Inspection system, judgment processing system, judgment processing device and inspection method

Abstract

This invention provides an inspection system, a judgment processing system, a judgment processing device, and an inspection method. In non-destructive inspection of defects in welded joints of piping or piping components, it improves the work efficiency of radiation transmission testing and enhances inspection accuracy by reducing the operator's workload. The system acquires imaging data by transmitting radiation through the welded joint of the piping to be inspected. Furthermore, based on the distribution of radiation transmission intensity obtained from the imaging data, processing is performed to correlate judgment data with image data representing the distribution of radiation transmission intensity. The judgment data indicates the result of determining defects in the welded joint of the piping to be inspected. Therefore, image data and corresponding judgment data can be obtained using the imaging data, reducing the operator's workload.

Description

Technical Field

[0001] This invention relates to a technique for non-destructive inspection of welded parts of piping, and more particularly to an inspection system, a judgment processing system, a judgment processing device, and an inspection method. Background Technology

[0002] The fluid processing plant includes natural gas plants, petroleum refining plants, and chemical plants. The natural gas plant liquefies natural gas or separates and recovers liquefied natural gas. The petroleum refining plant distills or desulfurizes crude oil or various intermediate products. The chemical plant produces petrochemical products or intermediate chemicals, polymers, etc.

[0003] These factories are structured, for example, as follows: a group of machines including static machines such as towers or heat exchangers, and dynamic machines such as pumps. Furthermore, the machines constituting these groups are connected by piping for fluid exchange.

[0004] In the construction of such plants, welding is sometimes used for the connection of the pipes to each other or to pipe components such as flanges and elbows. If the welded joint contains damage of size or shape exceeding the permissible limit (hereinafter, harmful damage exceeding the permissible limit and deemed unacceptable is referred to as a "defect"), it will become a major cause of trouble such as fluid leakage or weld breakage after the plant is put into operation. Therefore, after the welding of pipes or pipe components, in order to determine whether the welded joint contains defects, a radiation transmission test is performed at a sampling frequency set according to multiple factors such as fluid and pressure experienced by the reference pipes or pipe components.

[0005] Previously, in conducting radiation transmission tests on welded parts, the welded part was irradiated with radiation such as gamma rays or X-rays. The radiation passing through the welded part was used to expose an industrial film placed behind the welded part, thereby investigating whether there was any damage or defects in the welded part. The presence or absence of damage or defects was determined by a qualified person visually inspecting the film in a darkroom.

[0006] Here, the imaging of welded sections using radiation requires ensuring safe distances, and within the limited construction area, the number of imaging operations that can be carried out simultaneously is limited. Furthermore, the time required for arranging each imaging site necessitates consideration of the overall impact on the plant's construction schedule. Moreover, the sheer number of welded sections during plant construction necessitates a correspondingly large workforce for handling the membranes. Furthermore, the management and disposal of the developing solution used to develop the membranes require environmental considerations.

[0007] In order to solve the aforementioned problems, it is necessary to establish a welding inspection technique that can improve work efficiency and further reduce the burden on operators.

[0008] Patent Document 1 describes a radiation inspection device that, in a factory piping system, uses a detector to detect radiation transmitted through the piping from a radiation source to the inspected object, thereby checking the wall thickness of the piping. It also describes a structure in which the radiation source and detector are rotated relative to the piping, acquiring transmitted images at different rotation angles. Furthermore, Patent Document 2 describes a maintenance support device that measures the thinning depth of a structure in a factory, obtains a standard constant from the measured value, and, based on the standard constant, calculates the probability that the structure will become unusable for a specified service life and combination of plate thicknesses.

[0009] Thus, the previous method of not using membranes to detect radiation transmittance was only used for thinning management of piping in operating factories.

[0010] Existing technical documents

[0011] Patent documents

[0012] Patent Document 1: Japanese Patent Application Publication No. 2009-47424

[0013] Patent Document 2: Japanese Patent Application Publication No. 2018-25497 Summary of the Invention

[0014] The problem that the invention aims to solve

[0015] The present invention provides a technique that improves the efficiency of radiation transmission testing and enhances inspection accuracy by reducing the workload of operators in the non-destructive inspection of defects in welded parts of piping or piping components.

[0016] Technical means to solve the problem

[0017] The inspection system of the present invention is a non-destructive inspection system for defects in welded parts of piping or piping components, comprising:

[0018] The imaging data acquisition unit acquires imaging data obtained by passing radiation through the welded parts of the piping or piping components of the object being inspected; and

[0019] The determination processing unit performs processing based on the distribution of the transmitted intensity of the radiation obtained from the imaging data to correspond determination data with image data representing the distribution of the transmitted intensity of the radiation, wherein the determination data represents the result of determining a defect in the weld of the inspected object.

[0020] Furthermore, the inspection system may also include the following features.

