Metal pipe defect determination device and defect determination method

Abstract

To provide a technique capable of more accurately detecting harmful defects on a surface of a metal pipe.SOLUTION: A metal pipe defect determination device, which determines the presence / absence of defects on a metal pipe surface, comprises: a thickness expansion data acquisition unit 11A which acquires thickness expansion data in the two-dimensional direction made of thickness data for each unit area aligned in the circumferential direction and the longitudinal direction of the metal pipe from flaw detection data of the ultrasonic flaw detection of the metal pipe surface; an area setting unit 11C which sets a determination area consisting of regions where the unit areas are continuous in at least m or more (m: an integer of 2 or more) in the circumferential direction and n or more (n: an integer of 2 or more) in the longitudinal direction in the thickness expansion data; a determination criterion setting unit 11D which sets a determination criterion value for determining defects on the basis of the thickness data within the determination area; and a defect determination unit 11E which compares each thickness data in the determination area set with the determination criterion value with the determination criterion value to determine whether or not there is a defect.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a technique for determining the presence or absence of minute defects on the surface of a metal pipe, such as a steel pipe, by using the results of ultrasonic flaw detection on the surface of the metal pipe. The present invention is particularly useful for determining (detecting) the presence or absence of defects on the inner surface of a metal pipe, which are difficult to detect visually. [Background technology]

[0002] Some steel pipes are required by standards to have their wall thickness measured. The wall thickness of such steel pipes is mainly measured by ultrasonic testing. In ultrasonic testing, measurements are performed over the entire circumference of the steel pipe in the circumferential direction (also called the pipe circumferential direction) and the longitudinal direction (also called the pipe axial direction). There is an allowable range for the wall thickness of a steel pipe. In other words, there are upper and lower limit values ​​for the measured wall thickness. Then, for example, the pass / fail judgment of the wall thickness of the steel pipe is made depending on whether the measured wall thickness data is thicker than the upper limit thickness or thinner than the lower limit thickness.

[0003] Here, there are cases where minute defects (also simply called microdefects) are present on the inner surface of the steel pipe. Such microdefects can be confirmed as a sudden (localized) decrease in wall thickness on the steel pipe surface. If the depth of the microdefect is greater than the allowable depth, the microdefect must be detected as a harmful defect. However, the wall thickness of the part where such a microdefect exists may not fall below the lower limit wall thickness. In other words, such microdefects cannot be detected by determining whether or not the wall thickness is within the allowable range as described above. Conventionally, such harmful micro-defects have been detected by visual inspection of the exterior by operators, but because this is a visual inspection, there is a risk that steel pipes with harmful defects on the inside surface will be shipped out.

[0004] In response to this, methods for detecting such harmful microscopic defects include those described in Patent Documents 1 and 2, for example. In Patent Document 1, the change in wall thickness in the circumferential direction of the pipe and the steepness are monitored to determine pass / fail, while in Patent Document 2, it is determined to be harmful if there are consecutive points exceeding a threshold value. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2011-153974 A [Patent Document 2] JP 2003-222617 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, the method described in Patent Document 1 detects defects from a group of information arranged one-dimensionally in the circumferential or longitudinal direction, which may make it impossible to detect defects that run diagonally. Similarly, in the method described in Patent Document 2, defects are detected using information arranged one-dimensionally in the circumferential or longitudinal direction. For this reason, there is a risk that defects that are continuous diagonally cannot be detected. In addition, defects are recognized as abrupt changes in thickness relative to the surface surrounding the defect. Therefore, when defects are judged using a fixed threshold value as in Patent Document 2, there is a problem that the detection accuracy of harmful defects is poor.

[0007] The present invention has been made in light of the above-mentioned points, and has an object to provide a technique capable of detecting harmful defects on the surface of a metal pipe with higher accuracy. [Means for solving the problem]

[0008] In order to solve the problems, one aspect of the present invention is a metal pipe defect detection device that determines the presence or absence of defects on a surface of a metal pipe, and includes: a thickness expansion data acquisition unit that acquires two-dimensional thickness expansion data consisting of thickness data for each unit area arranged in the circumferential and longitudinal directions of the metal pipe from flaw detection data acquired by performing ultrasonic flaw detection in two-dimensional directions, that is, the longitudinal and circumferential directions, along the surface of the metal pipe; an area setting unit that sets a judgment area in the thickness expansion data, where the unit areas are continuous for m or more in the circumferential direction (m: an integer of 2 or more) and n or more in the longitudinal direction (n: an integer of 2 or more); a judgment criterion setting unit that sets a judgment reference value for judging defects based on the thickness data present in the judgment area; and a flaw detection unit that determines whether or not a defect exists by comparing each piece of thickness data in the judgment area for which the judgment reference value has been set with the judgment reference value. Effect of the Invention

[0009] According to an aspect of the present invention, a two-dimensional region along the surface of a metal pipe, i.e., a region having two or more consecutive unit areas in both the circumferential and longitudinal directions, is set as a judgment area. Then, a judgment reference value is set based on the wall thickness within the judgment area. Therefore, a judgment reference value is set for each area according to the state of change in wall thickness around the position where the presence or absence of a flaw is judged. Therefore, according to this aspect of the present invention, by avoiding the influence of electrical noise and ensuring the detection of oblique flaws, it becomes possible to more accurately detect harmful flaws on the surface of a metal pipe. Furthermore, according to this aspect of the present invention, since each judgment area is a two-dimensional area, even harmful defects that are diagonally continuous can be detected. [Brief description of the drawings]

