JP2955618B2 - Inspection method for weld surface defects of UO steel pipe - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting the surface of a welded portion of a UO steel pipe, for example, a method of determining the presence or absence of a flaw by imaging a surface flaw of a welded portion of a UO steel pipe by a light cutting method. It is something.

[0002]

2. Description of the Related Art Conventionally, various methods have been used to inspect flaws on the surface of a welded portion of a UO steel pipe. Current U
The surface flaw inspection of the welded portion of the O-steel pipe is performed by a visual inspection by a human.

[0003] By the way, general surface flaw inspection techniques include, for example, an optical method, an eddy current method, a leakage magnetic flux method and the like. Among these surface flaw inspection methods, the optical method scans and irradiates a laser spot or the like onto an inspection object and processes a time-series light reception signal obtained by receiving specular reflection light or scattered light. This is to determine the flaw, and is used as a means for inspecting the surface flaw of a steel sheet.

In the eddy current method, an eddy current is excited on the surface of the inspection object by a high-frequency magnetic field, a detection coil is brought close to the inspection object, and a magnetic flux value interlinking with the detection coil is detected. Is to be inspected,
It is used as a surface defect inspection means for various inspection objects.

[0005] The leakage magnetic flux method is a technique of detecting a change in leakage magnetic flux by moving a magnetic sensor close to the object to be inspected while moving a magnetizer close to the object to be inspected, and detecting a flaw. Yes, it is generally used as a method for detecting surface defects on steel sheets.

[0006]

In the above-described conventional flaw detection method, generally, an object to be inspected having a uniform shape is targeted. However, FIG. 10 to FIG. 12 show schematic shapes of undercut flaws which are examples of flaws generated at the end of the welded portion. Therefore, in order to inspect the fine undercut flaw 21 generated at the welded end, an inspection method capable of precisely inspecting the welded portion having undulations and its vicinity is required.

When the surface of the inspection object has undulations, for example, in the optical method, the light receiving intensity fluctuates due to a change in the intersection angle between the plane including the optical axis of the light emitting and receiving system and the inspection object surface. Therefore, it is difficult to inspect minute flaws.

Further, in order to obtain detection accuracy by a technique such as an eddy current method or a leakage magnetic flux method, it is necessary to keep the gap between the sensor and the surface of the inspection object short and to keep the gap constant. Detection of flaws on the surface is difficult.

As a method of measuring the shape, there is a light section method. In FIG. 10, the shape of the undercut flaw 21 is
Although it is shown large for easy understanding, it is actually necessary to detect very fine ones because it is harmful, and the shape of the welded part is not constant. No treatment method has been proposed.

SUMMARY OF THE INVENTION In view of the above problems, an object of the present invention is to make it possible to detect a surface flaw of a welded portion of a UO steel pipe with high reliability.

[0011]

According to the method of the present invention for inspecting the surface of a welded portion of a UO steel pipe, the slit light is applied to the welded portion of the UO steel pipe and the vicinity thereof, and the slit light is obtained by imaging with an imaging device. In the method of inspecting the surface of a welded portion of a UO steel pipe for detecting a flaw of a welded part using an optical cut image, while changing a position along a longitudinal direction of the UO steel pipe,
The light-section image is thinned to obtain an X-
A first process for sequentially generating a thinned data sequence Sn (x) in the direction, and a second process for storing the thinned data sequence Sn (x) sequentially generated by the first process in a data sequence storage unit And a point B (X _n , y _B ) on the thinned data string S (x) from the thinned data string stored in the data string storage unit and an arbitrary number of pixels d from the point B. _x, a first side point A ( _Xn
−d _x , y _A ) and the second side point C (X _n + d _x , y _C ) are set, and the calculation for obtaining the difference element dy of the difference data sequence for these Y coordinates is performed over the entire range of the X coordinates. A third processing to be sequentially performed; a fourth processing to create a difference data string D (x) in which the difference elements dy sequentially obtained by the third processing are arranged in correspondence with the X coordinate; Fifth processing in which a predetermined predetermined number of continuous moving sum calculations are performed on the difference data string D (x) created in the processing to create flaw detection data E (x) in which a flaw is emphasized; A sixth process of determining the presence or absence of a flaw by comparing the flaw detection data E (x) created in the above with a predetermined comparison value, and, when it is determined that there is a flaw in the sixth process, the thinning is performed. Performing the seventh process of measuring the depth of a flaw using the data sequence Sn (x). Features.