[0021] (a) The imaging data acquisition unit includes a radiation irradiation unit and a detector arranged facing each other across the weld portion of the object under inspection. The radiation irradiation unit irradiates radiation onto the weld portion, the detector detects the radiation that passes through the weld portion, and the imaging data acquisition unit acquires the imaging data in the form of digital data based on the transmission intensity of the radiation detected by the detector.

[0022] (b) The image data acquisition unit includes:

[0023] A clamp for arranging the radiation irradiation unit and the detector facing each other across the welded portion of the object being inspected; and

[0024] The moving mechanism, while maintaining the opposing arrangement of the clamps, synchronizes and moves the radiation irradiation unit and the detection unit in the circumferential direction of the welding unit, thereby moving the position irradiating the radiation in the circumferential direction.

[0025] The image data was acquired at multiple different locations along the circumference of the welded part.

[0026] (c) The inspection system includes an image processing unit that combines multiple images acquired at different locations along the circumferential direction of the welded part to create the image data.

[0027] (d) Irradiate the area containing the weld portion of the object under inspection with radiation, convert the analog image obtained by exposing the membrane to the radiation image transmitted through the weld portion into a digital image, and use the digital image as the imaging data.

[0028] (e) The determination data includes: information corresponding to the location, size, and shape characteristics of the weld line damage contained in the identifiable image data, and information indicating the type of defect, wherein the defect is damage to the weld portion that is determined to be unqualified or harmful.

[0029] The inspection system includes an automatic judgment unit that, based on machine learning results of damage contained in previous image data corresponding to the judgment data, performs temporary judgment data corresponding to new image data showing the distribution of radiation transmission intensity. The temporary judgment data determines the location, size, and shape characteristics of the weld line damage, thereby automatically determining the type of the anticipated defect.

[0030] The determination processing unit makes a determination based on the temporary determination data corresponding to the image data generated by the automatic determination processing unit, thereby performing processing to make the determination data correspond to the image data.

[0031] (f) The determination processing unit is configured to be subject to correction processing based on the temporary determination data corresponding to the image data by the automatic determination unit.

[0032] The determination processing apparatus of the present invention performs non-destructive inspection of defects in welded parts of piping or piping components, wherein the apparatus is configured such that...

[0033] Based on the distribution of radiation transmission intensity obtained by passing radiation through the welded portion of the piping or piping component of the inspection object, a process is performed to correspond determination data with image data representing the distribution of radiation transmission intensity, wherein the determination data represents the result of determining a defect in the welded portion of the piping of the inspection object.

[0034] The inspection method of the present invention is a non-destructive inspection method for defects in welded parts of piping or piping components, comprising the following steps:

[0035] Based on the distribution of radiation transmission intensity obtained by allowing radiation to pass through the welded portion of the piping or piping component of the inspection object, processing is performed to correspond determination data with image data representing the distribution of radiation transmission intensity, wherein the determination data represents the result of determining a defect in the welded portion of the inspection object.

[0036] The effects of the invention

[0037] This invention provides non-destructive inspection of weld defects in piping or piping components by acquiring imaging data obtained by allowing radiation to pass through the weld of the piping or piping component under inspection. Furthermore, the imaging data is used to determine defects in the weld of the inspected object, and processing is performed to correlate determination data representing the determination result with image data representing the distribution of radiation transmission intensity. Therefore, the determination result of the weld can be obtained without using a membrane, thereby improving the work efficiency of radiation transmission testing by reducing the workload of the operator. Moreover, by using image data, inspection accuracy and efficiency can be improved compared to visual inspection using a membrane. Attached Figure Description

[0038] Figure 1 This is a structural diagram of an inspection system for non-destructive inspection of welded parts in piping.

[0039] Figure 2 A side view of the data acquisition unit.

[0040] Figure 3 A process diagram illustrating an example of inspecting the welded parts of a piping system.

[0041] Figure 4 This diagram illustrates the methods for acquiring shooting data and creating image data.

[0042] Figure 5 This is an explanatory diagram showing the correspondence between image data and judgment data.

[0043] Figure 6 An illustrative diagram illustrating an example of defects contained in image data.

[0044] Figure 7 A process diagram illustrating an example of the automatic decision-making process of the automatic decision-making unit.

[0045] Figure 8 An explanatory diagram illustrating an example of the structure of temporary determination data obtained by the automatic determination unit.

[0046] Figure 9 A process diagram illustrating an example of inspecting welded parts of piping that includes automatic judgment processing.

[0047] Figure 10 Here is a structural diagram to illustrate another example of an inspection system.

[0048] Explanation of symbols

[0049] 1: Shooting Data Acquisition Department

[0050] 3: Judgment and Processing Department

[0051] 9: Image data

[0052] 90: Shooting data

[0053] 81: Welding section

[0054] 91: Image data analysis complete.

[0055] 100: Piping Detailed Implementation

[0056] Figure 1 The overall structure of an inspection system for a radiation transmission test indicating defects in welded parts of piping or piping components. In this embodiment, the radiation transmission test of welded parts of piping or piping components is performed based on the results of radiographic imaging. Furthermore, based on image data obtained from the imaging, which shows the distribution of radiation transmission intensity, the presence or absence of damage in the welded part is confirmed. If damage is found, it is determined whether it constitutes a "defect," which is damage that results in an unacceptable or harmful weld.