[0010] [Figure 1] 1 is a diagram showing an example of the configuration of a flaw detection facility including a flaw determination device according to an embodiment based on the present invention; [Diagram 2] FIG. 13 is a diagram showing an example of expanded wall thickness data. [Diagram 3] 11A and 11B are diagrams showing examples of judgment areas and examples of minute defects; [Figure 4]FIG. 11 is a diagram illustrating an example of a method for setting a plurality of determination areas. [Diagram 5] FIG. 13 is a diagram showing an example of actual flaw thickness data. [Figure 6] FIG. 13 is a diagram showing an analysis image of actual flaw thickness data. [Figure 7] FIG. 13 is a diagram showing an example of thickness data when actual noise occurs. [Figure 8] This is an image of the analysis of thickness data when actual noise occurs. [Figure 9] FIG. 11 is a diagram showing examples of variables to be set and the determination results at that time. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Next, an embodiment of the present invention will be described with reference to the drawings. In this embodiment, an oil well tubular good and other steel pipes are used as examples of metal pipes to be evaluated for the presence or absence of defects. However, the metal pipes to which the present invention is applied are not limited to steel pipes.

[0012] (composition) The equipment is equipped with flaw detection equipment 1 as equipment for evaluating steel pipes. The flaw detection facility 1 of this embodiment includes a flaw detection device 10 and an evaluation device 11, as shown in FIG.

[0013] <Flaw detection device 10> The flaw detection device 10 uses an ultrasonic flaw detector placed close to the surface (outer surface) of the steel pipe to scan the surface of the steel pipe to be inspected in the circumferential and longitudinal (axial) directions, continuously detecting flaws around the entire surface of the steel pipe. The above-mentioned scan is performed, for example, in a spiral manner along the surface of the steel pipe. Also, for example, detection signals from the flaw detector are continuously acquired at a predetermined sampling period in synchronization with the relative movement of the flaw detector with respect to the outer surface of the steel plate, and output information is continuously obtained at predetermined scanning intervals based on the acquired detection signals and sequentially stored in a storage unit. Here, as long as the entire circumference of the surface portion of the steel pipe can be inspected, the method is not particularly limited. In other words, as long as the configuration is capable of acquiring wall thickness data in the circumferential and longitudinal directions of the steel pipe surface, i.e., in two-dimensional directions, the equipment of the inspection device 10 and the inspection method may be any known means. Furthermore, each piece of inspection data includes circumferential and longitudinal coordinates, i.e., two-dimensional position information on the surface of the steel pipe, in addition to wall thickness information.

[0014] <Evaluation device 11> The evaluation device 11 of this embodiment includes a wall thickness developed data acquisition unit 11A, a wall thickness judgment unit 11B, an area setting unit 11C, a judgment criterion setting unit 11D, and a flaw judgment unit 11E. The wall thickness developed data acquisition unit 11A, the area setting unit 11C, the judgment criterion setting unit 11D, and the flaw judgment unit 11E constitute the flaw judgment device for metal pipes of this embodiment.

[0015] [Wall thickness development data acquisition section 11A] The wall thickness expansion data acquisition unit 11A acquires two-dimensional wall thickness expansion data tDATA consisting of a group of wall thickness data arranged in the circumferential direction (pipe circumferential direction) and longitudinal direction (pipe axial direction) of the steel pipe from the flaw detection data obtained by the flaw detection device 10. As shown in FIG. 2, the expanded thickness data tDATA is two-dimensional data in which thickness data, which is data on thickness values, is arranged continuously in the circumferential direction and also in the longitudinal direction for each unit area TA.

[0016] In Figure 2, each square corresponds to a unit area TA. The pitch of the unit areas TA should be the same as the measurement pitch of the flaw detection in the circumferential and longitudinal directions. In this way, the expanded wall thickness data tDATA is a group of wall thickness data in which the wall thickness data for each position along the entire length of the steel pipe and the entire circumferential direction are expanded onto a two-dimensional plane. Here, Figure 2 shows an image of the expanded wall thickness data tDATA. In Figure 2(c), the vertical direction is the longitudinal direction of the pipe, and the horizontal direction is the circumferential direction of the pipe. In addition, the numbers 1 to 9 in the first row of the top row and the leftmost column in Figures 2 to 4 indicate the coordinates of the measurement points. The other numbers written on the matrix are actually numerical values ​​of the measured thickness. However, in this figure, integers 1 to 5 are used to show the principle of flaw determination, and the larger the number, the thicker the thickness.

[0017] [Thickness determination section 11B] The thickness determination unit 11B compares the thickness of each piece of flaw detection data with the upper limit thickness and the lower limit thickness set in advance to determine whether or not there is any thickness data that falls outside the allowable thickness range between the upper limit thickness and the lower limit thickness. The thickness determination unit 11B performs a determination based on the thickness itself.

