is made and the image of the flaw is stored in the frame memory 17 at the next stage. This primary flaw detection unit 11j differentiates the scanning line luminance of the ITV image and detects the presence or absence of a value exceeding a set value by differential slicing, and calculates the area exceeding this set value or by a peak value). Next, the presence/absence of flaws is determined, and if it is determined that there is a flaw, the screen information is output to the frame memory 17.

Primary flaw detection unit 16 and frame memory 17 line 3 IT
Each is composed of 3 channels (3ch) to correspond to V2, but if it has a large processing capacity, 1 channel each
It is also possible to configure it with channels.

A secondary flaw detection unit 18 is connected to the output side of the frame memory 17, and a portion determined to be flawed by the primary flaw detection unit 16 stored in the frame memory 17 is stored in the frame memory 17. Secondary flaw detection is performed from the flaw detection results inputted from the eddy current flaw detection section 8 of the electromagnetic flaw detection device B. Through this secondary flaw detection, the presence or absence of flaws is checked, and the type, depth, and level of danger of the flaws are determined.

A monitor 19 and a flaw detection unit 2- are connected to the output side of the secondary flaw detection section 18.
A still image of the flaw site, the type of flaw, and the quantified size of the flaw are displayed on the seventh monitor 19 as a result of the secondary flaw detection by the secondary flaw detection unit 18. , activate the flaw alarm 20 as necessary to notify the operator of the abnormality.

Further, a primary detection setting value change signal is output from the secondary flaw detection unit 18 to the primary flaw detection unit 16 as a result of secondary flaw detection. This primary detection setting value change signal is used when it is determined that a defect is not detected in the electromagnetic flaw detection device B for an image that is suspected to be a flaw in the primary detection (erroneous detection in the primary detection). Software logic for secondary detection automatically changes settings such as dark area cross-sectional area, dark area length, bright area/dark area discrimination level, etc. In this way, by improving the false detection rate of the primary flaw detection section 16, it becomes possible to cope with a much higher speed inspection through the plate and to perform an inspection with fewer steps.

Furthermore, the arithmetic control section C includes a microcomputer 15.
is provided, and when a flaw detection signal is input from the primary flaw detection section 16, the optical flaw detection device A is also located downstream in the conveying direction by a predetermined distance from the threading tracking signal of the pulse generator 7. When the flaw detection section reaches the electromagnetic flaw detection device B, the -C servo circuit 14 is controlled. Thus, the servo circuit 4 is configured to operate each motor 11 to control the probe 9 to follow the flaw detection portion of the steel plate 1.

The microcomputer 15 detects welding points from the flaw detection signal from the eddy current flaw detection section 8 and the input signal from the primary flaw detection section 16, and outputs a welding point detection signal to the pickling process computer 21. In addition, the pickling process computer 21 outputs display signals to the timing control section 6 such as the width of the sheet being passed, the material, and the position of the welding point.

Furthermore, information on the presence or absence of flaws, their location, etc. is inputted to the cold rolling process computer 22 from the pickling process computer 21, and the cold rolling speed is lowered when passing the sheet through the flaw inspection position to prevent sheet breakage accidents.

In the configuration described above, this device moves the steel plate 1 on the conveyance line.
When the sheet is transported and reaches the optical flaw detection device A, the timing control section 6 outputs the input tracking signal from the pulse generator 7 to the synchronization control section 5. Therefore, the synchronous control section 5 performs synchronous control of the ITV 2 and the strobe 3 in accordance with the sheet passing speed.

In this way, the image signal captured by the ITV 2 is sent from the synchronization control section 5 to the timing control section 6.

The timing control section 6 subjects the imaging signal to a still image, which is then outputted to the primary flaw detection section 16 for primary flaw detection.

Here, Fig. 6 (a), Φ), ((ri, (d)
A perspective view showing flaws on the steel plate surface, a cross-sectional view thereof, imaging brightness, and an example of secondary detection are shown respectively. Each part of the surface of the steel plate shown in FIG. 6(a) shows: 1) rash, 2) linear gloss, 2, 2, 2, 2) dirt from water or oil, 2) ear cracks, and 2) ear marks. The surface shape of this steel plate A-A' section is shown in Fig. 6 ('b), and the scanning line brightness on the imaging screen that appears in the A-A' section is shown in Fig. 6 (e). . Primary flaw detection section 1
6, the scanning line luminance upper and lower limit values of the imaging screen (Fig. 6 (e)
The presence or absence of flaws is determined by comparing with the value set in advance as the limit value, based on the amount exceeding the upper dotted line (black squared part in the same figure), the peak value after differentiation, the number of peaks, etc. do.

In this way, areas suspected of having flaws are detected when the value exceeds the set value, and
The screen information is stored in the frame memory 17.

At the same time, the primary flaw detection section 16 outputs a primary flaw detection signal to the microcomputer 15, and the microcomputer 15, which has input this signal, sends a signal to the servo circuit 14 of the electromagnetic flaw detection device B to detect the 91st flaw detection signal. Eddy current flaw detection is performed by controlling the probe 9 to follow a portion deemed to be a flaw by the flaw detection section 16.

The electromagnetic flaw detection section 8 outputs the flaw detection results to the microcomputer 15 and the secondary flaw detection section 18. The secondary flaw detection unit 18, which receives the flaw detection results from the electromagnetic flaw detection unit B, compares the flaw detection results with the primary flaw detection data read from the frame memory 17 and makes a determination.

