Methods for generating a roundness prediction model for steel pipes, methods for predicting the roundness of steel pipes, methods for controlling the roundness of steel pipes, methods for manufacturing steel pipes, and devices for predicting the roundness of steel pipes.

Abstract

The method for generating a steel pipe roundness prediction model according to the present invention includes an operating condition dataset in the input data and sets the roundness of the steel pipe after the pipe expansion process as the output data. The calculation is performed multiple times while changing the operating condition dataset, thereby generating multiple sets of data corresponding to the operating condition dataset and the roundness of the steel pipe after the pipe expansion process as the learning data offline. Using multiple learning data, a roundness prediction model is generated offline through machine learning, with the operating condition dataset as the input data and the roundness of the steel pipe after the pipe expansion process as the output data.

Description

Technical Field

[0001] This invention relates to a method for generating a roundness prediction model of a steel pipe after the pipe expansion process in the manufacturing process of a steel pipe using the bending method, a method for predicting the roundness of a steel pipe, a method for controlling the roundness of a steel pipe, a method for manufacturing a steel pipe, and a device for predicting the roundness of a steel pipe. Background Technology

[0002] As a manufacturing technology for large-diameter and thick-walled steel pipes used in pipelines and the like, the following steel pipe manufacturing technology (so-called UOE steel pipe) is widely used: Steel plates with specified length, width, and thickness are stamped into a U-shape, then stamped into an O-shape, and the butt joints are welded to form a steel pipe. Furthermore, its diameter is increased (so-called pipe expansion) to improve roundness. However, in the manufacturing process of UOE steel pipes, the stamping of steel plates into U- and O-shapes requires enormous stamping pressure, thus necessitating the use of large-scale stamping machinery.

[0003] To address this, a technique has been proposed to reduce stamping pressure and form large-diameter, thick-walled steel pipes. Specifically, the following technique has been put into practical use: After bending the ends of a steel plate in the width direction (so-called end bending), a U-shaped cross-section is formed by multiple three-point bending stamping processes using a punch (hereinafter, sometimes referred to as a U-shaped forming body). Furthermore, a weld gap reduction process is performed to reduce the weld gap of the U-shaped forming body, forming a slotted pipe. The butt joints are then welded to form a steel pipe. Finally, a pipe expander is inserted into the steel pipe to enlarge its inner diameter. It should be noted that the pipe expander uses a device that includes multiple expanding tools with curved surfaces obtained by dividing an arc into multiple parts. The steel pipe is expanded and its shape adjusted by abutting the curved surfaces of the expanding tools against the inner surface of the steel pipe.

[0004] In the bending process, increasing the number of three-point bending presses improves the roundness of the steel pipe after the expansion process, but it takes a long time to form the pipe into a U-shaped cross-section. On the other hand, reducing the number of three-point bending presses results in a cross-sectional shape that is close to a polygon and difficult to achieve a perfect circle. Therefore, the number of three-point bending presses is determined empirically based on the size of the steel pipe (for example, 5 to 13 times for a steel pipe with a diameter of 1200 mm). Many suggestions have been made regarding the operating conditions for the bending process used to improve the roundness of the steel pipe after such an expansion process.

[0005] For example, Patent Document 1 describes the following method: This method is used to perform the three-point bending stamping as few times as possible, so that multiple tube expanding tools arranged in the circumferential direction of the tube expanding device come into contact with the undeformed part that has not been deformed by the three-point bending stamping to expand the tube.

[0006] In addition, Patent Document 2 describes the following method: by making the radius of curvature of the outer circumferential surface of the punch used for three-point bending stamping and the radius of curvature of the outer circumferential surface of the tube expanding tool satisfy a specified relationship, the roundness of the steel pipe after the tube expanding process is improved.

[0007] Furthermore, Patent Document 3 describes a method for efficiently manufacturing steel pipes with high roundness without requiring excessive pressing pressure during the bending process. The method involves providing a lightly machined section with an extremely small curvature compared to other areas, or an unmachined section where bending is omitted, on at least a portion of the steel plate during three-point bending stamping. Additionally, Patent Document 3 describes applying pressing pressure to a portion of the lightly machined or unmachined section that is a predetermined distance from its center, without binding the lightly machined or unmachined section. It should be noted that an O-type stamping device is typically used in the weld gap reduction process, which is performed after the bending process.

[0008] In contrast, Patent Document 4 describes a method where, after forming a non-circular preform (a U-shaped cross-section) using three-point bending stamping, instead of the usual O-type stamping process, the non-circular preform is supported by two lower support rollers, and a pressing tool positioned above the preform, facing the lower support rollers, applies pressure from the outside of the non-circular preform to reduce the weld gap (hereinafter referred to as the "closed stamping method"). This method is characterized by the simplified device structure, as the pressing force is applied from the outside of the non-circular preform using the pressing tool, eliminating the need for a mold based on the outer diameter of the steel pipe, as is required with O-type stamping devices. Furthermore, Patent Document 4 describes how, in three-point bending stamping, a region is intentionally formed with a relatively small area; in the closed stamping method used in the weld gap reduction process, the pressing force acts on this region of the U-shaped cross-section.

[0009] On the other hand, Non-Patent Document 1 describes the following method: the influence of the operating conditions of the tube expansion process on the roundness of the steel pipe after the tube expansion process is analyzed by using the finite element method.

[0010] Existing technical documents

[0011] Patent documents

[0012] Patent Document 1: Japanese Patent Application Publication No. 2012-170977

[0013] Patent Document 2: Japanese Patent No. 5541432

[0014] Patent Document 3: Japanese Patent No. 6015997

[0015] Patent Document 4: Japanese Patent Application Publication No. 2012-250285

[0016] Non-patent literature

[0017] Non-Patent Literature 1: Plasticity and Processing, Vol. 59, No. 694 (2018), pp. 203-208 Summary of the Invention

[0018] The problem that the invention aims to solve

[0019] The method described in Patent Document 1 improves the roundness of the steel pipe after the expansion process by establishing a correspondence between the pressing positions of the three-point bending stamping and the pressing positions of the pipe expanding tool. However, the manufacturing process of the steel pipe includes at least several processes: bending, reducing weld gap, welding, and expansion. Therefore, the method described in Patent Document 1 does not consider the influence of the operating conditions of other processes on the roundness of the steel pipe after the expansion process, and thus may not always improve the roundness of the steel pipe after the expansion process.

[0020] The method described in Patent Document 2, like that described in Patent Document 1, improves the roundness of the steel pipe after the pipe-expanding process by ensuring that the radius of curvature of the outer circumferential surface of the punch used in the three-point bending stamping (an operating condition of the bending process) and the radius of curvature of the outer circumferential surface of the pipe-expanding tool (an operating condition of the pipe-expanding process) satisfy a prescribed relationship. However, like the method described in Patent Document 1, the method described in Patent Document 2 has the problem of not being able to consider the influence of processes other than the bending process, such as the weld gap reduction process.

[0021] The method described in Patent Document 3 improves the roundness of the steel pipe after the expansion process by setting the processing conditions of the three-point bending stamping in the bending process to be changed according to the position of the steel plate and establishing a correlation with the forming conditions in the weld gap reduction process. However, the method described in Patent Document 3 has the following problem: if there are deviations in the thickness and material of the steel plate used as the blank, the roundness of the steel pipe after the expansion process will deviate even under the same forming conditions.

[0022] The method described in Patent Document 4 also improves the roundness of the steel pipe after the expansion process by establishing conditions that link the forming conditions of the U-shaped cross-section in the bending process with the forming conditions in the weld gap reduction process. However, the method described in Patent Document 4 also has the following problem: if there are deviations in the thickness or material of the steel plate used as the blank, the roundness of the steel pipe after the expansion process will deviate even under the same forming conditions.

[0023] On the other hand, by performing an offline calculation using the finite element method to analyze the tube expansion process, as described in Non-Patent Document 1, the impact of the operating parameters of the tube expansion process on roundness can be quantitatively predicted. However, the method described in Non-Patent Document 1 also has the problem that it cannot consider the impact of the operating conditions of other processes on roundness. Moreover, in the case of such numerical analysis, the calculation time required is long, thus making it difficult to predict roundness online.

[0024] This invention was made to solve the above-mentioned problems, and its object is to provide a method for generating a roundness prediction model for steel pipes, capable of generating a high-precision and rapid prediction model of the roundness of steel pipes after the expansion process in the manufacturing process of steel pipes consisting of multiple processes. Furthermore, another object of this invention is to provide a method and apparatus for predicting the roundness of steel pipes after the expansion process in the manufacturing process of steel pipes consisting of multiple processes, capable of highly accurate prediction. Additionally, another object of this invention is to provide a method for controlling the roundness of steel pipes after the expansion process in the manufacturing process of steel pipes consisting of multiple processes, capable of highly accurate control. Finally, another object of this invention is to provide a method for manufacturing steel pipes capable of producing steel pipes with desired roundness with high yield.

[0025] Methods for solving problems

[0026] The present invention relates to a method for generating a roundness prediction model for steel pipes. This method generates a roundness prediction model for steel pipes after the pipe expansion process in a steel pipe manufacturing process, which includes a bending process, a weld gap reduction process, a welding process, and a pipe expansion process. The bending process is a process of forming a U-shaped cross-section from a steel plate by multiple presses with a punch. The weld gap reduction process is a process of reducing the weld gap of the U-shaped cross-section to form a slotted pipe. The welding process is a process of joining the ends of the slotted pipe together. The pipe expansion process is a process of increasing the inner diameter of the steel pipe after the ends are joined together. The generation method includes a basic data acquisition step, which includes an operating condition dataset in the input data and sets the roundness of the steel pipe after the pipe expansion process as... The numerical calculation of the output data is performed multiple times while changing the operating condition dataset. As a result, multiple sets of data on the roundness of the steel pipe after the expansion process corresponding to the operating condition dataset are generated offline as learning data. The operating condition dataset includes one or more parameters selected from the attribute information of the steel plate, one or more parameters selected from the operating parameters of the bending process, and one or more parameters selected from the operating parameters of the weld gap reduction process. The roundness prediction model generation step uses the multiple learning data generated in the basic data acquisition step to generate a roundness prediction model offline through machine learning, with the operating condition dataset as input data and the roundness of the steel pipe after the expansion process as output data.

[0027] Preferably, the basic data acquisition step includes the step of calculating the roundness of the steel pipe after the pipe expansion process based on the operating condition dataset using the finite element method.

[0028] Preferably, the roundness prediction model includes one or more parameters selected from the operating parameters of the tube expansion process as part of the operating condition dataset.

[0029] Preferably, the manufacturing process of the steel pipe includes an end bending process that bends the end of the steel plate in the width direction prior to the bending process, and the roundness prediction model includes one or more parameters selected from the operating parameters of the end bending process as the operating condition dataset.

[0030] Preferably, the operating parameters of the bending process include the number of punches performed in the bending process, the punching position information of the punch used in the bending process pressing the steel plate, and the punching reduction amount.

[0031] Ideally, the machine learning method used is selected from neural networks, decision tree learning, random forests, Gaussian process regression, and support vector machine regression.

[0032] The method for predicting the roundness of steel pipes according to the present invention includes: an operation parameter acquisition step, which acquires an operation condition dataset set online as the operation conditions of the manufacturing process of the steel pipe as input to the steel pipe roundness prediction model generated by the method for generating the steel pipe roundness prediction model according to the present invention; and a roundness prediction step, which predicts the roundness information of the steel pipe after the pipe expansion process by inputting the operation condition dataset obtained in the operation parameter acquisition step into the roundness prediction model.

[0033] The first aspect of the present invention relates to a method for controlling the roundness of a steel pipe, comprising the following steps: using the steel pipe roundness prediction method of the present invention, before the start of the bending process, obtaining an operational condition dataset including actual values ​​of the steel plate's attribute information, set values ​​of the operation parameters for the bending process, and set values ​​of the operation parameters for the weld gap reduction process; predicting the roundness of the steel pipe after the expansion process by inputting the obtained operational condition dataset into the roundness prediction model; and resetting at least one of the set values ​​of the operation parameters for the bending process and the set values ​​of the operation parameters for the weld gap reduction process in a manner that reduces the predicted roundness.

[0034] The second aspect of the present invention relates to a method for controlling the roundness of a steel pipe, comprising the following steps: using the steel pipe roundness prediction method of the present invention, before the start of a resetting target process selected from the end bending process, pressure bending process, weld gap reduction process, and pipe expansion process constituting the steel pipe manufacturing process, predicting the roundness information of the steel pipe after the pipe expansion process; and based on the predicted roundness information of the steel pipe, resetting at least one or more operating parameters selected from the operating parameters of the resetting target process or one or more operating parameters selected from the operating parameters of a forming process downstream of the resetting target process.

[0035] The steps include manufacturing steel pipes using the roundness control method for steel pipes disclosed in this invention.