[0018] [Area setting section 11C] The area setting unit 11C sets a judgment area DA consisting of unit areas TA that are continuous in the circumferential direction by m or more (m: an integer of 2 or more) and in the longitudinal direction by n or more (n: an integer of 2 or more) in the wall thickness development data tDATA. The judgment area DA is configured as a two-dimensional area of ​​"m x n" unit areas TA. In the following description, the case where m=3 and n=3 will be described as an example.

[0019] In Figure 3, an example of a unit area TA is shown as the area within a thick frame. In addition, Figure 3 also shows an image of defect judgment. The example in Figure 3(a) shows an example where a decrease in thickness occurs at only one point if the measurement results are believed at face value. However, upon closer examination, it is highly likely that the thickness is not significantly different from other parts and is indicated by electrical noise. On the other hand, depending on the setting of the allowable accuracy of the steel pipe wall thickness, the thickness at only one point may be reduced. Data indicated as such a defect may also be mistaken for a harmful defect. Figure 3(b) is an example where the defect extends in the circumferential direction. Figure 3(c) is an example where the defect extends in the longitudinal direction. In Figure 3, the shaded area is an example where a defect occurs in the unit area TA. From the viewpoint of detecting locally existing harmful minute defects (hereinafter also simply referred to as defects), m and n are preferably 3 or more and 10 or less, and more preferably 3 or more and 5 or less.

[0020] In addition, the area setting unit 11C sets a judgment area DA every time the thickness developed data tDATA is displaced from a predetermined position in the thickness developed data tDATA along the development direction of the thickness developed data tDATA. "In the development direction" means at least one of the circumferential direction and the longitudinal direction. In this way, the area setting unit 11C sets multiple judgment areas DA in the thickness developed data tDATA. In this case, the amount of displacement in the circumferential direction is preferably smaller than the above m and an integer value of 1 or more. The amount of displacement in the longitudinal direction is preferably smaller than the above n and an integer value of 1 or more. It is particularly preferable that the above m and n are 1. In this case, there is an overlap between adjacent determination areas DA.

[0021] In the area setting unit 11C of the present embodiment, for example, as shown in FIG. 4, the initial position of the judgment area DA is set such that the upper left end of the judgment area DA is the first row in the longitudinal direction and the first column in the circumferential direction of the wall thickness development data tDATA. Then, the judgment area DA is set sequentially each time the upper left end of the judgment area DA is shifted in the circumferential direction by one square (one unit area TA). When it is determined that the circumferential right end of the judgment area DA is at the right end position of the wall thickness development data tDATA, the judgment area DA is shifted downward by one square in the longitudinal direction and the left end of the judgment area DA is set to the left end position of the wall thickness development data tDATA. Then, as described above, the judgment area DA is set sequentially each time the judgment area DA is shifted in the circumferential direction by one square (one unit area TA). The above process is repeated until the lower right position of the judgment area DA is at the left end and the lower end of the wall thickness development data tDATA, thereby sequentially setting a plurality of judgment areas DA. In this way, a plurality of determination areas DA are set so as to cover the entire area of ​​the thickness development data tDATA. The determination areas DA may be set so that there is no overlap between them. Also, although it is preferable that the size of each determination area DA is uniform, they may be different.

[0022] [Judgment criteria setting section 11D] The judgment criterion setting unit 11D sets, for each judgment area DA, a judgment criterion value DS for judging the presence or absence of minute defects of a harmful depth from the wall thickness data present in the judgment area DA. The judgment criterion setting section 11D of this embodiment includes a reference value calculation section 11Da and a thickness steepness correction section 11Db.

[0023] [Reference value calculation unit 11Da] The reference value calculation unit 11Da calculates the average or median value of the thickness data within the target determination area DA, and sets the calculated value as the reference value. Here, it is preferable to set the number of minimum value exclusion points and the number of maximum value exclusion points based on the measurement noise (measurement accuracy) estimated from the accuracy of ultrasonic flaw detection. In this case, the reference value calculation unit 11Da performs a pre-processing to calculate the reference value. In the pre-processing, the minimum value exclusion points and the maximum value exclusion points are used to remove thickness data that may be measurement noise from the thickness data in the judgment area DA. In the pre-processing, for example, the thickness data in the target judgment area DA is rearranged in order of thickness, for example, in ascending order. Then, the minimum value exclusion points are removed from the minimum value side of the thickness data, and the maximum value exclusion points are removed from the maximum value side. Furthermore, the reference value calculation unit 11Da calculates the average value or median value of the remaining thickness data to obtain the reference value.

[0024] The minimum value exclusion score and the maximum value exclusion score may simply be a numerical value less than or equal to 1 / 4 of the number of unit areas TA present in one judgment area DA. For example, when m=3 and n=3, the number of unit areas TA present in the judgment area DA is 9, so the minimum value exclusion score and the maximum value exclusion score are set to 1 or 2. Strictly speaking, the frequency of noise varies depending on the wall thickness measurement settings (detection speed, frequency, etc.) and the characteristics of the steel pipe (dimensions, shape, surface dirt, etc.). For this reason, it may be difficult to uniquely determine the minimum and maximum exclusion points.