[0057] For example, this technology is applicable to non-destructive inspection of welded parts during the construction and maintenance of piped plants that transport fluids via piping. There are no particular limitations on the types of plants to which this technology is applicable; it can be used for non-destructive inspection of welded parts in the following types of plants: natural gas plants, where natural gas liquefaction or separation and recovery of liquid natural gas is carried out; petroleum refining plants, where crude oil or various intermediate products are distilled or desulfurized; chemical plants, where petrochemical products or intermediate chemicals, polymers, etc., are produced; pharmaceutical plants, where pharmaceuticals or their intermediate products are produced; and waste treatment plants handling low-level radioactive waste.

[0058] Piping that is applicable to this technology can be connected by welding and subjected to radiation transmission tests, and there are no particular limitations on its materials, pipe diameter, wall thickness, or the presence or absence of an inner lining. Moreover, the fluid flowing in the piping can be a free-flowing powder or granular substance (powder, granules, or particles, etc.) in addition to gas and liquid.

[0059] Furthermore, the piping components to which this technology is applicable are parts other than piping used when connecting machine rooms in a factory using piping. Examples of piping components include flanges or elbows, T-joints, and pipe joints with different outer diameters. Regarding piping components, as long as they are connected by welding and a radiation transmission test can be performed, there are no particular limitations on their materials or dimensions, wall thickness, or the presence or absence of lining.

[0060] The welded part that is the object of the radiation transmission test can be any of the following: a welded part between pipes, a welded part between pipes and pipe components, or a welded part between pipe components and pipe components.

[0061] Next, the structural example of the inspection system in this case will be described. The inspection system includes: an image data acquisition unit 1 (1A-1C) that transmits radiation such as gamma rays or X-rays through the welded parts of the pipes or pipe components of the object under inspection and acquires image data in the form of digital data. It also includes: a judgment processing unit 3 that, based on image data representing the distribution of radiation transmission intensity obtained from the image data, generates judgment data indicating a defect in the welded parts of the pipes or pipe components of the object under inspection, and performs processing to correlate the judgment data with the image data. For example, the judgment processing unit 3 includes a computer.

[0062] Here, the damage contained in the weld 81 is equivalent to a defect, and the determination of whether the welding result is unqualified is carried out by a qualified person certified by a public or private institution.

[0063] In this regard, the inspection system in this example includes: an automatic judgment unit 5, which provides temporary judgment data to support the judgment when a qualified person makes a judgment. The automatic judgment unit 5 includes: a storage unit 51, which stores image data and judgment data; and a calculation unit 52, which performs machine learning related to the image information of the damage, and automatically determines the type of defect contained in the image data based on the results of the machine learning. For example, such as... Figure 1 As shown, the automatic determination unit 5 can be provided as a cloud computing function and is configured to be accessible by the determination processing unit 3 via the Internet or the like.

[0064] Figure 2 This represents an example of the structure of the data acquisition unit 1. Figure 2 The following example illustrates the acquisition of photographic data relating to the portion where the ends of two pipes 100 are connected to each other via a weld 81. The photographic data acquisition unit 1 includes: a radiation irradiation unit 11 that irradiates radiation onto the weld 81 of the pipes 100; and a detector 12 that detects the radiation transmitted through the weld 81. Figure 2 The symbol 13 in the text refers to a device or power source for irradiating radiation from the radiation irradiation unit 11. In the imaging data acquisition unit 1 of this example, a flat panel detector (FPD) can be used for the detector 12. This flat panel detector can detect radiation with a screen-shaped detector without using a membrane. Corresponding to the coordinates set in the detection surface of the detector, it outputs digital data representing the amount of radiation detected as imaging data.

[0065] As described above, the two pipes 100 under inspection are connected by butt joints between the ends of one pipe 100 and the other pipe 100 via a weld 81. This weld 81 is performed along the entire circumference of the pipe 100. Furthermore, Figure 2 The example describes the welded joint 81 that connects two pipes 100 to each other, but the image acquisition unit 1 in this example can also be used to inspect the welded joint 81 of pipe 100 and flange, or pipe 100 and elbow, flange and elbow, etc.

[0066] like Figure 2 As shown, the radiation irradiation unit 11 and the detector 12 are arranged facing each other across the welded portion 81 of the object being inspected by the clamp 16. If radiation is irradiated from the radiation irradiation unit 11 in this configuration, the radiation passing through the welded portion 81 is detected by the detector 12, and the imaging data described above is obtained as digital data based on the transmission intensity of the radiation detected by the detector 12.

[0067] Furthermore, the data acquisition unit 1 includes a movable body 14 that moves the radiation irradiation unit 11, the detector 12, and the clamp 16 along a guide rail 15, which is a movable track externally mounted along the circumferential direction of the piping 100. By moving while maintaining the state of being positioned opposite each other by the clamp 16, the radiation irradiation unit 11 and the detector 12 can be synchronized and move in the circumferential direction of the welding unit 81. The movable body 14 and the guide rail 15 correspond to the moving mechanism in this example.