Figure 6(d) shows an example of secondary flaw detection. Judgment is made and the area and depth are calculated.Similarly, ■'
Check the length as "one ear break", and check the length if the dark line and bright line are continuous and close with the edges.
The area and depth are calculated. ``Furthermore, the unclosed dark area ■'' and ■7.■, ■ in FIG. 1 (IL) are not considered as defects by secondary flaw detection.

In this way, the output of the secondary flaw detection section 18 is input to the monitor 19, which displays a still image of the flaw site, the type of flaw, and the quantified size of the flaw, thereby notifying the operator of the flaw detection.

Further, if necessary, an alarm is issued from the flaw alarm 20, and the rolling speed is reduced during cold rolling of the flawed area to prevent breakage of the steel plate during cold rolling.

Further, the secondary flaw detection unit 18 outputs a primary detection set value change signal to the primary flaw detection unit 16 to correct over-detection or under-detection in the primary flaw detection unit 16.

Further, a microcomputer 1 inputs the detection signal from the primary flaw detection section 16 and the flaw detection signal from the eddy current flaw detection section 8.
Since the welding point 5 can also detect the welding point of the steel plate, this welding point detection signal is outputted to the pickling process computer 21 and used for error correction of sheet threading tracking in the pickling process computer 21.

To explain this point in detail, when the pickling process computer 21 receives a signal indicating that welding has been performed from the welder on the pickling input side, the pickling process computer 21 calculates the movement of the pickling line and the position of the welding point over time. According to this calculation, when the welding point is considered to have arrived at the steel plate surface inspection apparatus, the microcomputer 15 is
Calculate the error from the previously calculated welding position (by sending 1 as the force and then obtaining the signal from the welding point detected by the steel plate surface inspection device itself). Based on this error, At the same time as the values such as the length of the steel strip to be wound are corrected, the threading tracking information sent to a device other than this device is reset for error correction.

As a result, for example, when cutting and removing a welding point on the pickling side shear, the positional information of the welding point becomes much more accurate. What used to be can be fully automated.

Note that while the electromagnetic flaw detector B [optical flaw detector A does not detect flaws, the flaws are detected near the edges of the steel plate in order to detect horizontal cracks (hair cracks) near the corners.

By employing the device of this embodiment, the amount of defects detected was reduced to 1/4 to 115 compared to the case using only the optical method, and the amount of overdetection was 1.2 when visually compared at the sample section. As a result of applying the flaw information from this device to the rolling speed adjustment during the cold rolling process, no plate breakage occurred during cold rolling.

In this example, the device of the present invention was installed between hot rolling and cold rolling, but by installing it after cold rolling, the final inspection of the product can be performed online, reducing false detections caused by rolling oil. be able to.

Effects of the Invention The embodiment of the steel sheet surface inspection apparatus according to the present invention is not only the above-mentioned, but also has the following effects.

In the steel plate surface inspection on a continuous steel plate conveyance line, it is fully compatible with the width of the steel plate and the conveyance speed, and it can accurately distinguish between stains and scratches caused by pickling, and it greatly reduces overdetection, making automation reliable. It becomes possible to bring one's sexuality to the forefront.

[Brief explanation of drawings]

Fig. 1 is a control block diagram of an embodiment of the steel plate surface inspection device according to the present invention, Fig. 2 is a schematic diagram showing the installation position of the ITV, Fig. 3 is a schematic diagram showing the irradiation angle and irradiation distance of the strobe, and Fig. 4
The figure is a graph showing the relationship between the irradiation angle and the flaw length.
Figure 5 is a graph showing the irradiation distance and the illuminance ratio of the left and right corners of the ITV field of view, Figure 6 (a) is a perspective view showing an example of a steel plate surface,
(in the same figure) is a cross-sectional view taken along the line A-A', (C) is a waveform diagram of the imaging luminance signal, and (d) is a pattern showing an example of secondary flaw detection. Equipment, B...Electromagnetic flaw detection device C...Calculation control unit, 1...Steel plate, 2...ITV. 3. Strobe, 4. Light shielding plate, 5. Synchronization control section, 6
...Timing control section, 7..Pulse generator,
8. Eddy current flaw detection section, 9. Probe 10 book, ball screw, 11. Motor, 12. Pulse 0 generator, 1
3. Tacho generator, 14. Servo circuit, 15. Micro computer, 16-. Primary flaw detection section, 17. 9 frame memory, 1
8. Secondary flaw detection unit, 19. Monitor, 20. Flaw alarm, 21. Pickling process reference computer, 22
@-Cold rolling 7' process-computer, 23...pridle roll. Figure 2 Figure 3 ;;-5+I super light sE#tノ

Claims (1)

[Scope of Claims] In a continuous steel plate conveyance line, an imaging device disposed upstream in the steel plate conveyance direction and configured to image the surface of the steel plate transferred on the conveyance line; and an imaging device disposed downstream of the imaging device in the steel plate conveyance direction. a flatness measuring device that measures the flatness of the surface of the steel sheet transported on the conveyance line; and a flatness measuring device that processes the captured screen input from the imaging device to select areas with a high probability of being surface flaws of the steel sheet. A steel plate surface inspection characterized by comprising: an arithmetic control unit that instructs the flatness measuring device to re-measure the selected portion and determines a flawed portion of the steel plate from the measurement value of the flatness measuring device. Device.