[0025] An example of a method for determining the minimum value exclusion points and the maximum value exclusion points will be shown. For example, the number of unit areas TA is 159 in the circumferential direction and 557 in the longitudinal direction, totaling 88,563, and noise occurs in an average of 99 locations from conventionally acquired data. Then, consider a case where an area of ​​about 0.1% of the total number of unit areas (88,563) is set as the determination area DA. In this example, the determination area DA can be expressed as a 3×3 area. In this case, since the probability of picking up noise in each determination area DA exceeds 1, it can be determined that it is better to set the minimum value exclusion points and the maximum value exclusion points to 2.

[0026] [Thickness steepness correction section 11Db] The thickness steepness correction unit 11Db performs a process of decreasing the reference value calculated by the reference value calculation unit 11Da by a thickness correction amount (steepness judgment value) corresponding to the thickness steepness amount due to harmful defects, and the corrected value is set as the judgment reference value DS for the judgment area DA. The thickness correction amount (steepness judgment value) is set in advance according to the depth of the flaw to be evaluated, etc. In the thickness steepness correction unit 11Db of this embodiment, the judgment reference value DS is calculated by the following formula. Judgment standard value DS = standard value - steepness judgment value (1) The steepness judgment value (thickness correction amount) is set to a value set as the steepness of the thickness based on the depth of the harmful flaw to be evaluated. For example, if the minimum depth of the harmful flaw to be detected is set to 0.3 mm, the steepness judgment value is set to 0.3 mm. In other words, the steepness of the thickness due to the harmful flaw is 0.3 mm.

[0027] The wall thickness correction amount may be set, for example, as a value obtained by multiplying the wall thickness by a steep judgment rate set based on the ratio of the depth of the flaw to be evaluated to the wall thickness of the metal pipe. This steep judgment rate is expressed as "(minimum depth of harmful flaw to be detected)÷wall thickness of steel pipe." The wall thickness of the steel pipe may be the minimum wall thickness allowed for the steel pipe, or may be the average wall thickness in the judgment area DA to be evaluated. When the steepness determination rate is used, the thickness correction amount is "reference value x steepness determination rate". Then, the thickness steepness correction section 11Db calculates the judgment reference value DS by the following formula. Judgment standard value DS = standard value - standard value × steepness judgment rate (2) = Reference value × (1 - steepness judgment rate)

[0028] Here, it is preferable to use formula (1) for thick-walled (wall thickness ≧ 5 mm) steel pipes, whereas it is preferable to use formula (2) for thin-walled (wall thickness < 5 mm) steel pipes, since the threshold value for steepness judgment may become too strict if formula (1) is applied. The steepness determination ratio is, for example, a value in the range greater than 0.0 and equal to or less than 0.10. In addition, the steepness judgment value and steepness judgment rate for setting the wall thickness correction amount may be optimized by repeatedly calculating using flaw detection data when a harmful flaw actually occurs so as to increase the detection accuracy.

[0029] Generally, defects that occur on steel pipes are divided into those that occur during rolling and those that occur at other times. Examples of defects that occur during rolling include those shown in (b) and (c) of Figure 3, which are caused by friction between the rolled material, the steel pipe, and tools such as rolls. In some cases, defects may occur diagonally. In addition, cracks may occur along grain boundaries. Examples of defects that occur due to reasons other than rolling include dents that occur during transportation. The inventors have found that, although there are various types of defects that usually occur, even if the defects have directionality or have little directionality like dents, they are unlikely to become extremely deep holes locally. From the above, it has been found that a thickness distribution in which the thickness is locally reduced at only one point, as shown in (a) of Figure 3, that is, a thickness distribution that changes locally with respect to the surrounding area, is unlikely to occur in the actual manufacturing process.

[0030] In addition, in the measurement, for example, one measurement area may be set to 1 mm in the circumferential direction and 10 mm in the axial direction, and the measurement location may be moved in a spiral pattern with a pitch of 1 mm in the circumferential direction and 5 mm in the axial direction. If the difference between the measured values ​​of adjacent wall thicknesses is measured as being locally excessive only in the relevant portion, or if the measured portion is measured as being locally thin by 20% or more of the nominal wall thickness, it is considered that this does not occur as a defect in the manufacturing process. Therefore, these can be judged as noise. Conversely, for fluctuations of less than 20%, the measured value can be used to detect defects regardless of noise or defects. This judgment method requires measuring the entire circumference of the pipe, and it is possible to set the measurement area of ​​the wall thickness to 8 mm in the circumferential direction and 20 mm in the axial direction, etc., and set the measurement location according to the conditions of the measurement area.

[0031] When making a more detailed judgment of whether or not a flaw exists, or of the type of flaw, regarding these criteria, the judgment can be supplemented with data accumulated during the manufacturing process. For example, a large amount of data on the occurrence of flaws that have actually occurred in the target steel pipe is obtained. Then, from the obtained data, a correlation between the flaw depth and the steepness judgment value or steepness judgment rate that makes the above formula (1) or (2) valid is obtained, and the steepness judgment value or steepness judgment rate is obtained and set from the correlation.

[0032] [Defect determination section 11E] The flaw determination unit 11E determines whether or not a flaw is present by comparing each thickness data within the determination area DA for which the determination standard value DS has been determined by the determination standard setting unit 11D with the determined determination standard value DS. That is, the flaw determination unit 11E compares each thickness data in the target determination area DA with the above-obtained determination reference value DS, and performs processing to determine that there is a harmful flaw in the unit area TA where "thickness data < determination reference value DS".