[0012]

According to the above-described flaw detection method, since processing for emphasizing and detecting only flaw parts is performed, even fine flaws in the undulated UO steel pipe welds such as overlays can be reliably obtained. In addition to being able to detect, when the presence of a flaw is determined, by measuring the depth of the flaw, information on the flaw depth can also be obtained.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the method for inspecting the surface of a weld of a UO steel pipe according to the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an example of a flaw inspection device for explaining one embodiment of the present invention.

In FIG. 1, 1 is a UO steel pipe, and 2 is a welded portion thereof. When such a surface flaw of the UO steel tube 1 is inspected, the slit light projector 3 emits light to form a slit light 4 at the welded portion and in the vicinity thereof. Further, an imaging device 5 is disposed at a position where the slit light 4 is captured.
Here, the case where the welded portion on the inner surface of the UO steel pipe is inspected is shown, but in the case of the outer surface, a slit light may be formed in the welded portion on the outer surface and in the vicinity thereof by a similar method and imaged. .

As the slit light projector 3, for example, FIG.
As shown in (1), the laser diode 13 emits light, and the laser light is collected by the collimator lens 14. Then, an apparatus for forming the slit light 4 by spreading the laser light into a fan shape by the cylindrical lens 15 is used. Thereby, for example, the slit light 4 having a half width of about 0.1 mm is formed.

The imaging device 5 is, for example, a CCD.
A camera is used. In a normal case, one imaging device may be used. However, in the case of detecting a minute flaw, it is desired that the imaging device 5 has a sufficient resolution. It is conceivable that this can be dealt with by arranging them. FIG. 1 shows an example in which two imaging devices are used.

By the way, welding flaws called "undercuts" often occur at the foot of the welded portion 2 as indicated by reference numeral 21 in FIG. The shape is generally V-shaped as shown in FIG. 4, which is a cross-sectional view taken along the line AA ′ in FIG.

Accordingly, when the slit light 4 between BB 'in FIG. 3 is photographed by the image pickup device 5, a light cut image 22 as shown in FIG. 5 is obtained. The thus-obtained light-section image 22 is input to the flaw detection processing unit 6.

In the flaw detection processing unit 6, first, the light-section image 22 is A / D-converted in the image storage unit 7 and stored as a two-dimensional digital value (x, y). The thinning is performed by the unit 8, and FIG. 6 shows the thinning processing performed here in detail.

As shown in FIG. 6, a process of selecting a point having the maximum luminance on one line in the screen y direction shown in FIG. 6 is performed on the two-dimensional image data p (x, y) in the screen x direction. To the maximum luminance point y corresponding to the x coordinate.
A thinned data sequence S (x) in which coordinate values are arranged is obtained.

At this time, if the luminance of the point selected as the maximum luminance does not reach the luminance threshold so that the noise outside the slit image is not erroneously selected for the portion where the luminance of the slit image is low, the maximum at x _n-1 Y coordinate S (x
_{As the} same value as _n-1 ), the thinning data is prevented from deviating from the slit image.

Next, in the difference processing section 9, the thinning processing section 8
Performs a calculation of the difference data sequence D (x) for the thinned data sequence S (x) output by the. FIG. 7 shows details of the calculation method. An appropriate number of pixels dx is set, and the thinned data sequence S
(X) a point on _B (x _n, y B) for the point A (x _n
−dx, y _A ) and a point C (x _n + dx, y _c ), and for these y coordinates, dy = (y _c −y _B ) − (y _B −y _A ) (1) Operate the element dy.