[0068] In this example, the clamp 16 can be adjusted flexibly according to the outer diameter of the piping 100 to create a distance between the radiation irradiation section 11 and the detector 12. Furthermore, the length of the moving track formed by the guide rail 15 is also configured to be freely adjustable according to the circumference of the piping 100.

[0069] Furthermore, the image data acquisition unit 1 includes: a control unit 2, which serves as an image processing unit, controls the acquisition of image data and performs the combination of the image data described later to create image data; or an operation unit 17, which serves as an operation terminal for operating the image data acquisition unit 1, such as a tablet computer (PC).

[0070] Next, while referring to Figure 3 The process diagram shown and Figure 4 An example of an inspection process for performing a radiation transmission test on a weld 81 using an inspection system including the aforementioned structure will be described. First, imaging data 90 of the weld 81 to be inspected at multiple positions along the circumferential direction is acquired using the imaging data acquisition unit 1 (processing P11). Furthermore, in the following description of the imaging data 90, additional identification symbols A and B are added for differentiation.

[0071] like Figure 4 As shown in (a-1), for example, an operator performing a radiation transmission test sets the radiation irradiation unit 11 and the detector 12 facing each other across the welded portion 81 of the pipe 100 being inspected, using a clamp 16. Radiation is irradiated from the radiation irradiation unit 11 onto the welded portion 81, and the detector 12 detects the radiation, acquiring imaging data 90A based on the intensity of the radiation transmitted through the welded portion 81.

[0072] like Figure 4 As shown in (b-1), the captured data 90A is raw data (RAW data) corresponding to the image along the forming direction of the weld portion 81. Capture data 90A indicates that in Figure 4 The distribution of radiation transmission intensity within the field of view captured by a single photograph when the radiation irradiation unit 11 and detector 12 are arranged at the position shown in (a-1). Figure 4The captured data 90A shown in (b-1) includes a weld line 810 representing the welded part 81 and a damage image 8 representing the damage.

[0073] Furthermore, for ease of explanation, the following will be explained. Figures 4-6 , Figure 8 In the text, for the images displayed based on image data 9 or the image data 91 and the image data 92 that have been determined and are temporarily determined, symbols representing these data 9, 91 and 92 are marked.

[0074] exist Figure 4 After acquiring imaging data at the position shown in (a-1), the imaging data acquisition unit 1 moves the radiation irradiation unit 11 and the detector 12 synchronously in the same direction along the circumference of the piping 100 via a moving mechanism. Next, in Figure 4 Data 90B was captured at different positions along the circumference of the piping 100, as shown in (a-2). Figure 4 The shooting data shown in (b-2) is 90B. Figure 4 The distribution of transmitted radiation intensity within the field of view of the location shown in (a-2) is captured by photography.

[0075] at this time, Figure 4 The configuration position shown in (a-2) is set in the following manner, namely: in Figure 4 The image data 90A was acquired at the location shown in (a-1), and in Figure 4 Between the shooting data 90B obtained at the position shown in (a-2), an overlapping area D is formed, which is part of the field of view of the shooting.

[0076] Thus, in this example, the image data acquisition unit 1 sequentially acquires image data 90 at multiple different positions in the circumferential direction of the piping 100. Figure 4 The text describes two shooting locations, but while moving the radiation irradiation unit 11 and detector 12 further, multiple shooting data 90 are acquired around the entire circumference of the welded part 81 of the piping 100.

[0077] After acquiring shooting data 90 at multiple different locations along the circumference of the piping 100, the control unit 2 combines these multiple shooting data 90 to create... Figure 4 Image data 9 shown in (c) (processing P12). At this time, for example, if image determination is performed on each captured data 90 and the captured data 90 are combined in such a way that the parts of the images overlap, the overlapping area D mentioned above can be automatically overlapped, and image data 9 combined along the circumferential direction of the piping 100 can be obtained.

[0078] From the viewpoint of combining the multiple captured data 90 described above, the control unit 2 provided in the captured data acquisition unit 1 constitutes the image processing unit of the inspection system of this example. Furthermore, along with the processing of image data 9 acquired from the multiple captured data 90, appropriate processing to remove overlapping portions may be performed.

[0079] Through the processing described above, image data 9 can be obtained. This image data 9 represents... Figure 2 The distribution of radiation transmission intensity in the circumferential direction of the welded portion 81 of the pipe 100 within the area enclosed by the dashed line. Furthermore, the control unit 2 can also add identification information to the image data 9, such as information about the welder performing the welding, the location where the welding was performed, and the date and time of the shooting.