[0036] The present invention relates to a steel pipe roundness prediction device for predicting the roundness of a steel pipe after the pipe expansion process in a steel pipe manufacturing process, which includes a bending process, a weld gap reduction process, a welding process, and a pipe expansion process. The bending process is a process of forming a U-shaped cross-section from a steel plate by multiple presses with a punch. The weld gap reduction process is a process of reducing the weld gap of the U-shaped cross-section to form a slotted pipe. The welding process is a process of joining the ends of the slotted pipe together. The pipe expansion process is a process of increasing the inner diameter of the steel pipe after the ends are joined together. The roundness prediction device includes a basic data acquisition unit that performs numerical calculations multiple times, taking an operating condition dataset as input data and setting the roundness information of the steel pipe after the pipe expansion process as output data, while changing the operating condition dataset. This allows the device to obtain the roundness information of the steel pipe after the pipe expansion process corresponding to the operating condition dataset. Multiple sets of data are generated as learning data. The operating condition dataset includes one or more parameters selected from the attribute information of the steel plate, one or more operating parameters selected from the operating parameters of the bending process, and one or more operating parameters selected from the operating parameters of the weld gap reduction process. A roundness prediction model generation unit uses the multiple learning data generated in the basic data acquisition unit to generate a roundness prediction model by machine learning, with the operating condition dataset as input data and the roundness information of the steel pipe after the expansion process as output data. An operating parameter acquisition unit acquires online the operating condition dataset set as the operating conditions of the manufacturing process of the steel pipe. A roundness prediction unit uses the roundness prediction model generated in the roundness prediction model generation unit to predict online the roundness information of the steel pipe after the expansion process corresponding to the operating condition dataset acquired by the operating parameter acquisition unit.

[0037] Preferably, a terminal device is provided, which has an input unit for acquiring input information based on user operation and a display unit for displaying the roundness information. The operation parameter acquisition unit updates part or all of the operation condition dataset in the manufacturing process of the steel pipe based on the input information acquired by the input unit, and the display unit displays the roundness information of the steel pipe predicted by the roundness prediction unit using the updated operation condition dataset.

[0038] Invention Effects

[0039] According to the method for generating a roundness prediction model for steel pipes disclosed in this invention, a roundness prediction model can be generated with high accuracy and rapid speed to predict the roundness of steel pipes after the expansion process in a steel pipe manufacturing process consisting of multiple processes. Furthermore, according to the steel pipe roundness prediction method and apparatus disclosed in this invention, the roundness of steel pipes after the expansion process in a steel pipe manufacturing process consisting of multiple processes can be predicted with high accuracy and rapid speed. Additionally, according to the steel pipe roundness control method disclosed in this invention, the roundness of steel pipes after the expansion process in a steel pipe manufacturing process consisting of multiple processes can be controlled with high accuracy. Finally, according to the steel pipe manufacturing method disclosed in this invention, steel pipes with desired roundness can be manufactured with high yield. Attached Figure Description

[0040] Figure 1 This is a diagram illustrating the manufacturing process of a steel pipe as an embodiment of the present invention.

[0041] Figure 2 This diagram illustrates an example of the process of forming a U-shaped cross-section using a bending device.

[0042] Figure 3 This diagram illustrates an example of the process of forming a U-shaped cross-section using a bending device.

[0043] Figure 4 This is a diagram showing an example of the structure of an O-type stamping device.

[0044] Figure 5 This is a diagram showing an example of the structure of a closed stamping device.

[0045] Figure 6 This is a diagram showing an example of the structure of a tube expander device.

[0046] Figure 7 This is a diagram illustrating an example of the structure of a device for measuring the outer diameter shape of a steel pipe.

[0047] Figure 8 This is a block diagram showing the structure of a steel pipe roundness prediction model generation device as an embodiment of the present invention.

[0048] Figure 9 It is shown Figure 8 The diagram shows the structure of the offline roundness calculation unit.

[0049] Figure 10 This is a diagram illustrating an example of how the relationship between the stamping amount and the roundness of the steel pipe after the expansion process changes with the change in operating conditions of the bending process.

[0050] Figure 11 This is a diagram showing an example of the pressing position and pressing amount for each pressing cycle.

[0051] Figure 12 This is a diagram showing the process of predicting the roundness of the steel pipe after the expansion process before the start of the bending process.

[0052] Figure 13 This is a diagram illustrating an example of a finite element model.

[0053] Figure 14 This is a perspective view showing the overall structure of the C-type stamping device.

[0054] Figure 15 This is a cross-sectional view showing the structure of the stamping mechanism.

[0055] Figure 16 This is a diagram illustrating a method for controlling the roundness of a steel pipe as an embodiment of the present invention.

[0056] Figure 17 This is a diagram showing the structure of a steel pipe roundness prediction device as an embodiment of the present invention. Detailed Implementation

[0057] [Steel pipe manufacturing process]

[0058] Figure 1 This is a diagram illustrating the manufacturing process of a steel pipe as an embodiment of the present invention. (See diagram for details.) Figure 1 As shown, in the steel pipe manufacturing process according to one embodiment of the present invention, the steel plate used as the billet is a thick steel plate manufactured by a thick plate rolling process, which is a preceding process in the steel pipe manufacturing process. Here, representative thick steel plates have a yield stress of 245–1050 MPa, a tensile strength of 415–1145 MPa, a plate thickness of 6.4–50.8 mm, a plate width of 1200–4500 mm, and a length of 10–18 m. Furthermore, the width end of the thick steel plate is pre-ground into a chamfered shape called a bevel. This is to prevent overheating of the outer surface corners of the width end during subsequent welding processes, thereby stabilizing the weld strength. Additionally, the width of the thick steel plate affects the outer diameter after being formed into a steel pipe, and is therefore adjusted to a specified range considering the deformation history in subsequent processes.

[0059] In the manufacturing process of steel pipes, an end-bending process is sometimes performed to bend the ends of the steel plate in the width direction. This end-bending process is carried out by a C-type stamping device, which performs end-bending processing (also known as edge curling) on ​​the ends of the steel plate in the width direction. The C-type stamping device has a pair of upper and lower dies and a pair of upper and lower clamping members that hold the center of the steel plate in the width direction. The length of the die is shorter than the length of the steel plate, so the end-bending process is repeated while feeding the steel plate sequentially in the length direction. This end-bending process is performed on both ends of the steel plate in the width direction. Since it is impossible to apply bending moment to the ends in the width direction in a three-point bending stamping process, the end-bending process uses the die to pre-inflate the bending deformation. This improves the roundness of the steel pipe in the final product. Examples of parameters used to determine the processing conditions include the length of the die contact between the end of the steel plate and the center in the width direction (i.e., the end-bending width), the clamping force, the feed rate of the steel plate when the end-bending process is repeated in the length direction, the feed direction, and the number of feeds.

[0060] The subsequent bending process involves using a bending device to perform multiple three-point bending presses with a punch to shape the steel sheet into a U-shaped cross-section. The subsequent weld gap reduction process typically uses an O-ring press to reduce the weld gap of the U-shaped cross-section, forming an open-slit tube. However, the closed-pressing method described in Patent Document 4 can be used instead of the O-ring press. The subsequent welding process involves binding the weld gap formed at the ends of the open-slit tube so that the ends contact each other and joining them together. Thus, the formed body becomes a steel pipe with its ends joined together. The subsequent tube expansion process uses a tube expansion device equipped with multiple tube expansion tools that divide an arc into multiple curved surfaces. The tube is expanded by bringing the curved surfaces of the tube expansion tools against the inner surface of the steel pipe. The steel pipe manufactured in this way is inspected to determine whether its material, appearance, dimensions, and other qualities meet the specified specifications before being shipped as a product. The inspection process includes a roundness measurement process to determine the roundness of the steel pipe.

[0061] In this embodiment, in the series of manufacturing processes that form a steel plate into a slotted pipe and then expand it after welding, the end bending process, the press bending process, the weld gap reduction process, and the pipe expansion process are referred to as "forming processing processes." These processes are common to each other as processes that impart plastic deformation to the steel plate to control the size and shape of the steel pipe. Hereinafter, each process of the steel pipe manufacturing process will be described in detail with reference to the accompanying drawings.

[0062] <End bending process>

[0063] Regarding the C-type stamping device used for end bending, Figure 14, Figure 15 Let me explain in detail. Figure 14 This is a perspective view showing the overall structure of the C-type stamping device. (Example) Figure 14 As shown, the C-type stamping device 30 includes a conveying mechanism 31 that conveys the steel plate S along its length direction; a stamping mechanism 32A that bends one width-direction end Sc to a predetermined curvature by making the downstream side of the steel plate S in the conveying direction forward; a stamping mechanism 32B that bends the other width-direction end Sd to a predetermined curvature; and an interval adjustment mechanism (not shown) that adjusts the interval between the left and right stamping mechanisms 32A and 32B according to the width of the steel plate S undergoing end bending. The conveying mechanism 31 consists of multiple rotary-driven conveying rollers 31a respectively arranged before and after the stamping mechanisms 32A and 32B. It should be noted that the reference numeral Sa in the figure indicates the front end (front end in the length direction) of the steel plate S.

[0064] exist Figure 15 (a) shows a cross-section of the stamping mechanism 32A, which bends one end Sc of the steel plate S in the width direction, viewed from the upstream side of the transport direction of the steel plate S to the downstream side of the transport direction. It should be noted that stamping mechanisms 32A and 32B are symmetrical and have the same structure. Stamping mechanisms 32A and 32B include an upper die 33 and a lower die 34, which are a pair of dies arranged opposite each other in the vertical direction, and a hydraulic cylinder 36, which is a die-moving unit that lifts the lower die 34 together with a tool holder 35 (moving it towards the upper die 33) and closes the die with a specified stamping force. It should be noted that stamping mechanisms 32A and 32B sometimes include a clamping mechanism 37 that holds the steel plate S inside the width direction of the upper die 33 and the lower die 34. The length of the upper die 33 and the lower die 34 in the length direction of the steel plate S is usually shorter than the length of the steel plate S. In this case, a transport mechanism 31 (see reference...) is used... Figure 14 The steel plate S is intermittently fed along its length to perform multiple end bending processes.

[0065] In the end bending process, the lower die 34, which is in contact with the outer surfaces of the width-direction ends Sc and Sd of the steel plate S to be end bent, has a pressing surface 34a facing the upper die 33. The upper die 33 has a convex curved forming surface 33a facing the pressing surface 34a and having a radius of curvature corresponding to the inner diameter of the manufactured steel pipe. The pressing surface 34a has a concave curved surface that approaches the upper die 33 as it moves outward in the width direction. However, although the pressing surface 34a of the lower die 34 is concave, it can be any surface that approaches the upper die 33 as it moves outward in the width direction, and it can also be an inclined plane. The curved surface shapes of the upper die 33 and the lower die 34 are sometimes designed according to the thickness of the steel plate S, the outer diameter of the steel pipe, etc., and are appropriately selected and used according to the workpiece being processed.

[0066] Figure 15 (b) is related to Figure 15 (a) A cross-section of the stamping mechanism 32A at the same location in the width direction, showing the state in which the lower die 34 is lifted by the hydraulic cylinder 36 and the die is closed. The lower die 34 is lifted by the hydraulic cylinder 36, and the width direction end Sc of the steel plate S is bent into a shape along the arc-shaped forming surface 33a of the upper die 33. The width of the end bending forming (end bending processing width) varies depending on the width of the steel plate S, but is generally about 100 to 400 mm.

[0067] <Bending Process>

[0068] Figure 2 This figure illustrates an example of a process for forming a U-shaped cross-section using a bending device. In the figure, reference numeral 1 indicates a die positioned within the transport path of the steel plate S. The die 1 consists of a pair of left and right rod-shaped members 1a and 1b supporting the steel plate S at point 2 along its transport direction; their spacing ΔD can be varied depending on the size of the steel tube to be formed. Reference numeral 2 indicates a punch that can move towards and away from the die 1. The punch 2 has a punch front end 2a with a downwardly convex processing surface that directly contacts the steel plate S and presses it into a concave shape, and a punch support body 2b connected to and supporting the back of the punch front end 2a. It should be noted that, typically, the maximum width of the punch front end 2a and the width (thickness) of the punch support body 2b are equal.

[0069] When using the bending device with the above-described structure to bend the steel plate S, the steel plate S is placed on the die 1, and the steel plate S is intermittently fed at a specified feed rate while... Figure 3 The technique shown involves successively bending and stamping the steel plate S at three points from both ends towards the center along its width using punch 2. It should be noted that... Figure 3This diagram illustrates the process of forming the shaped body S1 shown in Figure (j) on the right by bending a steel plate S, which has undergone pre-bending at the ends, from top to bottom in the left column (processing the first half (a) to (e)) and then from top to bottom in the middle column (processing the second half (f) to (i)) and feeding the steel plate S. It should be noted that in... Figure 3 In the diagram, the arrows marked on the steel plate S and the punch 2 indicate the direction of movement of the steel plate S and the punch 2 in each process. In addition, in the U-shaped cross-section formed body S1 after this process, the gap between the ends is called the "weld gap".

[0070] Here, the operational parameters that determine the operating conditions of the bending process include the number of stampings, stamping position information, stamping reduction, lower die interval, and punch curvature.

[0071] The number of stamping cycles refers to the total number of times the steel plate is pressed in the width direction using three-point bending stamping. The more stamping cycles, the smoother the U-shaped cross-section becomes, and the higher the roundness of the steel pipe after the expansion process.

[0072] The stamping position information refers to the position of the steel plate being pressed by the punch in the width direction. Specifically, it can be determined by the distance from one end of the steel plate in the width direction or by a distance based on the center of the steel plate in the width direction. The stamping position information is preferably processed as data that is associated with the number of presses (in the order of the first to the Nth presses).