This operation is performed over the entire area of the x-coordinate, and a difference data string D (x) in which the difference elements dy are arranged corresponding to the x-coordinate is created. In the difference data sequence D (x) obtained in this way, as shown in FIG. 7, the non-flaw portion has a value close to 0, and the V-shaped flaw in the thinned data sequence S (x). Only the part has a large value.

The determination of a flaw may be performed by using the difference data string D (x). However, in order to further improve the determination accuracy, the flaw emphasizing processing unit 10 performs the difference data string D (x) as shown in FIG. The moving sum calculation of k pieces of continuous data of (x) is performed by the following two equations.

[0025]

(Equation 1)

Then, a flaw detection data string E (x) in which the flaw portion is further emphasized is created. In the flaw presence determination unit 11,
The size of the flaw detection data sequence E (x) is compared with a preset flaw detection threshold. Then, when all the elements of the flaw detection data sequence E (x) do not exceed the flaw detection threshold value, it is detected that there is no flaw, and the flaw detection processing on the image is ended.

As shown in FIG. 8, when an element exceeding the flaw detection threshold value exists in the flaw detection data string E (x), it is detected that a flaw exists, and the flaw depth determination unit 12 The determination of the flaw depth is performed in.

In the flaw depth determining section 12, the thinned data sequence S
The flaw determination is performed by measuring the flaw depth using (x). Also,
As shown in FIG. 9, UO in the thinned data sequence S (x)
An approximate straight line of a portion corresponding to the base material portion of the steel pipe is created, and the depth of the flaw portion is calculated as the maximum number of pixels in which the element of the thinned data sequence S (x) is less than the value on the base material approximate straight line. Then, the maximum depth value of the flaw is calculated by taking the product of the length and the length in the flaw depth direction corresponding to one pixel, which is measured in advance.

Similarly, the above processing is performed over the entire length of the steel pipe, and the flaw presence / absence information and flaw depth information obtained thereby are transmitted to an external device such as a product management computer.

Next, an example in which flaws are inspected by setting specific numerical values in the above-described embodiment will be described. In the slit light projector 3, two 20 mW laser diodes are used.
The pulse was turned on for msec. In addition, CCD is used for the imaging device.
An image was taken with a camera in a visual field of 32 mm in the width direction and 45 mm in the depth direction, and then processed using an image processing device of 512 × 512 pixels.

The difference processing section was set to dx = 10, the flaw enhancement processing section was set to k = 6, and the flaw detection threshold value in the flaw presence / absence determination section was set to 30. The longitudinal direction of the UO steel pipe was inspected at a pitch of 2 mm. Undercut flaws having a depth of 0.2 mm or more in the longitudinal direction and 3 mm or more were reliably detected, and the flaw depth was determined with an accuracy of about ± 0.05 mm.

[0032]

According to the present invention, as described above, according to the present invention, an X-ray image is generated while changing the position along the longitudinal direction of the UO steel pipe, and the X-ray image is processed by thinning the X-ray image. thinning data sequence Sn (x) are sequentially generated in the X direction in the -Y coordinate, from among the thinned data string, a predetermined point B (X _{n, y B)} for any number of pixels from the point B d _x
Side point _{A (X n -d x, y} A) and the second side point C _(X n +
d _x , y _C ) are set, the calculation for obtaining the difference element dy of the difference data string is sequentially performed over the entire area of the X coordinate for these Y coordinates, and the difference elements dy are arranged in correspondence with the X coordinate. To generate a differential data sequence D (x), and further perform a predetermined number of consecutive moving sum calculations in the differential data sequence D (x) to generate flaw detection data E (x) in which flaw portions are emphasized. The presence or absence of a flaw is determined by comparing the flaw detection data E (x) with a predetermined comparison value. If there is a flaw, the depth of the flaw is measured using the thinned data sequence Sn (x). As a result, the flaw detection can be performed with high reliability even for fine flaws on the surface of the welded portion of the UO steel pipe having undulations such as buildup. Furthermore, according to the present invention, not only is the presence or absence of a flaw determined, but if there is a flaw, the depth of the flaw is measured, so that information on the flaw depth can also be obtained. By these, the present invention,
The present invention can be suitably applied to automatic inspection of a welded portion of a UO steel pipe.