[0080] The generated image data 9 is transmitted to the determination processing unit 3, for example, via a transferable storage medium such as a Universal Serial Bus (USB) or a Secure Digital (SD) card, or via a data communication path. For example, the determination processing unit 3 includes an operation display unit 31 that displays an image based on the acquired image data 9 on a screen (monitor) and accepts input of determination result information regarding the image data 9. Next, a qualified person 200, who performs the qualification or failure determination of the welding result, determines, via the operation display unit 31 of the determination processing unit 3, whether there are defects in the welded portion 81 of the piping 100, which is the object of the radiation transmission test. The determination processing unit 3 stores the determination data indicating the determination result corresponding to the image data 9.

[0081] Figure 5 of (a), Figure 5 (b) represents an example of the processing that makes the determination data correspond to the image data 9. Figure 5 (a) represents the image data 9 that has just been combined by the image processing unit 2.

[0082] If the welded section 81, which has been photographed using the image data acquisition unit 1, contains damage, a damage image 8 appears in the image data 9 within the weld line 810 representing the welded section 81. For example, the damage image 8 can be identified as a region where the intensity of radiation transmitted through the welded section 81 differs (a region with different image contrast).

[0083] The qualified user 200 operates the operation display unit 31 while visually inspecting the displayed image, such as... Figure 5As shown in (b), a marking process (labeling) is performed to annotate the damaged image 8 with a mark 80 in a manner that surrounds the image determined to be a defect. Thus, the qualified party 200 writes the mark 80 into the image data 9, the mark 80 indicating the location and size of the damaged image 8 corresponding to the defect. In the example described, all marked damages are determined to be "defects".

[0084] Furthermore, the structure can be configured such that all damages (damage images 8) that can be identified from the image are marked, and further identification information is added to the damages that are identified as defects. In this case, the damage image 8 with the identification information indicating defects can also be displayed as an image with a different color or shading than other damages that are not defects.

[0085] In each of the examples, the information indicating the location and size of the damaged image 8, determined by marking the damaged image 8 that is determined to be defective, is equivalent to the judgment data indicating the judgment result of the defect.

[0086] The image data 9 and the defect determination data are stored in a storage unit (processing P13) of the determination processing unit 3 (not shown). Hereinafter, the image data 9 corresponding to the determination data will also be referred to as the determination completed image data 91.

[0087] If the judgment processing unit 3 receives a request from a user, such as the person in charge of the factory construction, to display the judgment result of the welding unit 81, and wants to confirm the judgment result, it will retrieve the judgment completion image data 91 from the storage unit (not shown) and output it in the form of an image (processing P14). The position information of each welding line 810 of each welding unit 81 can be identified based on the product number of the piping 100 or piping component attached to the judgment completion image data 91, or the location of the welding unit, the date and time of shooting, and the distance from the starting point of the radiation transmission test (usually the starting point is set to 0, and the position information is identified as 1, 2, 3 in 10mm units). In addition, the output unit for the user to output the judgment result can also be configured such that it is located in a different location from the judgment processing unit 3 and obtains the judgment completion image data 91 via the cloud or the like.

[0088] The inspection system according to this embodiment has the following advantages. In the radiation transmission test of the welded portion 81 of the piping 100 or piping component, radiation is transmitted through the welded portion 81 of the piping 100 or piping component under inspection, and image data 90 of the welded portion 81 is obtained. Conventionally, each image capture requires preparation time for the radiation transmission test, such as placing the membrane on the piping or moving the equipment from the radiation control area. However, the inspection system according to this embodiment significantly improves work efficiency. Furthermore, the image data 90 is used to determine defects in the welded portion 81 of the inspected object, and processing is performed to correlate the determination data representing the determination result with image data 9 representing the distribution of radiation transmission intensity. Therefore, determination results can be obtained without using a membrane, improving the work efficiency of the radiation transmission test by reducing the burden on the operator. Moreover, by using image data, image magnification, contrast adjustment, and positive / negative reversal can be performed, thus improving inspection accuracy and efficiency compared to visual inspection using a membrane.

[0089] As mentioned above, in the past, after taking multiple radiographic images of the welded portion 81 of the object to be inspected along the circumference of the piping 100 at the factory construction site, a qualified person 200 would determine whether the welded portion 81 had defects based on the developed images. Therefore, in radiographic testing, not only is it necessary to maintain a storage facility for the images, but the development of the images or the treatment of the chemicals used also consume a lot of effort.

[0090] In this respect, by acquiring the shooting data 90A and 90B as digital data, it is not necessary to ensure the storage of the film or the chemicals required for the development of the film.

[0091] Furthermore, when the radiation irradiation unit and the retainer with the membrane are moved manually to photograph the welded part 81 around the entire circumference of the piping 100, from the point of view of ensuring a safe distance, there is also the problem that it is difficult to install many photographing devices in a limited construction area.