[0073] The punching reduction refers to the amount of material pressed in by the punch at each pressing position. The punching reduction is determined by... Figure 2 The line connecting the uppermost surface of the die 1 is used as a reference, and the amount by which the lower end face of the punch tip 2a protrudes downward from this point is defined. At this time, the pressing amount of the punch tip 2a can be set to a different value for each press; therefore, the number of presses and the pressing amount are preferably processed as data that are linked together. Thus, if the number of presses is set to N, the number of presses, the pressing position information, and the pressing amount are treated as a set of datasets, and the operating conditions of the bending process are determined by 1 to N datasets.

[0074] These datasets are used because, during the bending process, localized changes in the stamping position and punch depth, resulting in a slotted tube, alter the overall cross-sectional shape and affect the roundness of the steel pipe after the expansion process. However, it is not necessary to set all N datasets as input variables for the roundness prediction model described later. Alternatively, conditions with a significant impact on the roundness of the steel pipe after the expansion process can be selected, such as using the initial (1st) or final (Nth) stamping position information and stamping depth of the bending process to generate the roundness prediction model.

[0075] The gap between the lower punches is Figure 2 The spacing between the pair of left and right rod-shaped members 1a and 1b shown is represented by the parameter ΔD in the figure. If the lower die spacing increases, the curvature of the local steel plate will change even with the same stamping reduction, thus affecting the roundness of the steel pipe after the tube expansion process. Therefore, it is preferable to use the lower die spacing set according to the size of the steel pipe to be formed as the operating parameter for the bending process. In addition, when the lower die spacing is changed for each press of the punch, it can also be used as an operating parameter as data that establishes a relationship with the number of stampings.

[0076] Punch curvature refers to the curvature of the tip of the punch used for pressing. A greater punch curvature results in a greater localized curvature imparted to the steel sheet during three-point bending, which in turn affects the roundness of the steel pipe after the expansion process. However, it is difficult to change the punch curvature for each press when forming a single steel sheet. It is preferable to use the punch curvature set according to the dimensions of the steel pipe to be formed as an operating parameter for the bending process.

[0077] <Weld gap reduction process>

[0078] The weld gap reduction process is a process that reduces the weld gap of a U-shaped cross-section formed by a bending process. Bending and compressive forces are applied to bring the ends of the U-shaped cross-section closer together. Even when bending and compressive forces are applied to the U-shaped cross-section, the weld gap will expand due to springback when the load is removed. Therefore, by anticipating springback and applying enhanced bending and compressive forces, the U-shaped cross-section undergoes longitudinal collapse deformation as a whole.

[0079] exist Figure 4 The diagram shows a structural example of an O-ring stamping device commonly used in the process of reducing weld gaps. For example... Figure 4 As shown in (a), the O-type stamping device uses an upper die 3 and a lower die 4 to impart compressive deformation to the U-shaped cross-section formed body S1 in the longitudinal direction. At this time, the surfaces of the upper die 3 and the lower die 4 that are in contact with the U-shaped cross-section formed body S1 are processed into curved shapes. By bringing the upper die 3 and the lower die 4 closer together, the lower part of the U-shaped cross-section formed body S1 is constrained along the curved surface of the lower die 4. Furthermore, because the upper part of the formed body S1, including the ends, receives bending and compressive forces due to the upper die 3, the ends approach each other along the curved surface of the upper die 3.

[0080] Therefore, the weld gap between the circumferentially facing ends is temporarily reduced. Furthermore, by releasing the pressure based on the mold, the weld gap widens due to springback, thus determining... Figure 4(b) shows the final weld gap G of the slotted pipe S2. In this case, the O-punch reduction is a value obtained by subtracting the distance between the uppermost point of the inner surface of the upper die 3 and the lowermost point of the inner surface of the lower die 4 during die pressing from the outer diameter of the target steel pipe. Alternatively, the ratio of the O-punch reduction to the outer diameter of the steel pipe may be referred to as the O-punch reduction rate.

[0081] In addition to the O-punch pressing amount, other operational parameters used to determine the operational conditions for the weld gap reduction process include the O-punch pressing position and the O-punch die R.

[0082] The O-type stamping reduction position refers to the angle formed by the line connecting the end of the weld gap portion of the U-shaped cross-section of the formed body S1 and the center position in the width direction, and the vertical line. Additionally, the O-type stamping die R refers to the curvature of the areas of the upper die 3 and lower die 4 that abut against the formed body S1. Here, the greater the O-type stamping reduction in the O-type stamping device, the greater the curvature near the 3 o'clock and 9 o'clock positions of the formed body S1, and thus the smaller the final roundness of the steel pipe.

[0083] On the other hand, when a closed stamping method is used instead of an O-type stamping device, the device used to form the slit tube S2 from the formed body S1 is... Figure 5 The closed stamping device shown. For example... Figure 5 As shown, the closed stamping device includes lower tools 10a and 10b. The lower tools 10a and 10b are spaced apart from each other and each has a drive mechanism capable of reversing the rotation direction. Furthermore, the lower tools 10a and 10b are supported by spring units 11a and 11b, etc. An upper tool 13 equipped with a punch 12 is arranged opposite to the lower tools 10a and 10b. A pressing force is applied from the outside to the U-shaped cross-section formed body S1 via the punch 12.

[0084] At this point, the U-shaped cross-section forming body S1 is formed into a slotted tube S2 through two steps. In the first step, the forming body S1 is stamped in a manner that, as shown in a dashed line, the area R1 to which bending deformation is desired on the right side of the weld gap G is positioned near the three o'clock position in a clockwork analogy, using rotatable lower tools 10a and 10b. Then, a pressing force is applied by the punch 12, and the load on the punch 12 is removed after the pressing force is applied. Next, as the second step, similar to the first step, the area R2 to which bending deformation is desired on the left side of the weld gap G is stamped in a manner that, as shown in a clockwork analogy, is positioned near the nine o'clock position. Then, a pressing force is applied by the punch 12, and the load on the punch 12 is removed after the pressing force is applied, thereby forming the slotted tube S2.

[0085] Here, the stamping position in the first and second steps refers to the angle between the line connecting the center of the weld gap G and the center of the steel plate in the width direction. Figure 5 (The angle formed by the single-point dashed line and the vertical line). In addition, the pressing force in the first and second steps refers to the pressing force of the punch 12 on the formed body S1.

[0086] <Welding Process>

[0087] Regarding the slotted pipe S2, the end faces of the weld gap are then joined together and welded using a welding machine (joining unit) to form a steel pipe. For example, a welding machine consisting of three types of welding mechanisms can be used: a spot welding machine, an inner surface welding machine, and an outer surface welding machine. In these welding machines, the spot welding machine uses rollers to continuously and tightly press the mating surfaces together in a suitable positional relationship, welding the pressed portion along the entire axial length of the pipe. Next, the tack-pressed pipe is welded from the inner surface of the mating portion by the inner surface welding machine (submerged arc welding), and then welded from the outer surface of the mating portion by the outer surface welding machine (submerged arc welding).

[0088] <Pipe Expanding Process>

[0089] Regarding the steel pipe after the weld gap is welded, a pipe expander is inserted into the inside of the steel pipe to increase the diameter of the steel pipe (the so-called pipe expander). Figure 6 Figures (a) to (c) are examples of the structure of a tube expander. Figure 6 As shown in (a), the pipe expanding device has multiple expanding molds 16 along the circumferential direction of the conical outer peripheral surface 17, each having a curved surface obtained by dividing a circular arc into multiple parts. When expanding a steel pipe using the pipe expanding device, as... Figure 6 As shown in (b) and (c), firstly, the expansion mold 16 is aligned with the expansion start position by moving the steel pipe P using the steel pipe moving device, and the first expansion process is performed by retracting the pull rod 18 from the expansion start position.

[0090] As a result, the expanding molds 16, which slide in contact with the conical outer peripheral surface 17 through wedge action, are displaced radially, and the steel pipe P is expanded. Furthermore, the unevenness of the cross-sectional shape of the steel pipe P decreases, and the cross-sectional shape of the steel pipe P approaches a true circle. Next, the pull rod 18 is advanced to the expanding start position, and the expanding molds 16 are reset to their axially perpendicular inward position using the release mechanism. The steel pipe P is then moved further by an amount corresponding to the spacing (axial length) of the expanding molds 16. Then, the aforementioned actions are repeated after aligning the expanding molds 16 with the new expanding position. Thus, the first expanding process can be performed sequentially along the entire length of the steel pipe P according to the spacing of the expanding molds 16.

[0091] Operational parameters that determine the operating conditions of the pipe expansion process at this stage include the expansion ratio, the number of expansion dies, and the diameter of the expansion die. The expansion ratio is the ratio of the difference between the outer diameter after expansion and the outer diameter before expansion to the outer diameter before expansion. The outer diameters before and after expansion can be calculated by measuring the circumference of the steel pipe. The expansion ratio can be adjusted by the stroke of the expansion die in the radial direction. The number of expansion dies refers to the number of circumferentially arranged portions that abut against the steel pipe during expansion. The diameter of the expansion die refers to the curvature of the portion of each expansion die that abuts against the steel pipe.

[0092] Among the operational parameters that allow for easy adjustment of the roundness after the tube expansion process, the expansion ratio is crucial. Increasing the expansion ratio ensures that the curvature of the area in contact with the expansion die is evenly distributed throughout the entire circumference, according to the expansion die R, thereby improving roundness. Furthermore, a greater number of expansion dies effectively suppresses localized circumferential curvature variations in the steel pipe, resulting in better roundness after the expansion process.

[0093] However, if the expansion ratio is too large, the compressive yield strength of the steel pipe product may decrease due to the Bauschinger effect. When steel pipes are used in line pipes, high compressive stresses act on the circumference of the pipe; therefore, the steel pipe material itself requires high compressive yield strength, and unnecessarily increasing the expansion ratio is inappropriate. Therefore, in actual operation, the expansion ratio is set to a value lower than the pre-set upper limit, ensuring that the roundness of the steel pipe falls within the specified range.

[0094] <Roundness Measurement Procedure>

[0095] In the final inspection step of the steel pipe manufacturing process, a quality inspection is performed on the steel pipe, and its roundness is measured. The roundness measured in the roundness measurement step is an index of the degree to which the outer diameter shape of the steel pipe deviates from a perfect circle. Generally, the closer the roundness is to zero, the closer the cross-sectional shape of the steel pipe is to a perfect circle. Roundness is calculated based on the outer diameter information of the steel pipe measured by a roundness measuring machine. For example, by dividing the pipe circumferentially into equal parts at any point along its length and measuring the outer diameter at opposite positions, and setting the maximum and minimum diameters as Dmax and Dmin respectively, the roundness can be defined as Dmax-Dmin. In this case, the more equal parts there are, the more numerically quantifiable the small irregularities in the steel pipe after the expansion process becomes, which is preferable. Specifically, it is best to use information divided into 4 to 36,000 equal parts. More preferably, it is divided into 360 or more equal parts.

[0096] However, roundness may not be based on the difference between the maximum and minimum diameters. An equivalent hypothetical round circle (diameter) with the same area as the inner side of the curve can be calculated from a graph representing the outer diameter shape of the steel pipe as a continuous line graph. This hypothetical round circle is then used as a reference to define the area deviating from the outer diameter shape of the steel pipe as the content of the image. For example, the following methods can be used to determine the outer diameter shape of the steel pipe.

[0097] (a) such as Figure 7 As shown in (a), a device is used that has an arm 20 capable of rotating 360 degrees about the approximate central axis of the steel pipe P, displacement gauges 21a and 21b mounted on the front end of the arm 20, and a rotation angle detector 22 that detects the rotation angle of the rotation axis of the arm 20. The distance between the rotation center of the arm 20 and the measuring point on the outer periphery of the steel pipe P is measured by displacement gauges 21a and 21b every small angular unit of rotation of the arm 20. The outer diameter shape of the steel pipe P is determined based on the measured value.

[0098] (b) such as Figure 7 As shown in (b), a device is used that includes a rotating arm 25 that rotates about the central axis of the steel pipe P, a frame (not shown) that is provided at the end of the rotating arm 25 so as to be movable in the radial direction of the steel pipe P, a pair of pressing rollers 26a and 26b that abut against the outer and inner surfaces of the end of the steel pipe P and rotate with the rotation of the rotating arm 25, and a pair of pressing cylinders that press the pressing rollers 26a and 26b against the outer and inner surfaces of the steel pipe P and are fixed relative to the frame. The outer diameter shape of the steel pipe P is determined based on the amount of radial movement of the frame and the pressing position of the pressing rollers 26a and 26b implemented by each pressing cylinder.

[0099] In this embodiment, the prediction accuracy of the roundness prediction model described later can be verified by comparing it with the measured roundness value obtained in the aforementioned inspection process. Therefore, the prediction accuracy can also be improved by adding the actual value of the prediction error of the roundness prediction model to the prediction result of the roundness prediction model described later.

[0100] [Device for generating a roundness prediction model for steel pipes]

[0101] Figure 8 This is a block diagram showing the structure of a steel pipe roundness prediction model generation device as an embodiment of the present invention. Figure 9 It is shown Figure 8 A block diagram showing the structure of the offline roundness calculation unit 112.

[0102] like Figure 8As shown, the roundness prediction device 100 for steel pipes according to one embodiment of the present invention is composed of an information processing device such as a workstation, and includes a basic data acquisition unit 110, a database 120 and a roundness prediction model generation unit 130.