[0033] Here, if the number of flaws present within the judgment area DA (the number of unit areas TA judged to have flaws) is two or more (if the flaws are of a predetermined size or larger), it may be judged that there is a harmful flaw within the judgment area DA. In this case, the number of detections that are judged to be harmful flaws may be preset as a judgment score. The judgment score is set, for example, according to the size of the harmful flaws to be detected, and other conditions. In this embodiment, the presence or absence of a defect can be determined for each unit area TA individually. Therefore, based on the position coordinates of the unit area TA, unit areas TA that have been determined to have a defect may be determined to be harmful defects if the unit areas TA are continuously present in the longitudinal direction, circumferential direction, or diagonal direction.

[0034] (Operation etc.) In this embodiment, when determining the presence or absence of a flaw, a judgment area DA consisting of a two-dimensional surface area is set. The judgment area DA is set as an area in which two or more unit areas TA are consecutive in the circumferential direction and the longitudinal direction. Then, for each judgment area DA, a judgment reference value DS is set based on the wall thickness within the judgment area DA. For this reason, in this embodiment, the judgment reference value DS is set according to the state of change in wall thickness around the position where the presence or absence of a flaw is judged. That is, in this embodiment, it is possible to set a judgment reference value DS suitable for each judgment area DA where flaw judgment is performed. As a result, in this embodiment, the judgment reference value DS can be set with greater accuracy. Therefore, in this embodiment, it is possible to detect harmful flaws on the surface of the metal pipe with greater accuracy.

[0035] In this embodiment, each judgment area DA is a two-dimensional area, so even harmful defects that continue diagonally can be detected. In this embodiment, it is also possible to determine whether a defect is harmful or not from the number of defects that exist in a local area, such as within each judgment area DA that is set as a two-dimensional area. In this embodiment, the presence or absence of defects on the inner surface of the steel pipe is determined from the relative value of the wall thickness, in addition to the determination of whether the wall thickness is within the standard range based on the absolute wall thickness measurement by the wall thickness determination unit 11B. In other words, the determination of defects is not an evaluation method that replaces the wall thickness measurement.

[0036] In this embodiment, the judgment areas DA are set so that adjacent judgment areas DA overlap each other, which results in further improved accuracy in flaw detection. The reason for this will be explained. The flaw to be detected is recognized as a portion where the thickness changes steeply with respect to the surrounding surface. In other words, the recognition of the thickness change steepness due to the flaw differs depending on the location of the flaw and the thickness of the area surrounding the flaw. For this reason, even if a flaw is present within one judgment area DA, it may not be judged as a flaw depending on the thickness data of the same judgment area DA. However, in other judgment areas DA that contain the same flaw, the thickness conditions around the flaw may be different, and the flaw may be detected as a flaw. As a result, the accuracy of flaw detection is improved.

[0037] In particular, in this embodiment, by setting a judgment reference value DS for each judgment area DA based on the wall thickness data present in the judgment area DA, it is possible to further improve the accuracy of flaw detection. In this embodiment, there is a possibility that the measurement noise estimated from the accuracy of ultrasonic inspection is included in the thickness data present in the judgment area DA. Therefore, the judgment reference value DS is obtained from the remaining thickness data after removing one or more minimum values ​​and one or more maximum values ​​from the thickness data within the judgment area DA. This improves the accuracy of the judgment reference value DS.

[0038] (Example) Next, examples based on this embodiment will be described. When the thickness of a steel pipe is measured over its entire length and circumference using ultrasonic waves, the thickness data is obtained. The thickness data for each position is expanded to become the expanded thickness data tDATA. As mentioned above, the expanded thickness data tDATA is a two-dimensional display of the thickness data, as shown in Figure 2(c). As described above, the case of m=3 and n=3 will be described as an example. Since the wall thickness of the steel pipe varies, in this embodiment, flaws are judged for each judgment area DA. By setting the judgment area DA as a part of the entire area of ​​the entire length and circumference of the steel pipe, the effect of the change in wall thickness of the entire steel pipe can be reduced.

[0039] If the judgment area DA is free of flaws and has the same thickness (e.g., 5 mm), the reference value, which is the average or median value, will be a value equivalent to the thickness (e.g., 5 mm).On the other hand, if a unit area TA with a flaw exists within the judgment area DA and its thickness is, for example, 3 mm, the reference value will be approximately 4.7 mm, making it possible to detect the unit area TA with a flaw. However, in reality, there is variation in the thickness of the flaw-free portion, so in this embodiment, a value that is smaller than the reference value by the thickness correction amount is set as the determination reference value DS for threshold determination.

[0040] Figure 5 shows an example of thickness data of actual harmful defects. The numbers in the figure indicate the thickness value (mm) of each unit area TA. In this example, the judgment area DA was extracted at three points in the circumferential direction and three points in the tube axial direction. The steel pipe in question has an outer diameter of 139.8 mm (±0.8%) and a nominal wall thickness of 15.0 mm (±10%). In other words, the wall thickness tolerance is 13.5 mm≦wall thickness≦16.5 mm. Therefore, the minimum wall thickness within the judgment area DA is 13.72 mm. Therefore, all wall thickness data within the judgment area DA passes the wall thickness tolerance.