[0092] In this respect, such as using Figure 2As explained, if the imaging data acquisition unit 1 is equipped with a structure that allows the radiation irradiation unit 11 and detector 12 to move, remote operation can be performed via the control unit 2 while ensuring a safe distance. By using a detector with higher sensitivity than the membrane, more radiation transmission tests can be performed simultaneously without compromising work efficiency, even with a smaller radiation source. Furthermore, without setting up the membrane on the welding section 81 or requiring the operator to move the radiation irradiation unit 11 and detector 12 each time they enter or leave the radiation area, imaging data 90 can be continuously captured around the entire circumference of the welding section 81. Moreover, imaging data 90 with less uniformity in imaging accuracy can be acquired without relying on the skill of the operator performing the imaging, thus improving inspection accuracy. Furthermore, by directly acquiring imaging data 90 as digital data, it is easy to confirm the image on-site, reducing the risk of re-imaging in case of imaging failure.

[0093] Furthermore, in the past, when visually inspecting the developed film to determine defects, it was necessary to dispatch 200 qualified personnel to the site, which was sometimes difficult to do in a timely manner due to the remote location.

[0094] In this respect, image data 9, as digital data, can also be sent to distant locations, significantly reducing the limitations on the dispatch of qualified personnel 200. From this perspective, it is also possible to... Figure 1 The determination processing unit 3 shown is configured as an independent determination processing device that transmits and receives image data 9 or determined image data via the cloud, and only provides services for adding determination data to image data 9.

[0095] Here, as an example of a standard for determining defects in the weld 81 of the piping 100, there is American Society of Mechanical Engineers (ASME) standard B31.3. Table 1 shows the types of defects in the weld 81 based on ASME B31.3.

[0096] [Table 1]

[0097] Types of defects 1 rupture 2 Poor fusion 3 Poor solder penetration 4 circular shape 5 linear shape 6 Undercut 7 dent

[0098] Table 1 shows that "Crack" refers to a cracked defect in the weld 81, and "Lack of Fusion" refers to a gap created inside the weld 81 due to incomplete melting of the welding material. Furthermore, "Incomplete Penetration" refers to a depression formed in the weld 81 on the inner side of the pipe 100 due to the welding material not completely filling the space between the pipes 100. Damage remaining on the surface of the weld 81 is categorized as "Rounded Indication" and "Elongated Indication." Other defects include "Undercutting," which creates a gap between the welding material and the pipe 100 on the surface, and "Concave Surface," which causes a depression on the surface of the weld 81.

[0099] Figure 6 (a)~ Figure 6 (c) is an example of image data 9 of weld 81 containing these defects. Figure 6 In (a), the damage image 8A contained in weld line 810 represents damage equivalent to a defect "crack". Figure 6 Damage image 8B in (b) represents "poor fusion" damage. Furthermore, Figure 6 In (c), damage image 8C represents "poor penetration" damage. Thus, the qualified person 200 can determine the location or size of the damage formed on the weld 81 based on the position, size, and shape characteristics of damage image 8 in the image, and, if the damage is equivalent to a defect, add judgment data indicating this condition. Identification information identifying the defect category can also be added to the judgment data.

[0100] When determining the category of these defects, it is sometimes difficult to identify which type of defect corresponds to the damaged images 8A to 8C contained in the image data 9 at first glance, which can sometimes place a heavy burden on the qualified personnel 200.

[0101] Therefore, the inspection system of the embodiments of the present invention is as follows: Figure 1 As shown, it may also include: an automatic determination unit 5, which determines the location, size, and shape characteristics of the damaged image 8 contained in the image data 9, and identifies which type of defect shown in Table 1 it corresponds to, and adds information to identify the type of defect and performs marking.

[0102] An example of automatic decision processing using the automatic decision unit 5 is shown below. Figure 7The process diagram is shown. The automatic determination unit 5 acquires the determined image data 91, which is the image data 9 corresponding to the determination processing unit 3 and has determination data, and stores it in its storage unit 51.

[0103] Next, the automatic determination unit 5 reads the determined image data 91 stored in the storage unit 51 and performs machine learning (processing P21) related to the position, size, and shape features of the damaged image 8 that has been determined to be defective. Moreover, machine learning related to the position, size, and shape features of the damaged image 8 that has not been determined to be defective can also be performed at this time.

[0104] Furthermore, the automatic determination unit 5 constructs an Artificial Intelligence (AI) determination algorithm based on the results of machine learning for damage assessment. This AI determination algorithm is a calculation procedure for determining whether the damage contained in the new image data 9 is equivalent to a defect and identifying the type of defect. According to the AI ​​determination algorithm, for example, the damage contained in the weld 81 can be identified based on the area or shape, formation direction, or grayscale variation tendency of the part contained in the image data 9 that has a different contrast from the surrounding weld lines 810, and the determination of whether the damage is equivalent to a defect and the identification of the type of defect are performed.

[0105] After the AI ​​algorithm is constructed, if new image data 9 is acquired from the image data acquisition unit 1, for example, an operator or qualified person 200 conducting a radiation penetration test may request the automatic judgment unit 5 to automatically judge the image data 9. If the automatic judgment unit 5 receives the request, it receives the image data 9 before judgment processing from the judgment processing unit 3 and performs automatic judgment on the acquired image data 9 based on the AI ​​judgment algorithm.