[0103] The basic data acquisition unit 110 includes an operation condition dataset 111 obtained by quantifying the factors that affect the roundness of the steel pipe after the bending process, the weld gap reduction process, the welding process, and the pipe expansion process, and an offline roundness calculation unit 112 that uses the operation condition dataset 111 as input conditions to output the roundness after the pipe expansion process.

[0104] In this embodiment, the operating condition dataset 111 includes at least the attribute information of the steel plate that becomes the billet, the operating parameters of the bending process, and the operating parameters of the weld gap reduction process. This is because these are factors that have a significant impact on the roundness of the steel pipe after the pipe expansion process and can affect the deviation of the roundness. However, it may also include the operating parameters of the welding process and the operating parameters of the pipe expansion process. The data used in the operating condition dataset 111 will be described later.

[0105] The basic data acquisition unit 110 calculates the roundness of the steel pipe after the expansion process corresponding to multiple operating condition datasets 111 by changing various parameters contained in the operating condition dataset 111 and performing numerical calculations by the roundness offline calculation unit 112. The range of parameters to be changed in the operating condition dataset 111 is determined based on the size of the manufactured steel pipe, the specifications of the equipment for each process, etc., and is based on the range that can be changed as normal operating conditions.

[0106] The offline roundness calculation unit 112 calculates the shape of the steel pipe after the expansion process, which involves a series of manufacturing processes from bending to expansion, based on numerical analysis. The roundness of the steel pipe is then determined based on this expanded shape. These manufacturing processes include bending, weld gap reduction, and expansion. Figure 9 As shown, the roundness offline calculation unit 112 includes finite element model generation units 112a to 112c corresponding to each process and a finite element analysis solver 112d. It should be noted that the roundness offline calculation unit 112 may also include a finite element model generation unit corresponding to the end bending process.

[0107] In the case where the finite element model generation unit corresponding to the end bending process is included in the roundness offline calculation unit 112, the finite element model generation unit for the end bending process performs element segmentation within the steel plate based on the steel plate's property information. Element segmentation is performed automatically based on pre-set element segmentation conditions. The finite element model of the end bending process after element segmentation, along with the calculation conditions in the end bending process, is sent to the finite element analysis solver 112d. The calculation conditions in the end bending process include the operation parameters of the end bending process, as well as all information required for performing finite element analysis, including determining the physical property values, geometric boundary conditions, mechanical boundary conditions, etc. of the workpiece, tools, etc. The shape and stress-strain distribution of the steel plate obtained through the finite element analysis of the end bending process are sent to the finite element model generation unit 112a for the bending process as initial conditions related to the workpiece in the bending process.

[0108] As for the finite element analysis solver 112d, since many commercially available general-purpose analysis software programs exist, they can be utilized by appropriately selecting and incorporating them. Alternatively, the finite element analysis solver 112d can be mounted on a computer independent of the roundness offline calculation unit 112, and input data including the finite element model and output data as calculation results can be exchanged between the two. This is because if the finite element model corresponding to each process is generated, numerical analysis can be performed using a single finite element analysis solver.

[0109] The finite element method (FEM) is an approximate solution method that divides a continuum into a finite number of elements. Although it is an approximate method, the FEM finds solutions that satisfy the equilibrium of forces and the continuity of displacements at the nodes of the elements, and can obtain highly accurate solutions even when the deformation is non-uniform. In the FEM, stress, strain, and displacement within each element are defined independently and are formulated as a problem of solving simultaneous equations by establishing a relationship with the displacements (velocities) at the nodes. Therefore, the method of setting the displacements (velocities) at the nodes of the elements as unknowns and evaluating the strain (increment) and stress is widely used.

[0110] Furthermore, the finite element method (FEM) is characterized by its calculation based on the principle of virtual work expressed in integral form, relative to the stress equilibrium conditions within an element. The accuracy of the analytical results varies depending on conditions such as element segmentation. Additionally, the computation time required for analysis is typically long. However, the FEM is characterized by its ability to provide solutions to problems that are difficult to solve using other methods, as it satisfies the fundamental equations of plasticity within nodes or elements. Therefore, even for complex machining processes in steel pipe manufacturing, solutions to the displacement, stress, and strain fields of the workpiece can be obtained, closely resembling actual phenomena.

[0111] It should be noted that a portion of the finite element analysis solver can also be replaced with various numerical analytical methods or approximate solutions, such as the slip line field method or the energy method. This can shorten the overall computation time. Furthermore, the finite element analysis used in this embodiment performs elasto-plastic analysis and does not include temperature field analysis such as thermal conduction analysis. However, in cases where the processing speed is high and the temperature rise of the workpiece due to processing heat is large, analysis obtained by coupling thermal conduction analysis and elasto-plastic analysis can be performed. Additionally, the elasto-plastic analysis in this embodiment is a two-dimensional cross-sectional analysis regardless of whether it pertains to the bending process, the weld gap reduction process, or the pipe expansion process. Numerical analysis is sufficient for the cross-section of the stable portion along the length direction when the steel plate is formed into a U-shaped cross-section, a slotted pipe, or a steel pipe. However, in cases where the shape of unstable portions such as the front and rear ends of the steel pipe needs to be predicted with high accuracy, a finite element model generation unit that performs three-dimensional analysis including the front and rear ends can also be included.

[0112] Regarding the steel sheet that is the workpiece in the bending process, its attribute information is given as input data. In this case, if the bending process includes an end-bending process as a preceding process, the shape and stress-strain distribution of the steel sheet obtained from the finite element analysis of the end-bending process become the initial conditions for the workpiece in the bending process. Here, the finite element model generation unit 112a of the bending process performs element segmentation within the steel sheet based on the dimensions and shape of the steel sheet before the bending process. Element segmentation is performed automatically based on pre-set element segmentation conditions. Alternatively, the distribution of residual internal stress and strain can be allocated to each element based on the manufacturing history assigned to the steel sheet by the preceding process. This is because, in the bending process, where bending is the main function, the initial residual stress also affects the shape of the U-shaped formed body of the processed steel sheet.

[0113] Along with the finite element model of the bending process thus generated, the calculation conditions of the bending process are sent as input data to the finite element analysis solver 112d. At this time, the calculation conditions of the bending process include the operating parameters of the bending process, as well as all the information required to perform finite element analysis, including the physical property values ​​of the workpiece, tools, etc., geometric boundary conditions, mechanical boundary conditions, and all other boundary conditions.

[0114] In the finite element analysis solver 112d, numerical analysis is performed under the aforementioned calculation conditions to determine the shape of the U-shaped body after the bending process and the distribution of residual stress and strain within it. The results of this calculation are used as input data for the next weld gap reduction process in the roundness offline calculation unit 112. Based on the calculated shape after the bending process, the internal features of the U-shaped body are segmented in the finite element model generation unit 112b for the weld gap reduction process. Feature segmentation is performed automatically based on pre-set feature segmentation conditions. At this time, it is preferable to distribute the stress and strain distribution calculated in the previous process to each feature, for the same reasons as described above.

[0115] Along with the finite element model of the weld gap reduction process generated in this way, the calculation conditions in the weld gap reduction process are sent as input data to the finite element analysis solver 112d. At this time, the calculation conditions in the weld gap reduction process include the operating parameters of the weld gap reduction process, as well as all the information required to perform finite element analysis, including determining the physical properties, geometric boundary conditions, mechanical boundary conditions, and other boundary conditions of the workpiece, tools, etc.

[0116] In the finite element analysis solver 112d, numerical analysis is performed under the aforementioned calculation conditions to determine the shape of the weld gap reduction tube and the distribution of residual internal stress and strain. The results of this calculation are then used as input data in the finite element model generation unit 112c for the next tube expansion process. Furthermore, the welding process of welding the weld gap portion of the weld gap tube can also be analyzed numerically to determine the residual stress and strain generated in the welded steel tube.

[0117] However, rigorous numerical analysis of the welding process is often difficult due to factors such as heat conduction behavior associated with the melting of the steel plate during welding and the influence of the heat-affected zone on mechanical properties. Furthermore, the heat-affected zone only affects the shape of a portion of the steel pipe, having a minor impact on the overall shape. Therefore, the influence of the heat-affected zone of the steel plate during welding on the roundness of the steel pipe after the expansion process can also be ignored.

[0118] During the welding process, the welded tube is bound from the outside while being welded to reduce the weld gap. Therefore, in the area outside the weld gap, the stress and strain distribution changes based on elastic deformation. Thus, using the finite element analysis solver 112d, the behavior of binding the tube from the outside to achieve a zero weld gap can be numerically analyzed using the finite element method, and the results can be set as the stress-strain state after the welding process.

[0119] On the other hand, when the reduction of the weld gap in such a welding process is an elastic deformation, the analytical solution of stress and strain relative to the bending beam based on beam theory can be superimposed on the stress and strain distribution inside the slotted tube calculated analytically by finite element method, thus obtaining the stress and strain distribution after the welding process. This can shorten the calculation time.

[0120] Based on the shape of the steel pipe obtained after the welding process as described above, the finite element model generation unit 112c of the pipe expansion process performs element segmentation inside the steel pipe. Element segmentation is performed automatically based on pre-set element segmentation conditions. At this time, it is preferable to distribute the stress and strain distributions calculated as described above to each element. The generated finite element model of the pipe expansion process, along with the calculation conditions in the pipe expansion process, is sent to the finite element analysis solver 112d. The calculation conditions in the pipe expansion process include the operating parameters of the pipe expansion process in this embodiment, as well as all information required for performing finite element analysis, including determining the physical property values ​​of the workpiece, tools, etc., geometric boundary conditions, mechanical boundary conditions, and all other boundary conditions.

[0121] In the finite element analysis solver 112d, numerical analysis is performed under the calculation conditions given above to determine the shape of the steel pipe after the expansion process and the distribution of internal stress and strain. The calculated shape of the steel pipe has a non-uniform curvature distribution in the circumferential direction, and the roundness of the steel pipe is determined according to the definition of roundness in the roundness measurement process. It should be noted that in the numerical analysis using the finite element method in the roundness offline calculation unit 112, the calculation time is sometimes about 1 to 10 hours for one operating condition dataset (1 example).

[0122] However, since the processing is performed offline, there is no constraint on computation time. Nevertheless, to shorten the computation time relative to many operational condition datasets, multiple computers can be used to perform numerical computations corresponding to multiple operational condition datasets in parallel. This allows for the rapid construction of a database for generating roundness prediction models. Furthermore, in recent years, the use of GPGPU (General-Purpose computing on Graphics Processing Units) has reduced the computation time per instance to approximately 1 / 2 to 1 / 10 compared to previous methods, making such computing tools readily available.

[0123] return Figure 8Database 120 stores the operating condition dataset 111 and the corresponding data related to the roundness of the steel pipe after the pipe expansion process. The data stored in database 120 can be obtained offline. Unlike databases that accumulate actual operational performance values, this database allows for arbitrary setting of operating condition datasets. Therefore, statistical bias is less likely to occur in the operating conditions of the dataset, making it suitable for machine learning. Furthermore, the data is accumulated based on rigorous numerical analysis calculations, not time-varying learning data; therefore, the more data accumulated, the more beneficial the database becomes.

[0124] The roundness prediction model generation unit 130 generates a roundness prediction model M, learned through machine learning, based on the relationship between multiple sets of operating condition datasets 111 stored in database 120 and the roundness of the steel pipe. This model calculates the roundness of the steel pipe after the expansion process relative to the input operating condition datasets 111. It should be noted that the relationship between the operating conditions in each process and the roundness of the steel pipe after the expansion process sometimes exhibits complex nonlinearity. Modeling based on a first-order linear assumption results in low accuracy. However, by using machine learning techniques with nonlinear functions, such as neural networks, high-precision predictions can be achieved. Here, modeling means replacing the input-output relationship in numerical calculations with an equivalent functional form.

[0125] The number of data points required to generate the roundness prediction model M varies depending on the size of the manufactured steel pipes, but 500 or more data points are sufficient. Preferably, 2000 or more data points are used, and more preferably, 5000 or more data points are used. Known learning methods can be used for machine learning. For example, known machine learning techniques such as neural networks can be used. Other methods include decision tree learning, random forest, Gaussian process regression, support vector machine regression, and k-nearest neighbor algorithm. It should be noted that the roundness prediction model M is generated offline, but the roundness prediction model generation unit 130 can also be integrated into an online control system, using a database that is constantly computed and accumulated offline, to periodically update the roundness prediction model.

[0126] The roundness prediction model M for steel pipes after the expansion process, generated as described above, has the following characteristics: The properties of the steel plate used as the billet, such as yield stress and plate thickness, deviate during steel plate manufacturing. These properties affect the curvature of the steel plate and its curvature after load removal during the three-point bending stamping process. Therefore, by selecting these steel plate properties as input parameters for the offline-generated roundness prediction model M, the influence of the billet's yield stress, plate thickness, and other properties on the roundness of the steel pipe after the expansion process can be predicted. Furthermore, the weld gap reduction process also involves applying bending and compressive forces using molds, etc. The yield stress and plate thickness change the curvature of the steel plate after load removal; therefore, these are also used as input parameters for the roundness prediction model M.