[0041] Also, Fig. 6 shows an image of the analysis of the thickness data of the actual flaws, rather than the measured values. In Fig. 6, the scans are (1), (2), and (3) in the circumferential direction, and (4), (5), and (6) in the longitudinal direction. Scans (1) and (6) in Fig. 6 detect 13.72 mm, and scans (2 and (5) detect 13.92 mm. Figure 7 shows the thickness data obtained by measuring the same area, but with noise during measurement. In Figure 7, there is a data of 13.72 mm at the center, but this is not a correct value due to noise; the correct value is 15.05. An image of the analysis of thickness data when actual noise occurs is shown in Fig. 8. Scans (2) and (5) in Fig. 8 detect 13.72 mm.

[0042] Here, in the method of Patent Document 1, in order to detect the flaw in Fig. 6, it is necessary to set the threshold value for determining that the flaw is harmful at 13.72 mm. In that case, the noise in Fig. 8 would also be determined to be harmful, so there is a risk that it would not be possible to determine that only the flaw is harmful. Furthermore, in the method described in Patent Document 2, since the defective parts are not continuous in any of the cases shown in FIG. 8, there is a risk that defects that cannot be determined to be harmful may not be detected. As described above, the methods described in Patent Documents 1 and 2 may not be able to detect a defect (defective portion) that is oblique. In contrast, in this embodiment, the judgment reference value DS calculated from the thickness within the judgment area DA is used as a threshold value, and the judgment of defects in each unit area TA is performed regardless of the scanning direction shown in Figure 8, etc., and if the defect (defective part) is oblique, it is possible to judge the defect area as a defect.

[0043] Next, a method of judging flaws in an embodiment based on this embodiment will be described. In this embodiment, the number of unit areas TA is expressed by points. Here, the judgment area DA is set as m: the number of points in the circumferential direction (points), and n: the number of points in the axial direction (points). In addition, the allowable number of defects in the judgment area DA is set as a: the number of steepness judgment points (points). In addition, b: the number of minimum value excluded points (points), c: the number of maximum value excluded points (points), e: steepness judgment value (%), and f: steepness judgment value (mm). Here, a can be an integer of 1 or more, b and c can be integers of 1 or more, and e and f can be positive numbers.

[0044] These variables are set so that the defects that actually occur can be detected from the thickness data shown in Fig. 5. Here, it is assumed that the values ​​of each variable are set as shown in Fig. 7. In ultrasonic testing, noise may occur when the inspection signal is sent from the probe to the flaw detector. This noise is transmitted as several abnormal values ​​in the maximum and minimum wall thickness directions. As most of the noise is electrical noise, there is no continuity and only one abnormal value is transmitted, but noise caused by poor water film formation between the probe and steel pipe may be transmitted as several or more consecutive values. Therefore, on the other hand, depending on the equipment layout of the flaw detector and the properties of the steel pipe, noise may not occur and it may not be necessary to exclude minimum and maximum values. Therefore, by setting b: minimum value exclusion points (points) and c: maximum value exclusion points (points) to integers greater than or equal to 0, optimization can be performed according to the equipment and circumstances.

[0045] In addition, depending on the rolling process of steel pipes, there are some with high elongation rates and some with low elongation rates, and the shapes of harmful defects to be detected, such as length, width, and depth, also differ accordingly. Therefore, a: number of steepness judgment points (points), e: steepness judgment value (%), and f: steepness judgment value (mm) are appropriately set according to the steel pipe and defect to be evaluated. The data used in the calculation of this example are the actual flaw thickness data in Fig. 5 and the actual thickness data when noise occurs in Fig. 7. It was found that when the determination process is performed using the method of this embodiment, flaws can be detected and noise is not determined as a defect and is not over-detected. The calculation results for each level are shown below. The judgment method is to calculate using formula 1 and formula 2, and if it is judged as poor by either formula, it is considered as poor.

[0046] Next, specific calculation results when the settings shown in FIG. 7 are used will be shown. First, the calculation results for the thickness data of the actual flaw in Fig. 5 are as follows. b=1 eliminates 15.52, and c=1 eliminates 13.72. If the above formula (2) is used to calculate the judgment standard value DS, the following calculation results. Judgment criteria value DS = {14.96 + 15.04 + 14.96 + 13.92 + 15.10 + 15.02 + 15.07} ÷ 7 -{6.95+6.96+6.60+7.08+6.96+6.99+7.11}÷7×0.02 =14.87-0.30=14.57 5, out of the nine points, two points are equal to or less than the judgment standard value DS=14.57. In this example, since the steep judgment score a: is set to 2 (points) or more, it is judged as defective.

[0047] Moreover, if the above formula (1) is used to calculate the judgment reference value DS, the following calculation results. Criterion value DS = {14.96 + 15.04 + 14.96 + 13.92 + 15.10 + 15.02 + 15.07} ÷ 7 -0.30 =14.87-0.30=14.57 5, two out of nine points are below the judgment standard value DS=14.57. Therefore, they are also judged as defective.