[0106] The system automatically determines the location, size, and shape features of the damaged image 8 to determine whether the damage corresponding to the damaged image 8 is equivalent to a defect. If the damage is determined to be equivalent to a defect, the type of defect is then identified.

[0107] Next, the image data 9 that is determined to contain defects is marked, that is, a symbol 80 containing information indicating the type of defect is used as temporary determination data. The temporary determination data is information that can identify the location of the damaged image 8, the size of the damaged image 8, and the shape features of the damaged image 8 (processing P22). Hereinafter, the image data 9 corresponding to the temporary determination data is also referred to as the temporarily determined image data 92.

[0108] Figure 8In the example shown, the area corresponding to the defect identified as "cracked" is marked with symbol 80A, and the area corresponding to the defect identified as "poor fusion" is marked with symbol 80B. These symbols 80A and 80B correspond in advance to the various defects illustrated in Table 1 and are set with different types of colors or outlines in a visually recognizable manner.

[0109] Image data 9 (temporary judgment completed image data 92) with attached temporary judgment data containing information indicating the type of defect is sent to the judgment processing unit 3 and stored in the storage unit 31.

[0110] When using this temporarily determined image data 92, which has been automatically determined by the automatic determination unit 5, the determination processing unit 3, as follows: Figure 9 As shown in P13', the image of the temporarily determined image data 92 is displayed, and the qualified person 200 performs a defect determination. The result is that the temporary determination data is directly adopted based on the determination of the qualified person 200, or the temporary determination data is corrected and saved as the official determination data corresponding to the image data 9.

[0111] Furthermore, the arithmetic unit 52 can also be located in the determination processing unit 3. That is, the AI ​​determination algorithm can also be located in the determination processing unit 3. As a result, the time for sending and receiving image data 9 can be reduced. Moreover, the AI ​​determination algorithm can be used in places where the Internet is not available.

[0112] Use the above Figure 1 Furthermore, the image data acquisition unit 1 included in the inspection system can also be composed of an image reading device that scans the captured analog image to acquire image data 90. For example, Figure 10 In the example of the inspection system shown, radiation is irradiated onto the area containing the weld portion 81 of the object to be inspected, obtaining a simulated image 93. This simulated image 93 is obtained by exposing the membrane to the radiation image transmitted through the weld portion 81. By reading the simulated image 93 via the image reading device 19, photographic data 90, converted into digital data, can be acquired. Furthermore, Figure 10 The example described herein illustrates how a single image reading device 19 can be used to acquire image data 90 from analog images 93 taken at multiple locations. Furthermore, not limited to the example shown, multiple different image reading devices 19 can be used to read these analog images 93 and acquire image data 90 separately.

[0113] The inspection system can also perform image processing on the captured data 90 read by the image reading device 19, and the determination processing unit 3 performs determination data matching on the obtained image data 90. Thus, the user can easily obtain the determined image data 91 with corresponding determination data, which is convenient for users. Figure 1The inspection system described in this embodiment is the same.

Claims

1. An inspection system for non-destructive inspection of defects in welded parts of piping or piping components, characterized in that, include: The imaging data acquisition unit acquires imaging data, which is obtained by passing radiation through the pipes or welded parts of the pipe components of the object being inspected. The determination processing unit performs processing based on the distribution of the transmitted intensity of the radiation obtained from the imaging data to correspond the determination data with image data representing the distribution of the transmitted intensity of the radiation. The determination data includes: information representing the result of a qualified person who has the qualification to determine the defects of the welded part of the inspected object, corresponding to information that can identify the location, size, and shape characteristics of the damage to the weld line contained in the image data, and information representing the type of defect, wherein the defect is a defect in which the welding result of the welded part is determined to be unqualified or harmful. as well as The automatic determination unit acquires image data corresponding to the determination data from the determination processing unit. Based on the damage contained in previous image data corresponding to the determination data, and performing machine learning, it generates temporary determination completion image data corresponding to temporary determination data for new image data representing the distribution of radiation transmission intensity. The temporary determination data determines the location, size, and shape characteristics of the weld line damage and automatically determines the type of the anticipated defect. The automatic determination unit sends the temporarily determined image data to the determination processing unit. When the determination processing unit receives the temporarily determined image data from the automatic determination unit, the determination processing unit performs a defect determination on the temporary determination data corresponding to the temporarily determined image data, and processes it as formal determination data corresponding to the new image data.

2. The inspection system according to claim 1, characterized in that, The imaging data acquisition unit includes a radiation irradiation unit and a detector arranged facing each other across the welded portion of the object under inspection. The radiation irradiation unit irradiates the welded portion with radiation, the detector detects the radiation that passes through the welded portion, and the imaging data acquisition unit acquires the imaging data in the form of digital data based on the transmission intensity of the radiation detected by the detector.