[0127] Furthermore, the bending process involves imbuing the steel plate with discontinuous curvature multiple times along its width, thus creating a localized distribution of curvature along the plate's width. Subsequently, if a combined deformation of compression and bending is applied, as in the weld gap reduction process, the bending moment acting on a so-called "bending beam" varies due to the beam's curvature before deformation. Similarly, the bending moment imposed by the weld gap reduction process will be locally distributed based on the localized curvature distribution of the steel plate imposed by the bending process. Therefore, the operating conditions of the bending process affect the curvature distribution of the steel plate along its width after the weld gap reduction process. At this point, it is meaningful to use both the operating parameters of the bending process and the weld gap reduction process as input parameters for the roundness prediction model M.

[0128] For example, Figure 10 The results were obtained by measuring the roundness of a steel pipe with an outer diameter of 30 inches and a thickness of 44.5 mm after an expansion process (under the same operating conditions) during the manufacturing of a steel pipe. The expansion process involved setting the number of punching cycles to 9 during the bending operation, and using an O-punch device to adjust the O-punch reduction rate as a process to reduce weld gap. Figure 10 The diagram shows the results of setting other operating conditions in the bending process to constant while changing the three levels of the final (9th) pressing amount (final pass pressing amount).

[0129] like Figure 10As shown, it can be seen that among the O-type stamping reduction rates representing the stamping amount in the weld gap reduction process, there exists an optimal value for reducing the roundness of the steel pipe after the expansion process. However, this optimal value varies depending on the final stamping reduction amount of the bending process, which is an operating condition of the bending process. That is, it can be seen that in order to reduce the roundness of the steel pipe after the expansion process, the operating conditions of the weld gap reduction process need to be changed according to the operating conditions of the bending process. However, if we only consider the operating conditions of the bending process and the weld gap reduction process as independent parameters affecting the roundness of the steel pipe after the expansion process, it may not be possible to set appropriate operating conditions.

[0130] In contrast, the roundness prediction model of this embodiment can consider the influence of the operating parameters of multiple manufacturing processes on the roundness of the steel pipe after the expansion process, and can achieve high-precision roundness prediction. Furthermore, since the roundness prediction model is generated through machine learning, even if the variables used as input conditions are changed, the roundness of the output can be calculated immediately. Therefore, it has the feature that operating conditions can be set and corrected immediately even when used online. The parameters used as inputs to the roundness prediction model will be explained below.

[0131] <Steel plate property information>

[0132] As for the property information of the steel sheet that becomes the billet, any parameter that affects the roundness of the steel pipe after the expansion process can be used, such as the yield stress, tensile strength, longitudinal elastic modulus, plate thickness, thickness distribution within the plate surface, yield stress distribution in the thickness direction of the steel sheet, degree of Bauschinger effect, and surface roughness. In particular, it is preferable to use the deformation state of the steel sheet caused by the three-point bending stamping in the bending process and the factor affecting springback, as well as the deformation state of the steel sheet caused by the compression and bending processing in the weld gap reduction process and the factor affecting springback, as indicators.

[0133] The yield stress, thickness distribution, and thickness of the steel plate directly affect the stress and strain state during three-point bending stamping. Tensile strength, as a parameter reflecting the work hardening state during bending, influences the stress state during bending deformation. The Bauschinger effect affects the yield stress and subsequent work hardening behavior under load reversal during bending deformation, thus impacting the stress state during bending deformation. Furthermore, the longitudinal elastic modulus of the steel plate affects the springback behavior after bending. Moreover, the thickness distribution within the plate surface influences the distribution of bending curvature during the pressing process, and surface roughness affects the friction state between the die and the steel plate in the weld gap reduction process, thereby affecting the roundness of the steel pipe after the expansion process.

[0134] Among these attribute information, yield stress, representative plate thickness, plate thickness distribution information, and representative plate width are particularly preferred. These are information measured during the quality inspection process of the steel plate manufacturing process, namely the thick plate rolling process, which produces the billet. They affect the deformation behavior in the bending process and the weld gap reduction process, and affect the roundness of the steel pipe after the expansion process. Therefore, they are preferred to be used as the attribute information of the steel plate in the basic data acquisition unit 110.

[0135] Yield stress is information obtained from tensile tests of small specimens collected from a thick steel plate used for quality verification, and a representative value within the plane of the steel plate can be used. Furthermore, the representative plate thickness is the plate thickness representing the in-plane thickness of the steel plate. This can be the plate thickness at any point along the length of the plate, the thickness at the center of the width direction, or the average value of the length direction thickness. Alternatively, the average value of the overall in-plane thickness of the steel plate can be calculated and used as the representative plate thickness.

[0136] In addition, plate thickness distribution information refers to information representing the thickness distribution along the width direction of the steel plate. A representative example is the convexity of the steel plate. Convexity represents the difference in plate thickness between the center of the steel plate along its width direction and a position located a specified distance (e.g., 100mm, 150mm, etc.) from the edge of the steel plate along its width direction. Furthermore, the representative plate width is a representative value relating to the width of the steel plate used as a billet. Deviations in the width of the thick steel plate used as a billet, and variations in the width of the steel plate during grinding through beveling, can affect the outer diameter accuracy of the finished steel pipe.

[0137] The above-mentioned steel plate attribute information, collected by a host computer during online operations, is used to set the operating conditions in the steel pipe manufacturing process. The basic data acquisition unit 110 preferably selects from this steel plate attribute information in a manner consistent with that collected by the online host computer.

[0138] <Operating parameters for end bending process>

[0139] When the operating parameters of the end bending process are used as input to the roundness prediction model, parameters that determine the shape of the forming surface 33a of the upper die 33 and the pressing surface 34a of the lower die 34 used in the C-type stamping apparatus 30 can be used as operating parameters. Additionally, the end bending processing width (width for end bending forming), the lifting force (C-type stamping force), and the holding force of the clamping mechanism 37 in the end bending process can also be used as operating parameters. This is because these are factors that can affect the deformation of the end of the steel sheet in the width direction during the end bending process. Furthermore, when performing three-dimensional deformation analysis regarding the end bending process, the feed rate, feed direction, and number of feeds of the steel sheet can also be set as operating parameters for the end bending process.

[0140] Regarding the shape of the forming surface 33a of the upper mold 33, there are cases where it is given by a shape formed by continuous arcs with multiple radii of curvature, or by an involute, etc., and parameters for determining the geometric cross-sectional shape can be used. For example, when the cross-sectional shape is formed by using a parabolic shape, the cross-sectional shape can be determined by using the coefficients of the first and second terms of the quadratic equation representing the parabola passing through the origin, and such coefficients can be set as operating parameters for the end bending process.

[0141] On the other hand, when multiple molds are held and used to replace the forming surface 33a of the upper mold 33 according to the outer diameter, wall thickness, and steel type of the manufactured steel pipe, the mold management number used to determine the mold used in the end bending process can be set as the operation parameter of the end bending process.

[0142] <Operating parameters for bending process>

[0143] In this embodiment, the operating parameters of the bending process are used as inputs to the roundness prediction model. As operating parameters for the bending process, various parameters that affect the local bending curvature of the steel sheet and its distribution in the width direction, such as the number of punches, punching position information, punching reduction, lower die interval, and punch curvature described above, can be used. In particular, it is preferable to use all information including the punching position information, punching reduction, and the number of punches performed through the bending process. Examples of this "all information" include... Figure 11 The method shown.

[0144] Figure 11(a) and (b) show examples of the pressing position and pressing amount when the punch was pressed 16 times and 10 times respectively relative to a steel plate of the same width. In this case, the pressing position is information indicating the distance from the end of the steel plate in the width direction (as referenced), and this is used as the pressing position information. On the other hand, the pressing amount is recorded corresponding to each pressing position, and these "number of pressings," "pressing position," and "pressing amount" can be set as a set of data. Figure 11 In the examples shown in (a) and (b), the operating parameters of the bending process are determined by 16 and 10 sets of data, respectively, for 16 and 10 stamping cycles.

[0145] In this embodiment, such a dataset is used as input to the roundness prediction model in the following form. For example, as input to the roundness prediction model, the stamping position and stamping amount when stamping is performed at the position closest to the end on one side of the steel plate, and the stamping position and stamping amount when stamping is performed at the position closest to the end on the other side of the steel plate, can be used.

[0146] In three-point bending stamping, when the stamping reduction at one end of the steel plate is increased, Figure 2 The curvature of the steel pipe increases at approximately the 1 o'clock and 11 o'clock positions, resulting in a transversely elongated shape as a U-shaped cross-section. Furthermore, the closer these stamping positions are to the end of the steel plate, the lower the weld gap, again resulting in a transversely elongated shape as a U-shaped cross-section. As a result, the steel pipe, after being formed into a slotted pipe and undergoing welding and expansion processes, retains a transversely elongated shape, affecting its roundness. Moreover, the curvature of the punch during stamping, the overall number of stamping operations, and the spacing of the lower die during stamping also affect roundness.

[0147] On the other hand, by using all the stamping reduction position information and stamping reduction amount data along with the number of stamping operations as input to the roundness prediction model, the prediction accuracy of the roundness prediction model can be further improved. For example, based on the assumed maximum number of stamping operations, when stamping is performed, the stamping reduction position and stamping reduction amount data are saved according to the number of stamping operations. Furthermore, the stamping reduction position and stamping reduction amount in subsequent stamping processes where no stamping is performed are set to zero. For example, in... Figure 11 In the examples shown in (a) and (b), when the maximum number of stampings is assumed to be 16, the data for the 11th to 16th stampings are set to zero when the number of stampings is 10, and become the input of the roundness prediction model.

[0148] The above-mentioned bending process operation parameters are used as operating conditions set by the host computer during online operation. The basic data acquisition unit 110 preferably selects parameters from the bending process operation parameters collected by the online host computer to use as input for the roundness prediction model.

[0149] <Operating parameters for reducing weld gap>

[0150] In this embodiment, the operating parameters of the weld gap reduction process are used as inputs to the roundness prediction model. When using an O-punch device as the weld gap reduction process, the operating parameters can be the O-punch reduction amount, the O-punch reduction position, and the O-punch die R. On the other hand, when using the closed punching method, the operating parameters are the closed punch reduction position and the closed punch pressing force from each of the above steps. In particular, when using an O-punch device, it is preferable to use the O-punch reduction amount. This is because if the O-punch reduction amount is increased, the area between the point where the upper die receives the restraint and pressing force and the point where the lower die restrains the force (mainly near the 3 o'clock and 9 o'clock positions of the steel pipe) is not restrained, and the bending and compression deformation is concentrated. Therefore, the increased curvature in this area will affect the final roundness.

[0151] The above-mentioned operation parameters for reducing weld gap are used as operating conditions set by the host computer during online operation. The basic data acquisition unit 110 preferably selects parameters from the operation parameters for reducing weld gap collected by the online host computer to be used as input for the roundness prediction model.

[0152] <Operating parameters for the tube expansion process>

[0153] In addition to the operating parameters mentioned above, when using the operating parameters of the tube expansion process as input to the roundness prediction model, the tube expansion ratio can be used as an operating parameter for the tube expansion process. A higher tube expansion ratio results in a higher roundness of the steel pipe after the tube expansion process. However, from the viewpoint of compressive yield strength of the steel pipe product, the upper limit of the tube expansion ratio is limited. Therefore, values ​​within this range are used to determine the calculation conditions of the basic data acquisition unit 110. At this time, the tube expansion ratio is information required to control the tube expansion device, and therefore can be determined by setting a value set by the host computer. It should be noted that, in addition to the tube expansion ratio, the number of tube expansion dies and the diameter of the tube expansion die can also be used as operating parameters for the tube expansion process.

[0154] [Roundness Prediction Method]

[0155] In this embodiment, a roundness prediction model M, generated offline by the roundness prediction model generation unit 130 as described above, is used to predict the roundness of the steel pipe after the expansion process online. When predicting the roundness of the steel pipe after the expansion process, firstly, an operating condition dataset (operating parameter acquisition step) is obtained online, which sets the operating conditions for the steel pipe manufacturing process. This step involves obtaining the necessary data from the host computer overseeing the steel pipe manufacturing process or the control computer for each forming process, as the operating condition dataset that becomes the input to the roundness prediction model generated as described above. Here, "online" means the period from before the start of the steel pipe manufacturing process to the completion of the expansion process, encompassing a series of manufacturing processes. Therefore, it is not necessary for processing to be performed in any forming process. The period of waiting to transport the steel plate to the next process during each forming process is also included in "online". Furthermore, the period before the start of the steel pipe manufacturing process and after the completion of the thick plate rolling process that produces the steel plate as a billet can also be included in "online". This is because: once the thick plate rolling process, which produces steel plates as billets, is completed, the system becomes capable of obtaining the operating condition dataset that serves as the input to the roundness prediction model of this embodiment. The roundness prediction model M, learned through machine learning, is used online. By setting the operating parameters as input conditions, the roundness of the output can be calculated immediately, allowing for rapid resetting of the operating conditions, etc.

[0156] The roundness prediction of the steel pipe after the tube expansion process can be performed at any time before or during the steel plate manufacturing process. Depending on the timing of the prediction, an appropriate set of operating conditions is generated as input to the roundness prediction model M. That is, when the roundness prediction of the steel pipe after the tube expansion process is performed before the bending process, actual values ​​(measured values) of the property information of the steel plate that becomes the billet can be used as operating parameters for subsequent manufacturing processes, including the bending process, using preset values ​​of operating conditions established by the host computer.