[0048] Next, the calculation results of the thickness data when the actual noise occurs in Figure 7 are as follows. b=1 eliminates 15.04, and c=1 eliminates 13.72. If the above formula (2) is used to calculate the judgment standard value DS, the following calculation results. Judgment criteria value DS = {14.98+14.98+15.04+14.96+14.98+14.98+14.96}÷7 -{6.95+6.96+6.60+7.08+6.96+6.99+7.11}÷7×0.02 =14.98-0.30=14.68 7, the number of points that is equal to or less than the judgment standard value DS=14.68 is 1 out of 9 points. In this example, the number of steep judgment points a: is 2 (points) or more, so the product is judged as non-defective.

[0049] Moreover, if the above formula (1) is used to calculate the judgment reference value DS, the following calculation results. Criterion value DS = {14.98+14.98+15.04+14.96+14.98+14.98+14.96}÷7 -0.30 =14.98-0.30 =14.68 7, the number of points that is equal to or less than the judgment standard value DS=14.68 is 1 out of 9 points. In this example, the number of steep judgment points a: is 2 (points) or more, so the product is judged as non-defective.

[0050] The number of steepness judgment points (points) a is a value set according to the condition of the flaw to be detected, and in this example, a is set to 2, since two consecutive points are evaluated as a flaw to be detected. As described above, according to the defect determination method of the present invention, defects can be detected more accurately by determining the surface (an area defined by an integer of 2 or more vertically and an integer of 2 or more horizontally) and by excluding one or more maximum and minimum values ​​within the determination area DA in order to separate defects from noise.

[0051] (others) The present disclosure may also have the following configuration. (1) A flaw detection device for metal pipes that detects the presence or absence of flaws on the surface of a metal pipe, comprising: a wall-thickness development data acquisition unit that acquires two-dimensional wall-thickness development data consisting of wall-thickness data for each unit area arranged in the circumferential and longitudinal directions of the metal pipe from inspection data acquired by performing ultrasonic inspection in two-dimensional directions, longitudinal and circumferential, along the surface of the metal pipe; an area setting unit that sets a judgment area consisting of a region in which the unit area is continuous for m or more (m: an integer of 2 or more) in the circumferential direction and n or more (n: an integer of 2 or more) in the longitudinal direction in the thickness development data; a judgment criterion setting unit that sets a judgment criterion value for judging defects based on the wall thickness data present in the judgment area; a flaw determination unit that determines whether or not a flaw exists by comparing each thickness data in the determination area in which the determination reference value is set with the determination reference value; A defect detection device for metal pipes comprising: (2) the area setting unit sets a judgment area each time the sensor element is displaced in at least one of the circumferential direction and the longitudinal direction; The amount of displacement in the circumferential direction is an integer value of 1 or more that is smaller than m, The amount of displacement in the upper longitudinal direction is an integer value of 1 or more and smaller than the above n. (3) The judgment criterion setting unit sets the judgment criterion value to a value obtained by subtracting a preset steep judgment value from a reference value that is the average or median value of the thickness data within the target judgment area. (4) The steepness judgment value is set based on the depth of the flaw to be evaluated. (5) The steepness judgment value is set as a value obtained by multiplying the reference value by a steepness judgment rate that is set based on the ratio of the depth of the flaw to be evaluated to the wall thickness of the metal pipe. (6) setting the minimum and maximum exclusion points based on the measurement noise estimated from the accuracy of the ultrasonic flaw detection; The above-mentioned judgment reference value is determined by arranging the thickness data within the target judgment area in thickness order, excluding the above-mentioned minimum value exclusion points from the minimum value side and the above-mentioned maximum value exclusion points from the maximum value side, and then determining the above-mentioned reference value from the remaining thickness data. (7) A flaw detection method for determining the presence or absence of flaws on a surface of a metal pipe, comprising the steps of: From the inspection data obtained by performing ultrasonic inspection in two-dimensional directions, namely, in the longitudinal direction and the circumferential direction along the surface of the metal pipe, two-dimensional wall thickness development data consisting of wall thickness data for each unit area arranged in the circumferential direction and the longitudinal direction of the metal pipe is obtained; In the thickness expansion data, a plurality of judgment areas are set in the thickness expansion data, each of which is a region in which the unit area is continuous for m or more in the circumferential direction (m: an integer of 2 or more) and n or more in the longitudinal direction (n: an integer of 2 or more); For each judgment area, a judgment reference value for judging flaws is set based on the wall thickness data present in that judgment area, and the set judgment reference value is compared with each wall thickness data in that judgment area to judge whether or not a flaw exists. A method for determining defects in a metal pipe comprising: (8) A judgment area is set for each displacement in at least one of the circumferential direction and the longitudinal direction, thereby setting a plurality of judgment areas in the wall thickness development data; The amount of displacement in the circumferential direction is an integer value of 1 or more that is smaller than m, The amount of displacement in the upper longitudinal direction is an integer value of 1 or more and smaller than the above n. (9) The judgment reference value is determined by subtracting a preset steep judgment value from a reference value that is the average or median value of the thickness data within the target judgment area. (10) The steepness judgment value is set based on the depth of the flaw to be evaluated. (11) The steepness judgment value is set as a value obtained by multiplying the reference value by a steepness judgment rate that is set based on the ratio of the depth of the flaw to be evaluated to the wall thickness of the metal pipe. (12) In order to remove noise near the maximum and minimum thicknesses, one or more thickness data items with minimum thickness values ​​are removed from the thickness data within the target judgment area based on the measurement noise estimated from the accuracy of the ultrasonic inspection, and one or more thickness data items with maximum thickness values ​​are removed, and the above-mentioned reference value is obtained from the remaining thickness data. [Explanation of symbols]