[Brief description of the drawings]

FIG. 1 is a block diagram of a flaw inspection device for explaining one embodiment of the present invention.

FIG. 2 is a configuration diagram schematically showing an example of a slit light projector.

FIG. 3 is a partial perspective view of a UO steel pipe showing a schematic shape of an undercut flaw in a weld portion, which is a flaw to be detected.

FIG. 4 is a sectional view taken along line A-A 'in FIG.

FIG. 5 is a diagram showing a light-section image taken along line BB ′ in FIG. 3;

FIG. 6 is an explanatory diagram showing the contents of processing performed by a thinning processing unit.

FIG. 7 is an explanatory diagram illustrating the content of a difference process performed by a difference processing unit.

FIG. 8 is an explanatory diagram illustrating a process based on a moving sum of a plurality of pixels performed by a flaw enhancement processing unit.

FIG. 9 is an explanatory diagram illustrating a method of measuring the flaw depth performed by the flaw depth determination unit.

FIG. 10 is an explanatory diagram showing a schematic shape of an undercut flaw generated at an end of a welded portion.

FIG. 11 is a sectional view taken along line A-A 'in FIG.

FIG. 12 is a sectional view taken along the line C-C 'in FIG.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 UO steel pipe 2 UO steel pipe welded part 3 slit light projector 4 slit light 5 imaging device 6 flaw detection processing part 7 image storage part 8 thinning processing part 9 difference processing part 10 flaw enhancement processing part 11 flaw presence / absence judgment part 12 flaw depth Judgment unit

Continued on the front page (72) Inventor Kazuo Takashima 8-1-1 Tsukaguchi Honcho, Amagasaki City Mitsubishi Electric Corporation In-house Research Institute of Industrial Systems (72) Inventor Katsuya Ueki 1-1-2 Wadazakicho, Hyogo-ku, Kobe Mitsubishi Electric (56) References JP-A-1-227910 (JP, A) JP-A-1-224649 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G01N 21 / 84-21/91

Claims (1)

(57) [Claims] 1. A welded part of a UO steel pipe, which irradiates a slit light to a welded part of a UO steel pipe and its vicinity, and detects a flaw in the welded part by using a light section image obtained by imaging the slit light by an imaging device. In the surface defect inspection method, the light-section image is subjected to thinning processing while changing the position along the longitudinal direction of the UO steel pipe to sequentially generate a thinned data string Sn (x) in the X direction on the XY coordinate. A first process to store the thinned data sequence Sn (x) sequentially generated by the first process in a data sequence storage unit; and a thinning process stored in the data sequence storage unit. From the data sequence, a point B (X _n ,
y _B ), a first side point A (X _n −d _x , y) separated from the point B by an arbitrary number of pixels d _{x in} both side directions.
_A ) and a second side point C (X _n + d _x , y _C ) are set, and for these Y coordinates, a calculation for obtaining a difference element dy of a difference data sequence is sequentially performed over the entire X coordinate. Processing, a fourth processing of creating a difference data sequence D (x) in which the difference elements dy sequentially obtained by the third processing are arranged in correspondence with the X coordinates, and a difference created by the fourth processing. Fifth processing in which a predetermined predetermined number of consecutive moving sum calculations in the data string D (x) are performed to generate flaw detection data E (x) in which flaws are emphasized; and flaw detection generated in the fifth processing. A sixth process of comparing the data E (x) with a predetermined comparison value to determine the presence or absence of a flaw; and, when it is determined in the sixth process that there is a flaw, the thinned data sequence Sn (x )) And a seventh process of measuring the depth of the flaw using Weld surface defect inspection method of UO pipe.