[0157] Furthermore, when predicting the roundness of the steel pipe after the expansion process following the bending process but before the weld gap reduction process begins, the actual values ​​(measured values) of the property information of the steel plate that becomes the billet and the actual values ​​of the operating parameters of the bending process are used. Additionally, the preset values ​​of the operating conditions, pre-set by the host computer, are used as the operating parameters for subsequent manufacturing processes, including the weld gap reduction process. It should be noted that the preset operating condition values ​​are set based on past operating performance and are therefore preset values ​​stored in the host computer.

[0158] In this embodiment, as described above, a set of operating condition datasets obtained based on the timing of predicting the roundness of the steel pipe after the expansion process is used as input to the roundness prediction model, and the roundness of the steel pipe after the expansion process is predicted online as the output. Therefore, the operating conditions of subsequent manufacturing processes can be reset based on the predicted roundness of the steel pipe, thus making the roundness of the steel pipe after the expansion process even smaller.

[0159] [Roundness Control Method]

[0160] The method for controlling the roundness of the steel pipe after the expansion process in this embodiment will be explained. Figure 12 This illustrates the process of predicting the roundness of the steel pipe after the expansion process, performed before the bending process begins. For example... Figure 12 As shown, in this process, actual data related to the property information of the steel plate that becomes the billet, preset values ​​(operational setting values) for the operation conditions of the bending process, and preset values ​​(operational setting values) for the operation conditions of the weld gap reduction process are obtained from the host computer 140, and this information is acquired as the operation condition dataset 111. In addition, a preset roundness target value for the roundness of the steel pipe after the pipe expansion process is sent from the host computer 140 to the operation condition resetting unit 150.

[0161] The obtained operating condition dataset 111 is used as input to the roundness prediction model M to predict the roundness of the steel pipe after the pipe expansion process. Then, the predicted roundness (predicted roundness value) is compared with the target roundness (target roundness value). If the predicted roundness is smaller than the target roundness value, the steel pipe is manufactured without changing the set values ​​of the operating conditions for the bending process, the weld gap reduction process, and the pipe expansion process. On the other hand, if the predicted roundness is larger than the target roundness value, the operating conditions for the bending process or the weld gap reduction process are reset.

[0162] Specifically, the operating conditions for the bending process are reset by increasing the number of stamping operations by one or two times or more and shortening the interval between stamping positions. This improves the roundness of the steel pipe after the expansion process. Furthermore, the reset operating conditions for the bending process can be used again as input data for the roundness prediction model M to predict roundness again, confirming whether the predicted roundness is smaller than the target roundness value, and thus determining the reset values ​​for the bending process's operating conditions.

[0163] Then, the re-set operating conditions for the bending process are sent to the bending process operating condition control unit, which determines the operating conditions for the bending process. By repeating the roundness determination in the operating condition re-setting unit 150 multiple times, even if the roundness target value is set small, appropriate operating conditions for the bending process can be set, and steel pipes with better roundness can be manufactured.

[0164] On the other hand, when resetting the operating parameters for the weld gap reduction process, the O-punch reduction amount is changed. For example, multiple O-punch reduction amount conditions are input into the roundness prediction model M, and the condition that yields the best roundness is selected and set. The reset operating conditions for the weld gap reduction process are then sent to the weld gap reduction process operating condition control unit, which determines the operating conditions for the weld gap reduction process. It should be noted that both the operating parameters for the bending process and the operating parameters for the weld gap reduction process can be reset. The resetting method is the same as described above. The more operating parameters are reset, the wider the control range for the roundness of the steel pipe after the pipe expansion process becomes, and the smaller the roundness can be.

[0165] As described above, according to the roundness control method of the present invention, since a roundness prediction model is used that can simultaneously consider the influence of the deviation of the property information of the steel plate that will become the billet, the bending process and the weld gap reduction process on the roundness, it is possible to set appropriate operating conditions for making the steel pipe after the expansion process have good roundness, and the operating conditions can be quickly reset by applying it online, thus enabling the manufacture of steel pipes with high roundness.

[0166] Next, refer to Table 1 and Figure 16 As an embodiment of the present invention, the roundness control method is described in the case where the roundness control method includes the end bending process of the steel plate before the bending process.

[0167] In this embodiment, firstly, a re-setting target process is selected from among multiple forming processes constituting the steel pipe manufacturing process. Then, before the start of the re-setting target process, the roundness prediction model M is used to predict the roundness of the steel pipe after the expansion process. Next, in order to reduce the roundness of the steel pipe after the expansion process, at least one or more operating parameters selected from the operating parameters of the re-setting target process or at least one operating parameter selected from the operating parameters of the forming process downstream of the re-setting target process are re-set.

[0168] Here, the multiple forming processes constituting the steel pipe manufacturing process refer to the end bending process, pressure bending process, weld gap reduction process, and pipe expansion process, which plastically deform the steel plate to form the steel pipe into a specified shape. Regarding the re-setting target process, any one of these forming processes is selected. Furthermore, before performing the forming process in the selected re-setting target process, the roundness prediction model M of the steel pipe is used to predict the roundness of the steel pipe after the pipe expansion process. At this time, regarding the forming process upstream of the re-setting target process, since the forming process of the steel plate has already been completed, the performance data of the upstream forming process can be used as the input to the roundness prediction model M when using the operating parameters of the upstream forming process. On the other hand, regarding the forming process downstream of the re-setting target process, since operational performance data cannot be collected, the preset values ​​set in a host computer or the like are used as the input to the roundness prediction model M of the steel pipe. In this way, the roundness of the steel pipe after the pipe expansion process for the target part can be predicted.

[0169] Then, it is determined whether the predicted roundness of the steel pipe after the expansion process falls within the allowable roundness for the product. Therefore, if the roundness of the steel pipe after the expansion process is smaller than the predicted value, the operating conditions in the resetting target process and the forming process downstream of the resetting target process can be reset. Here, the reset operating parameters can be either the operating parameters in the resetting target process or the operating parameters in the forming process downstream of the resetting target process. Based on the difference between the predicted roundness and the allowable roundness for the product, the operating parameters of the forming process suitable for changing the roundness of the steel pipe after the expansion process can be selected. Alternatively, the operating parameters of both the resetting target process and any forming process downstream of the resetting target process can be reset. This is because when the difference between the predicted roundness and the allowable roundness for the product is large, the roundness of the steel pipe after the expansion process can be effectively changed.

[0170] Table 1 shows examples of forming processes selected as resetting targets and examples of forming processes whose operating parameters can be reset accordingly. Example 1 is an example of selecting the end bending process as the resetting target process in the manufacturing process of a steel pipe, which includes an end bending process. In this case, before the end bending process begins, the roundness of the steel pipe after the pipe expansion process is predicted using the set values ​​of the operating parameters in the forming processes, which include the pressure bending process and the weld gap reduction process. If the predicted roundness is large, any operating parameter in each forming process, including the end bending process, the pressure bending process, the weld gap reduction process, and the pipe expansion process, can be reset. The operating parameters to be reset are not limited to the operating parameters of the end bending process, but can also be the operating parameters of other forming processes. It should be noted that when the property information of the steel plate is included as input to the roundness prediction model M, actual data, including measured values ​​related to the property information of the steel plate, can be used as input before the end bending process, which is the resetting target process.

[0171] Examples 2 and 3 can also be approached using the same logic as Example 1 to select the target process for resetting and the resetting operation parameters. On the other hand, Example 4 involves setting the pipe expansion process as the target process for resetting. In this case, before the start of the pipe expansion process, a roundness prediction model M is used to predict the roundness of the steel pipe after the expansion process. In this case, as input to the roundness prediction model M, at least the operational performance data from the bending process and the weld gap reduction process can be used. Alternatively, performance data from the steel plate's property information and operational performance data from the end bending process can also be used. Thus, by comparing the predicted roundness of the steel pipe after the expansion process with the roundness allowed for the product, if a reduction in roundness is desired, the operation parameters in the pipe expansion process are reset. As the reset operation parameter for the pipe expansion process, the expansion rate is preferred. It should be noted that the change in the reset expansion rate relative to the initial setting value can be set based on empirical insights. However, if the roundness prediction model M includes the expansion rate of the expansion process in its input, the value of the re-set expansion rate can also be used as the input of the roundness prediction model M to re-predict the roundness of the steel pipe after the expansion process and determine whether the re-set conditions are appropriate.

[0172] [Table 1]

[0173] (Table 1)

[0174]

[0175] ○: Forming process that allows for resetting operating parameters

[0176] Here, refer to Figure 16The roundness control method for steel pipes, as one embodiment of the present invention, will be described. Figure 16 The example shown is as follows: The weld gap reduction process is selected as the resetting target process. After the bending process is completed, the U-shaped formed body is transferred for the weld gap reduction process. At this time, the operation performance data of the bending process is sent to the operation condition resetting unit 150. The operation performance data can be sent via a network from the control computer of each forming process. However, it can also be temporarily sent from the control computer of each forming process to the host computer 140 that oversees the steel pipe manufacturing process, and then sent from the host computer 140 to the operation condition resetting unit 150. Furthermore, the operation condition resetting unit 150 sends performance data regarding the steel plate's properties from the host computer 140 as needed. Moreover, the operation performance data of the end bending process can also be sent as needed. Additionally, the setting values ​​for the operation parameters of the resetting target process and the forming processes downstream of the resetting target process, namely the weld gap reduction process and the pipe expansion process, are sent from the control computer of each process to the operation condition resetting unit 150. However, if the set values ​​of the operating parameters for the weld gap reduction process and the pipe expansion process are stored in the host computer 140, they can also be sent from the host computer 140 to the operating condition resetting unit 150. It should be noted that the roundness target value determined according to the specifications of the steel pipe that will become the product is sent from the host computer 140 to the operating condition resetting unit 150.

[0177] The operating condition resetting unit 150 uses the roundness prediction model M online to predict the roundness of the steel pipe after the pipe expansion process based on this information, and compares the predicted roundness (predicted roundness value) with the target roundness (target roundness value). If the predicted roundness value is smaller than the target roundness value, the operating condition resetting unit 150 determines the operating conditions for the remaining forming processes without changing the set values ​​of the operating conditions for the bending process, weld gap reduction process, and pipe expansion process, and manufactures the steel pipe. On the other hand, if the predicted roundness is larger than the target roundness value, the operating condition resetting unit 150 resets at least the operating conditions for the weld gap reduction process or the pipe expansion process. Specifically, the O-punch reduction amount, etc., of the weld gap reduction process can be reset. Additionally, the pipe expansion rate of the pipe expansion process can be reset. Furthermore, both the O-punch reduction amount and the pipe expansion rate can be reset.

[0178] It should be noted that the operation condition resetting unit 150 can also reuse the reset operation parameters as input data for the roundness prediction model M to perform roundness prediction again, confirm whether the predicted roundness is smaller than the target roundness value, and determine the resetting values ​​for the operation conditions of the weld gap reduction process and the pipe expansion process. The reset operation conditions for the weld gap reduction process and the pipe expansion process are sent to their respective control computers, becoming the operation conditions for the weld gap reduction process and the pipe expansion process. By repeating the roundness determination in the operation condition resetting unit 150 multiple times, even if the target roundness value is set small, appropriate operation conditions for the weld gap reduction process and the pipe expansion process can be set, thus enabling the manufacture of steel pipes with better roundness. Moreover, after performing roundness control on the steel pipe after the pipe expansion process, in which the weld gap reduction process is set as the resetting target process, the roundness control on the steel pipe after the pipe expansion process, which has been formed into a slotted pipe and then welded, can be performed again. This is because: by obtaining operational data on reducing weld gaps, the accuracy of predicting the roundness of the steel pipe is further improved.

[0179] As described above, according to the steel pipe roundness control method of one embodiment of the present invention, since a roundness prediction model M that considers the influence of the interaction between the bending process and the weld gap reduction process on roundness is used, suitable operating conditions for ensuring good roundness of the steel pipe after the expansion process can be set, and steel pipes with high roundness can be manufactured. Furthermore, high-precision roundness control reflecting deviations in the property information of the steel plate that becomes the billet can be achieved.

[0180] <Steel Pipe Roundness Prediction Device>

[0181] Next, refer to Figure 17 The roundness prediction device for steel pipes, which is one embodiment of the present invention, will be described.

[0182] Figure 17 This is a diagram showing the structure of a steel pipe roundness prediction device as an embodiment of the present invention. Figure 17 As shown, the roundness prediction device 160 for steel pipes according to one embodiment of the present invention includes an operation parameter acquisition unit 161, a storage unit 162, a roundness prediction unit 163, and an output unit 164.

[0183] The operation parameter acquisition unit 161, for example, has an interface capable of acquiring the roundness prediction model M generated by the machine learning unit from the roundness prediction model generation unit 130. For example, the operation parameter acquisition unit 161 preferably has a communication interface for acquiring the roundness prediction model M from the roundness prediction model generation unit 130. In this case, the operation parameter acquisition unit 161 can also receive the roundness prediction model M from the machine learning unit 100b using a prescribed communication protocol. Furthermore, the operation parameter acquisition unit 161 acquires the operating conditions of the forming equipment (the equipment performing the forming process) from the control computer or host computer of the equipment used in each forming process, for example. For example, the operation parameter acquisition unit 161 preferably has a communication interface for acquiring the operating conditions. Additionally, the operation parameter acquisition unit 161 can acquire input information based on user operations. In this case, the roundness prediction device 160 for steel pipes also has an input unit that includes one or more input interfaces that detect user input and acquire input information based on user operations. Examples of input units include physical buttons, electrostatic capacitive buttons, touch screens integrated with the output unit's display, and microphones that accept voice input, but these are not limited to these. For example, the input unit accepts input relative to the operating conditions of the roundness prediction model M obtained by the operation parameter acquisition unit 161 from the roundness prediction model generation unit 130.