[0052] 1 Flaw detection equipment 10 Flaw detection equipment 11 Evaluation equipment 11A Thickness expansion data acquisition section 11B Thickness determination section 11C Area setting section 11D Judgment criteria installation section 11Da Reference value calculation unit 11Db Thickness steepness correction section 11E Defect Judgment Section DA Judgment Area DS Criteria TA Unit Area tDATA Wall thickness expansion data

Claims

1. A metal pipe defect detection device for determining the presence or absence of defects on a metal pipe surface, comprising: a wall-thickness development data acquisition unit that acquires two-dimensional wall-thickness development data consisting of wall-thickness data for each unit area arranged in the circumferential and longitudinal directions of the metal pipe from inspection data acquired by performing ultrasonic inspection in two-dimensional directions, namely, the longitudinal and circumferential directions, along the surface of the metal pipe; an area setting unit that sets a judgment area of ​​the wall thickness development data, the unit area being a continuous area of ​​m or more (m: an integer of 2 or more) in the circumferential direction and n or more (n: an integer of 2 or more) in the longitudinal direction; a judgment criterion setting unit that sets a judgment criterion value for judging defects based on the wall thickness data present in the judgment area; a flaw determination unit that determines whether or not a flaw exists by comparing each thickness data in the determination area in which the determination reference value is set with the determination reference value; A metal pipe defect detection device comprising:

2. the area setting unit sets a determination area each time the sensor element is displaced in at least one of the circumferential direction and the longitudinal direction; The amount of displacement in the circumferential direction is an integer value of 1 or more that is smaller than m, The amount of displacement in the upper longitudinal direction is an integer value of 1 or more that is smaller than the above n.

2. A flaw detection device for metal pipes according to claim 1.

3. The judgment criterion setting unit sets the judgment criterion value to a value obtained by subtracting a preset steep judgment value from a reference value that is an average value or a median value of the thickness data within the target judgment area.

2. A flaw detection device for metal pipes according to claim 1.

4. The steepness judgment value is set based on the depth of the flaw to be evaluated.

4. A flaw detection device for metal pipes according to claim 3.

5. The steepness judgment value is set as a value obtained by multiplying the reference value by a steepness judgment rate set based on the ratio of the depth of the flaw to be evaluated to the wall thickness of the metal pipe.

4. A flaw detection device for metal pipes according to claim 3.

6. Based on the measurement noise estimated from the accuracy of the ultrasonic flaw detection, a minimum value exclusion point number and a maximum value exclusion point number are set, The above-mentioned judgment reference value is obtained by arranging the wall thickness data in the target judgment area in thickness order, excluding the minimum value exclusion points from the minimum value side and excluding the maximum value exclusion points from the maximum value side, and determining the above-mentioned reference value from the remaining wall thickness data. A flaw detection device for a metal pipe according to any one of claims 3 to 5.

7. A flaw detection method for determining the presence or absence of flaws on a metal pipe surface, comprising the steps of: From the inspection data obtained by performing ultrasonic inspection in two-dimensional directions, namely, in the longitudinal direction and the circumferential direction along the surface of the metal pipe, two-dimensional thickness development data is obtained, which is composed of thickness data for each unit area arranged in the circumferential direction and the longitudinal direction of the metal pipe; A plurality of judgment areas are set in the thickness expansion data, each of which is a continuous area of ​​the unit area of ​​m or more (m: an integer of 2 or more) in the circumferential direction and n or more (n: an integer of 2 or more) in the longitudinal direction; For each judgment area, a judgment reference value for judging flaws is set based on the wall thickness data present in that judgment area, and the set judgment reference value is compared with each wall thickness data in that judgment area to judge whether or not a flaw exists. A method for determining defects in a metal pipe comprising:

8. A determination area is set for each displacement in at least one of the circumferential direction and the longitudinal direction, thereby setting a plurality of determination areas in the wall thickness development data; The amount of displacement in the circumferential direction is an integer value of 1 or more that is smaller than m, The amount of displacement in the upper longitudinal direction is an integer value of 1 or more that is smaller than the above n. A method for determining defects in a metal pipe according to claim 7.

9. The judgment reference value is determined by subtracting a preset steep judgment value from a reference value consisting of an average value or a median value of the wall thickness data within the target judgment area. A method for determining defects in a metal pipe according to claim 7.

10. The steepness judgment value is set based on the depth of the flaw to be evaluated. A method for determining defects in a metal pipe according to claim 9.

11. The steepness judgment value is set as a value obtained by multiplying the reference value by a steepness judgment rate set based on the ratio of the depth of the flaw to be evaluated to the wall thickness of the metal pipe. A method for determining defects in a metal pipe according to claim 9.

12. Based on the measurement noise estimated from the accuracy of the ultrasonic flaw detection, one or more thickness data on the minimum value side and one or more thickness data on the maximum value side are removed from the thickness data in the target judgment area, and the reference value is obtained from the remaining thickness data. A method for determining defects in a metal pipe according to any one of claims 9 to 11.

Images