3. The inspection system according to claim 2, characterized in that, The image data acquisition unit includes: A clamp for arranging the radiation irradiation unit and the detector facing each other across the welded portion of the object being inspected; and The moving mechanism, while maintaining the opposing configuration of the clamps, synchronizes and moves the radiation irradiation unit and the detector in the circumferential direction of the welding unit, thereby moving the position irradiating the radiation in the circumferential direction. The image data was acquired at multiple different locations along the circumference of the welded part.

4. The inspection system according to claim 3, characterized in that, include: The image processing unit combines multiple images captured at different locations along the circumferential direction of the welded part to create the image data.

5. The inspection system according to claim 1, characterized in that, Irradiate the area containing the weld portion of the object under inspection with radiation, and convert the analog image obtained by the membrane through the radiation image transmitted through the weld portion into a digital image, and use the digital image as the imaging data.

6. The inspection system according to claim 1, characterized in that, The determination processing unit is configured to be capable of undergoing correction processing based on the temporary determination data corresponding to the image data by the automatic determination unit.

7. A judgment processing system for non-destructive inspection of defects in welded parts of piping or piping components, characterized in that, include: The determination processing unit performs processing based on the distribution of radiation transmission intensity obtained by passing radiation through the welded portion of the piping or piping component of the inspection object, and the processing of the determination data to correspond with image data representing the distribution of radiation transmission intensity. The determination data includes: information representing the result of a qualified person who has been qualified to determine defects in the welded portion of the inspection object, corresponding to information that can identify the location, size, and shape characteristics of the damage to the weld line contained in the image data, and information representing the type of defect, wherein the defect is a defect in which the welding result of the welded portion is determined to be unqualified or harmful. as well as The automatic determination unit acquires image data corresponding to the determination data from the determination processing unit. Based on the damage contained in previous image data corresponding to the determination data, and performing machine learning, it generates temporary determination completion image data corresponding to temporary determination data for new image data representing the distribution of radiation transmission intensity. The temporary determination data determines the location, size, and shape characteristics of the weld line damage and automatically determines the type of the anticipated defect. The automatic determination unit sends the temporarily determined image data to the determination processing unit. When the determination processing unit receives the temporarily determined image data from the automatic determination unit, the determination processing unit performs a defect determination on the temporary determination data corresponding to the temporarily determined image data, and processes it as formal determination data corresponding to the new image data.

8. A judgment processing apparatus for non-destructive inspection of defects in welded parts of piping or piping components, characterized in that, Composed of, Based on the distribution of radiation transmission intensity obtained by allowing radiation to pass through the welded portion of the piping or piping component of the inspection object, processing is performed to correlate judgment data with image data representing the distribution of radiation transmission intensity. The judgment data includes: information indicating the result of a qualified person determining defects in the welded portion of the inspection object; information corresponding to the location, size, and shape characteristics of damage to the weld line contained in the image data; and information indicating the type of defect, where the welded portion is determined to be unqualified or harmful. The determination processing device sends image data corresponding to the determination data to the automatic determination unit. The automatic determination unit, based on machine learning results of damage contained in previous image data corresponding to the determination data, acquires temporary determination completion image data corresponding to the temporary determination data for new image data representing the distribution of radiation transmission intensity. The temporary determination data determines the location, size, and shape characteristics of the weld line damage, and automatically determines the type of the anticipated defect. When the determination processing device receives the temporarily determined image data from the automatic determination unit, the determination processing device performs a defect determination on the temporary determination data corresponding to the temporarily determined image data, and processes it as formal determination data corresponding to the new image data.

9. The determination and processing device according to claim 8, characterized in that, Composed of, It can accept and process corrections to the temporary determination data corresponding to the image data.

10. An inspection method for non-destructive inspection of defects in welded parts of piping or piping components, characterized in that, The process includes the following steps: Based on the distribution of radiation transmission intensity obtained by allowing radiation to pass through the welded portion of the piping or piping component of the inspection object, processing is performed to correspond judgment data with image data representing the distribution of radiation transmission intensity. The judgment data includes: a result indicating a defect in the welded portion of the inspection object as determined by a qualified person qualified to determine defects, corresponding to information that can identify the location, size, and shape characteristics of the damage to the weld line contained in the image data, and information indicating the type of defect, wherein the defect is damage to the welded portion that is determined to be unqualified or harmful. The image data corresponding to the determination data is sent to the automatic determination unit, and the automatic determination unit performs machine learning based on the damage contained in the previous image data corresponding to the determination data. For the new image data representing the distribution of radiation transmission intensity, the unit obtains temporary determination completed image data corresponding to the temporary determination data. The temporary determination data determines the location, size, and shape characteristics of the weld line damage and automatically determines the type of the envisioned defect. as well as When the temporary judgment completed image data is received from the automatic judgment unit, the qualified person who accepted the temporary judgment data corresponding to the temporary judgment completed image data is judged for defects, and processed as formal judgment data corresponding to the new image data.

11. The inspection method according to claim 10, characterized in that, The process of making the determination data correspond to the image data includes a correction process for the temporary determination data corresponding to the image data.

Images