[0184] Storage unit 162 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of these. Storage unit 162 functions as, for example, a main storage device, an auxiliary storage device, or a flash memory. Storage unit 162 stores any information used in the operation of the roundness prediction device 160 for steel pipes. Storage unit 162 stores, for example, the roundness prediction model M obtained by the operation parameter acquisition unit 161 from the roundness prediction model generation unit 130, the operating conditions obtained by the operation parameter acquisition unit 161 from the host computer, and the roundness information predicted by the roundness prediction device 160 for steel pipes. Storage unit 162 may also store system programs and application programs, etc.

[0185] The roundness prediction unit 163 includes one or more processors. In this embodiment, the processor can be a general-purpose processor or a dedicated processor for a specific process, but is not limited to either. The roundness prediction unit 163 is communicatively connected to each structural component constituting the roundness prediction device 160 of the steel pipe, and controls the overall operation of the roundness prediction device 160 of the steel pipe. The roundness prediction unit 163 can be, for example, any general-purpose electronic device such as a PC (Personal Computer) or a smartphone. The roundness prediction unit 163 is not limited to these; it can also be one or multiple server devices capable of communicating with each other, or other electronic devices dedicated to the roundness prediction device 160 of the steel pipe. The roundness prediction unit 163 uses the operating conditions obtained through the operating parameter acquisition unit 161 and the roundness prediction model M obtained from the roundness prediction model generation unit 130 to calculate the predicted value of the roundness information of the steel pipe.

[0186] The output unit 164 outputs the predicted value of the roundness information of the steel pipe calculated by the roundness prediction unit 163 to a device for setting the operating conditions of the forming processing equipment. The output unit 164 may include one or more output interfaces for notifying the user of output information. The output interface is, for example, a display. The display is, for example, an LCD or an organic EL display. The output unit 164 outputs data obtained by the operation of the roundness prediction device 160 of the steel pipe. The output unit 164 can replace the roundness prediction device 160 installed on the steel pipe and be connected to the roundness prediction device 160 of the steel pipe as an external output device. As a connection method, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used. For example, as the output unit 164, examples include a display that outputs information in the form of images, a speaker that outputs information in the form of voice, etc., but it is not limited to them. For example, the output unit 164 prompts the user with the predicted value of the roundness information calculated by the roundness prediction unit 163. Based on the predicted roundness value provided by the output unit 164, the user can appropriately set the operating conditions of the forming processing equipment.

[0187] A more preferred form of the roundness prediction device 160 for steel pipes after the expansion process described above is a terminal device such as a tablet terminal, which has an input unit 165 that acquires input information based on user operation and a display unit 166 that displays the predicted value of the roundness information calculated by the roundness prediction unit 163. It acquires user operation-based input information from the input unit 165 and uses this information to update part or all of the operating parameters of the forming equipment that have been input into the roundness prediction device 160 for steel pipes. That is, for steel plates being processed in the forming equipment, if the roundness information of the steel pipe has been predicted by the roundness prediction unit 163, the operator uses the terminal device to receive a correction input of a portion of the operating parameters of the forming equipment that have been input into the operating parameter acquisition unit 161. At this time, the operating parameter acquisition unit 161 maintains the original input data for operating parameters that are not corrected from the terminal device in the operating parameters of the forming equipment, and only changes the operating parameters that have been corrected. Therefore, in the operation parameter acquisition unit 161, new input data for the roundness prediction model M is generated, and the roundness prediction unit 163 calculates the predicted value of the roundness information based on the input data. Furthermore, the calculated predicted value of the roundness information is displayed on the display unit 166 of the terminal device via the output unit 164. Thus, the operator or factory manager of the forming equipment can immediately confirm the predicted value of the roundness information when the operating parameters of the forming equipment are changed, and quickly adjust to suitable operating conditions.

[0188] Example

[0189] [Example 1]

[0190] In this embodiment, an offline roundness prediction model after the pipe expansion process was generated, corresponding to the manufacturing conditions of using pipeline steel plates (API grade X60) with a thickness of 38.0–38.4 mm and a width of 2700–2720 mm to manufacture a 36-inch diameter steel pipe after the pipe expansion process, which involves bending, weld gap reduction, welding, and pipe expansion. An example of the finite element model generated by the finite element model generation unit for the weld gap reduction process used in this embodiment is shown below. Figure 13 The example is shown below. The finite element analysis solver used is Abaqus 2019, and the computation time per case is approximately 3 hours. The dataset stored in the database contains 300 data points, and Gaussian process regression with dynamic path basis functions was used as the machine learning model.

[0191] In addition, for the steel plate's property information, representative plate thickness (average thickness within the plane), plate width, and yield stress were selected. The range of variations for these operating conditions was determined based on manufacturing performance, and the input data for calculations was modified accordingly. For the bending process's operating parameters, the number of pressing strokes and the pressing position were selected. At this point, the number of pressing strokes was based on 11 strokes, and the conditions were varied to a range of 7 to 15 strokes. Regarding the pressing position, pressing strokes were performed at equal intervals along the plate width direction based on the number of pressing strokes, and the pressing position was determined based on the number of pressing strokes. The pressing amount was set to the amount by which the tip of the punch reached a position 15.8 mm from the top of the connecting rod-shaped member, with each stroke resulting in a 30° bend.

[0192] Then, a steel plate is placed on a die with the spacing of the rod-shaped members set at 450 mm. Using a punch with a machined surface having a radius of 308 mm, the stamping process begins with a position 1120 mm away from the center of the steel plate in the width direction. With 11 stamping passes, from... Figure 2 The paper is pressed down five times on the right side towards the center of the width direction, with a sheet feed spacing of 224mm. Afterwards, ... Figure 2 The paper moves from the left end to the vicinity of the rod-shaped member, and from a position 1120mm from the end about the left half of the steel plate, it is pressed down 6 times with a plate feed spacing of 224mm.

[0193] For the weld gap reduction process, an O-punch device was used. The operating parameters for this process were selected based on the O-punch reduction rate, with operating conditions varying from 1.0% to 3.0%. Other operating conditions were set to use an upper die with an arc surface having a radius R of 457.2 mm and a central angle θc of 60°, and a flat surface connected to the arc surface at angle θd of 30°, and a lower die with a concave arc surface having a radius R of 502.9 mm. On the other hand, a constant value of 1.0% was used for the expansion rate, which was a parameter for the pipe expansion process.

[0194] In this embodiment, the above-described analytical conditions are set for the offline roundness calculation unit. Within the range of the aforementioned operating conditions, the analytical conditions are changed, and the roundness calculation results obtained through analysis after the pipe expansion process are stored in a database. Furthermore, a roundness prediction model is generated based on the stored database. In this embodiment, the generated roundness prediction model is applied online. Regarding the roundness in this embodiment, the roundness is defined as Dmax - Dmin, where the outer diameter at each of the pipe's circumferentially divided into 3600 equal parts and opposite positions is selected, with the maximum and minimum diameters among them set as Dmax and Dmin, respectively.

[0195] In the online process, before the bending process begins, the representative thickness and width of the steel plate are obtained from the host computer as actual data on the properties of the steel plate that will become the billet. Additionally, test data on the yield stress obtained during the inspection process of the thick plate rolling process are acquired. Furthermore, the setting values ​​for the operating conditions of the bending process and the weld gap reduction process are obtained from the host computer. In the steel pipe manufacturing process, which is the subject of this embodiment, the operating conditions preset by the host computer are as follows: the number of punches in the bending process is 11; the first pressing position is set at a position 1120 mm away from the center of the steel plate in the width direction; and the punching positions are set at 224 mm intervals in the width direction of the steel plate. Furthermore, the punching reduction amount at each punching position is preset to a value of 15.8 mm. On the other hand, regarding the weld gap reduction process using an O-ring punch, the O-ring punching reduction rate is set to 2.0% as a preset operating condition value by the host computer.

[0196] In this embodiment, before the bending process begins, these set values ​​and representative plate thickness and width, which are actual data representing the properties of the steel plate, are used as inputs to the roundness prediction model to predict the roundness of the steel pipe after the expansion process. On the other hand, in the host computer, the roundness target value is set to 10 mm. The predicted roundness of the steel pipe (predicted roundness value) is compared with the roundness target value. If the predicted roundness exceeds the roundness target value, the operating conditions for the bending process are then set. The number of stamping operations is selected as the re-set operating condition. As a result, in this example, the average roundness is confirmed to be 4.0 mm, and the pass rate is 100%. In contrast, as a comparative example, when the bending process operating conditions are maintained at the preset values ​​by the host computer, the average roundness is 11.2 mm, and the pass rate is 80%.

[0197] [Example 2]

[0198] In this embodiment, a roundness prediction model after the pipe expansion process is generated offline for a case where a 56-inch diameter steel pipe is manufactured using pipeline steel plates (API grade X60) with a thickness of 50.0–50.4 mm and a width of 4450–4460 mm. This is achieved by bending, reducing weld gap, welding, and expanding the pipe. In this case, the parameters for the steel plate's property information, the bending process's operation parameters, and the weld gap reduction process's operation parameters, which are set as inputs to the roundness prediction model of the steel pipe, are the same as in Example 1. However, the ranges of these operation parameters differ from those in Example 1. For the representative plate thickness, which serves as the parameter for the steel plate's property information, the operation condition dataset is set in the range of 50.0–50.4 mm; for the plate width, the operation condition dataset is set in the range of 4450–4460 mm.

[0199] As operating conditions for the bending process, the die spacing of the bending equipment was set to 620mm. A punch with a front-end radius of curvature of 478mm was used, and the starting point for the pressing was set at a position 1824mm away from the center of the steel plate in the width direction. The number of pressing operations in the bending process was set to a range of 7 to 15 times. In this case, the number of pressing operations preset by the host computer was 11. Under these conditions, the following operating conditions were set: starting from the pressing start point on one side of the steel plate and moving towards the center of the steel plate in the width direction, 5 pressing operations were performed with a feed interval of 365mm; then, starting from the pressing start point on the other side and moving towards the center of the steel plate in the width direction, 6 pressing operations were performed with a feed interval of 365mm. Furthermore, the pressing depth was set to 33.8mm for all pressing operations. With different number of pressing cycles, only the feed interval is changed, while the starting position and pressing amount are set to constant values.

[0200] For the weld gap reduction process, an O-punch device was used. The upper die of the O-punch device had an arc surface with a radius R of 704.0 mm and a central angle θc of 60°, and a flat surface connected to the arc surface at an angle θd of 30°. The lower die had a concave arc surface with a radius R of 704.0 mm. The operating parameter for the weld gap reduction process was the O-punch reduction rate, and the operating condition dataset was set in the range of 1.0% to 3.0%. It should be noted that the O-punch reduction rate preset by the host computer was 2.0%. Additionally, the expansion rate, which became an operating parameter in the pipe expansion process, was set to 0.9%.

[0201] The manufacturing conditions in the forming process of the steel pipe described above were set. Multiple operational condition datasets were prepared, including parameters representing steel plate properties such as plate thickness and width, operational parameters for the bending process such as the number of punching reductions, and operational parameters for the weld gap reduction process such as the O-punch reduction rate. Finite element analysis was performed in the basic data acquisition section to accumulate the roundness information of the steel pipe after the expansion process in a database. The number of datasets accumulated in the database was 300. Using Gaussian process regression with dynamic diameter basis functions as the basis functions, a roundness prediction model was generated offline, with the roundness of the steel pipe after the expansion process as the output.

[0202] The roundness prediction model generated as described above is sent to the online operation condition resetting unit, configured to accept the operation condition dataset obtained from the host computer as input and output the predicted roundness value of the steel pipe after the expansion process. Furthermore, in this embodiment, the bending process is selected as the resetting target process. Before the bending process begins, the representative plate thickness and width (as actual data of the steel plate's attribute information), along with the operation setting values ​​for the bending process and the weld gap reduction process, are obtained from the host computer to form an operation condition dataset. The roundness prediction model is used to predict the roundness of the steel pipe after the expansion process. Here, in the host computer, the roundness target value for the target steel pipe is set to 14.2 mm, and the roundness is determined by comparing it with the predicted roundness value. If the predicted roundness value exceeds the target roundness value, the operation conditions for the forming process downstream of the resetting target process, namely the weld gap reduction process, are reset. The reset operation condition is set to the O-punch reduction rate.

[0203] One hundred steel pipes were manufactured using this roundness control method. As a result, the average roundness of the steel pipes after the expansion process was 10.0 mm, compared to the target roundness value of 14.2 mm, resulting in a pass rate of 90%. On the other hand, when steel pipes were manufactured without using the roundness control method of this embodiment and without changing the operating conditions pre-set by the host computer, the average roundness was 14.4 mm, and the pass rate was 60%.

[0204] [Example 3]

[0205] In this embodiment, a roundness prediction model after the pipe expansion process is generated offline for a case where a 36-inch diameter steel pipe is manufactured using pipeline steel plates (API grade X60) with a thickness of 38.0–38.4 mm and a width of 2700–2720 mm. This is achieved by performing end bending, pressure bending, weld gap reduction, welding, and pipe expansion processes. In this case, the parameters for the steel plate properties, the operation parameters for the pressure bending process, and the operation parameters for the weld gap reduction process, which are set as inputs to the roundness prediction model of the steel pipe, are the same as in Example 1. Furthermore, the ranges of the operation parameters constituting the operation condition dataset are also the same as in Example 1.

[0206] On the other hand, regarding the end bending process, instead of changing the upper and lower pairs of dies used in the C-type stamping device for each steel plate, the same die is used. The end bending width is selected as the operating parameter for the end bending process, and the operating condition dataset in the basic data acquisition unit is changed to vary from 180 to 240 mm. Furthermore, in this embodiment, the tube expansion rate is selected as the operating parameter for the tube expansion process, and the operating condition dataset in the basic data acquisition unit is changed to vary from 0.6% to 1.4%.

[0207] Under the above conditions, finite element analysis was performed on the operating condition dataset defined within the aforementioned range using Abaqus 2019 as the finite element analysis solver. The roundness of the pipe after the expansion process obtained through analysis was mapped to the operating condition dataset and stored in a database. The number of datasets stored in the database is 600. For the machine learning model, Gaussian process regression with dynamic diameter basis functions was used to generate an offline roundness prediction model with the roundness of the steel pipe after the expansion process as the output.

[0208] The roundness prediction model generated as described above is incorporated into a roundness prediction device that performs online roundness prediction of steel pipes. The roundness prediction device takes the operating parameters from the steel pipe manufacturing process, obtained online from a host computer, as input and outputs a predicted value for the roundness of the steel pipe after the pipe expansion process. In this embodiment, the roundness prediction device is a tablet terminal, capable of obtaining input information based on the operator's actions from the input unit, and using this input information to update part or all of the operating parameters of the forming processing equipment input to the roundness prediction device. This tablet terminal has the function of recognizing operating parameters that have been corrected by the operator's actions and displaying a predicted value of roundness information reflecting the corrected input value on the display unit.

[0209] In this embodiment, during the online manufacturing process, firstly, before starting the end bending process, the operation parameter acquisition unit acquires the actual data of the steel plate's property information, the pre-set operation settings for the bending process, weld gap reduction process, and tube expansion process from the host computer. Then, the roundness prediction unit outputs the predicted roundness value of the steel pipe after the tube expansion process, obtained by inputting the acquired operation condition dataset, to the display unit of the aforementioned tablet terminal. The operation manager, who oversees the steel pipe manufacturing process, confirms the displayed predicted roundness value and compares it with the set value (in this case, 7.0 mm) for the target roundness value of the steel pipe. Furthermore, if the predicted roundness is high, the operation conditions for the end bending processing width, which are operation parameters for the end bending process, can be corrected by inputting from the operation panel of the C-type stamping device.

[0210] In this embodiment, the operator corrects the end-bending processing width, the operating parameter for the end-bending process displayed on the tablet terminal, to a range of 180-240 mm. The roundness prediction device in this embodiment updates the corrected end-bending processing width to the corrected value, using it as an input parameter to the roundness prediction model. Then, while maintaining the values ​​already obtained by the operating parameter acquisition unit for other inputs, the roundness prediction device re-displays the roundness prediction value on the tablet terminal. The operator confirms the displayed roundness prediction value, determines the condition for the end-bending processing width as an operating parameter for the end-bending process, and sets a new setting value for the operation dial of the C-type stamping device.

[0211] Thus, using the roundness prediction device of this embodiment, the operator appropriately modified the operating conditions in the forming process of the steel pipe to manufacture 50 steel pipes. As a result, relative to the roundness target value of 7.0 mm, the average roundness of the steel pipes after the pipe expansion process was 4.2 mm, and the pass rate was 100%. On the other hand, when the steel pipes were manufactured without using the roundness prediction device of this embodiment and without changing the operating conditions pre-set by the host computer, the average roundness was 9.1 mm, and the pass rate was 35%. That is, it is confirmed that the roundness prediction device of this embodiment is effective in assisting the operator's judgment in the steel pipe manufacturing process.

[0212] Industrial availability

[0213] According to the present invention, a method for generating a roundness prediction model for a steel pipe is provided, which can generate a roundness prediction model that can generate a high-precision and rapid prediction of the roundness of the steel pipe after the expansion process in the manufacturing process of a steel pipe consisting of multiple processes. Furthermore, according to the present invention, a method and apparatus for predicting the roundness of a steel pipe after the expansion process in the manufacturing process of a steel pipe consisting of multiple processes are provided, which can predict the roundness of the steel pipe after the expansion process in the manufacturing process of a steel pipe consisting of multiple processes with high precision. Additionally, according to the present invention, a method for manufacturing a steel pipe that can produce steel pipes with a desired roundness with high yield is provided.

[0214] Explanation of reference numerals in the attached figures

[0215] 1. Stamping Die

[0216] 1a, 1b Rod-shaped components

[0217] 2 Punch

[0218] 2a Punch front end

[0219] 2b Punch support body

[0220] 3 upper mold

[0221] 4. Lower mold

[0222] 10a, 10b Lower Tools

[0223] 11a, 11b Spring Units

[0224] 12 punches

[0225] 13. Upper tool

[0226] 16. Pipe Expanding Mold

[0227] 17. Conical outer circumference

[0228] 18 pull rods

[0229] 20 arms

[0230] 21a, 21b Displacement gauges

[0231] 22 Rotation Angle Detector

[0232] 25 Rotating Arm

[0233] 26a, 26b Pressing Rollers

[0234] 30 C-type stamping device

[0235] 31. Transportation agencies

[0236] 31a Conveyor Roller

[0237] 32A and 32B stamping mechanisms

[0238] 33 upper mold

[0239] 33a Forming surface

[0240] 34 Lower mold

[0241] 34a Pressing surface

[0242] 36 Hydraulic cylinders

[0243] 37 Clamping Mechanism

[0244] 110 Basic Data Acquisition Department

[0245] 111 Operational Conditions Dataset

[0246] 112 Offline Roundness Calculation Department

[0247] 112a Finite element model generation unit for bending process

[0248] 112b Finite element model generation section for weld gap reduction process

[0249] 112c Finite element model generation section for pipe expansion process

[0250] 112d Finite Element Analysis Solver

[0251] 120 Database

[0252] 130 Roundness Prediction Model Generation Unit

[0253] 140 host computers

[0254] 150 Operating Condition Resetting Section

[0255] 160 steel pipe roundness prediction device

[0256] 161 Operation Parameter Acquisition Unit

[0257] 162 Storage Department

[0258] 163 Roundness Prediction Department

[0259] 164 Output Section

[0260] 165 Input Section

[0261] 166 Display Department

[0262] G Weld gap

[0263] M-Roundness Prediction Model

[0264] P steel pipe

[0265] R1 and R2 areas

[0266] S steel plate

[0267] S1 Molded Body

[0268] S2 slotted pipe.

Claims

1. A method for generating a roundness prediction model for steel pipes, comprising generating a roundness prediction model for predicting the roundness of a steel pipe after the pipe expansion process in a steel pipe manufacturing process including a bending process, a weld gap reduction process, a welding process, and a pipe expansion process, wherein the bending process is a process of processing a steel plate into a U-shaped cross-section by multiple pressing with a punch, the weld gap reduction process is a process of reducing the weld gap of the U-shaped cross-section to form a slotted pipe, the welding process is a process of joining the ends of the slotted pipe together, and the pipe expansion process is a process of expanding the inner diameter of the steel pipe after the ends are joined together, wherein... The generation method includes: The basic data acquisition step involves calculating the roundness of the steel pipe after the expansion process as the output data, including an operation condition dataset in the input data, and performing the calculation while changing the operation condition dataset multiple times. This generates multiple sets of data corresponding to the operation condition dataset, representing the roundness of the steel pipe after the expansion process, offline as learning data. The operation condition dataset includes one or more parameters selected from the steel plate's attribute information, one or more parameters selected from the operation parameters of the bending process, and one or more parameters selected from the operation parameters of the weld gap reduction process. The roundness prediction model generation step uses multiple learning datasets generated in the basic data acquisition step to generate a roundness prediction model offline through machine learning, with the operating condition dataset as input data and the roundness of the steel pipe after the pipe expansion process as output data.

2. The method for generating the roundness prediction model of steel pipes according to claim 1, The basic data acquisition step includes using the finite element method to calculate the roundness of the steel pipe after the expansion process based on the operating condition dataset.

3. The method for generating the roundness prediction model of steel pipes according to claim 1 or 2, The roundness prediction model includes one or more parameters selected from the operating parameters of the tube expansion process, which are part of the operating condition dataset.

4. The method for generating the roundness prediction model of steel pipes according to claim 1 or 2, The manufacturing process of the steel pipe includes an end bending process that bends the end of the steel plate in the width direction prior to the bending process. For the roundness prediction model, the operating condition dataset includes one or more parameters selected from the operating parameters of the end bending process.

5. The method for generating the roundness prediction model of steel pipes according to claim 1 or 2, The operating parameters of the bending process include the number of punches performed in the bending process, the punching position information of the punch used in the bending process, and the punching reduction amount.

6. The method for generating the roundness prediction model for steel pipes according to claim 1 or 2, As the machine learning method, machine learning methods selected from neural networks, decision tree learning, random forests, Gaussian process regression, and support vector machine regression are used.

7. A method for predicting the roundness of a steel pipe, comprising: The operation parameter acquisition step involves obtaining an operation condition dataset online, which is set as the operation conditions for the manufacturing process of the steel pipe, as input to the steel pipe roundness prediction model generated by the method for generating the steel pipe roundness prediction model according to any one of claims 1 to 6. and The roundness prediction step involves inputting the operational condition dataset obtained in the operational parameter acquisition step into the roundness prediction model to predict the roundness information of the steel pipe after the pipe expansion process.

8. A method for controlling the roundness of a steel pipe, comprising the following steps: Using the roundness prediction method for steel pipes according to claim 7, before the start of the bending process, an operational condition dataset is obtained, including actual values ​​of the steel plate's attribute information, set values ​​of the operating parameters for the bending process, and set values ​​of the operating parameters for the weld gap reduction process. The obtained operational condition dataset is input into the roundness prediction model to predict the roundness of the steel pipe after the expansion process. At least one of the set values ​​of the operating parameters for the bending process and the set values ​​of the operating parameters for the weld gap reduction process is re-set in a way that reduces the predicted roundness.

9. A method for controlling the roundness of a steel pipe, comprising the following steps: Using the roundness prediction method for steel pipes as described in claim 7, before the start of a resetting target process selected from the end bending process, pressure bending process, weld gap reduction process, and pipe expansion process constituting the steel pipe manufacturing process, the roundness information of the steel pipe after the pipe expansion process is predicted. Based on the predicted roundness information of the steel pipe, at least one or more operating parameters selected from the operating parameters of the resetting target process or one or more operating parameters selected from the operating parameters of the forming process downstream of the resetting target process are reset.

10. A method for manufacturing a steel pipe, The steps include manufacturing the steel pipe using the roundness control method for the steel pipe as described in claim 8 or 9.

11. A steel pipe roundness prediction device, for predicting the roundness of a steel pipe after the pipe expansion process in a steel pipe manufacturing process including a bending process, a weld gap reduction process, a welding process, and a pipe expansion process, wherein the bending process is a process of processing a steel plate into a U-shaped cross-section by multiple pressing with a punch, the weld gap reduction process is a process of reducing the weld gap of the U-shaped cross-section to form a slotted pipe, the welding process is a process of joining the ends of the slotted pipe together, and the pipe expansion process is a process of expanding the inner diameter of the steel pipe after the ends are joined together, wherein... The roundness prediction device includes: The basic data acquisition unit takes the operating condition dataset as input data and sets the roundness information of the steel pipe after the pipe expansion process as output data. It performs numerical calculations while changing the operating condition dataset multiple times. As a result, multiple sets of data corresponding to the operating condition dataset of the roundness information of the steel pipe after the pipe expansion process are generated as learning data. The operating condition dataset includes one or more parameters selected from the attribute information of the steel plate, one or more operating parameters selected from the operating parameters of the bending process, and one or more operating parameters selected from the operating parameters of the weld gap reduction process. The roundness prediction model generation unit uses multiple learning data generated in the basic data acquisition unit to generate a roundness prediction model by machine learning, with the operating condition dataset as input data and the roundness information of the steel pipe after the pipe expansion process as output data. The operation parameter acquisition unit acquires online the operation condition dataset set as the operation conditions for the manufacturing process of the steel pipe; and The roundness prediction unit uses the roundness prediction model generated in the roundness prediction model generation unit to predict online the roundness information of the steel pipe after the expansion process, which corresponds to the operating condition dataset obtained by the operating parameter acquisition unit.

12. The roundness prediction device for steel pipes according to claim 11, The device includes a terminal unit that receives input information based on user operations and a display unit that displays the roundness information. The operation parameter acquisition unit updates part or all of the operation condition dataset in the steel pipe manufacturing process based on the input information obtained by the input unit. The display unit shows the roundness information of the steel pipe predicted by the roundness prediction unit using the updated operating condition dataset.

Images