STEEL PIPE OUT-OF-ROUNDNESS PREDICTION MODEL GENERATION METHOD, STEEL PIPE OUT-OF-ROUNDNESS PREDICTION METHOD, STEEL PIPE OUT-OF-ROUNDNESS CONTROL METHOD, STEEL PIPE MANUFACTURING METHOD, AND STEEL PIPE OUT-OF-ROUNDNESS PREDICTION DEVICE

FIELD

The present invention relates to a steel pipe out-of-roundness prediction model generation method, which is a method of generating an out-of-roundness prediction model that predicts an out-of-roundness of a steel pipe after a pipe expanding step in a steel pipe manufacturing process using a press bending method, a steel pipe out-of-roundness prediction method, a steel pipe out-of-roundness control method, a steel pipe manufacturing method, and a steel pipe out-of-roundness prediction device.

BACKGROUND

Techniques of manufacturing a steel pipe having a large diameter and a large thickness used for a line pipe and the like include a wide-spread technique of manufacturing a steel pipe (referred to as UOE steel pipe) obtained by press working on a steel sheet having a predetermined length, width, and thickness into a U shape, press forming the sheet into an O shape and welding a butt portion to form the sheet into a steel pipe, and expanding the diameter of the steel pipe (referred to as pipe expansion) to improve out-of-roundness. However, the manufacturing process of the UOE steel pipe requires a large pressing pressure in the step of performing press working on the steel sheet to be formed into a U shape and an O shape, making it necessary to use a large-scale press machine.

To manage this issue, there is a proposed technique in manufacturing a steel pipe having a large diameter and a large thickness, which is a technique of press forming with reduced pressing pressure. Specifically, the technique that has been put into practical use is a technique in which, after bending is applied to the end of a steel sheet in the width direction (referred to as end bending), a press bending step of performing a 3-point bending press a plurality of times by punching is conducted to produce a formed body having a U-shaped cross section (hereinafter, referred to as a U-shaped formed body), a seam gap reduction step of reducing a seam gap portion of the formed body having a U-shaped cross section is conducted to form the pipe into an open pipe, butt portions are welded to form a steel pipe, and finally a pipe expanding device is inserted into the steel pipe to expand an inner diameter of the steel pipe. The pipe expanding device used is a device which includes a plurality of pipe expanding tools each having a curved surface obtained by dividing an arc into a plurality of portions and brings the curved surface of the pipe expanding tool into contact with the inner surface of the steel pipe so as to expand the steel pipe and fix the shape of the steel pipe.

In the press bending step, increasing the number of times of the 3-point bending press improves the out-of-roundness of the steel pipe after the pipe expanding step, although it takes a long time to form the steel pipe to have a U-shaped cross section. On the other hand, reducing the number of times of the 3-point bending press would result in a problem that the cross-section of the steel pipe takes a substantially polygonal shape, making it difficult to form a circular cross-sectional shape. To handle this, the number of times of 3-point bending press (for example, 5 to 13 times in a steel pipe having a diameter of 1200 mm) is empirically determined in operation depending on the dimensions of the steel pipe. Regarding the operational conditions of the press bending step for improving the out-of-roundness of the steel pipe after the pipe expanding step, many proposals have been made regarding the setting method thereof.

For example, Patent Literature 1 describes a method for performing 3-point bending press as few times as possible and a method of expanding a pipe by bringing a plurality of pipe expanding tools disposed in a circumferential direction of a pipe expanding device into contact with a non-deformed portion, which is not deformed by the 3-point bending press.

Patent Literature 2 describes a method of improving the out-of-roundness of the steel pipe after the pipe expanding step by setting the curvature radius of the outer circumferential surface of the punch used in the 3-point bending press and the curvature radius of the outer circumferential surface of the pipe expanding tool satisfy a predetermined relational expression.

Patent Literature 3 describes a method, as a manufacturing method capable of efficiently manufacturing a steel pipe having satisfactory out-of-roundness without requiring an excessive pressing force in the press bending step, in which there is provided, in execution of 3-point bending press, a light-worked portion having an extremely small curvature compared to other regions, or an unworked portion in which bending processing is omitted, in at least a part of a steel sheet. Patent Literature 3 also describes an operation, in the seam gap reduction step, which is application of a pressing force to a portion separated by a predetermined distance from the center of the light-worked portion or the unworked portion without restraining the light-worked portion or the unworked portion. Incidentally, an O-press device is typically used in the seam gap reduction step performed after the press bending step.

On the other hand, Non Patent Literature 1 describes a method of analyzing, with calculation using a finite element method, an influence of operational conditions of the pipe expanding step on the out-of-roundness of the steel pipe after the pipe expanding step.

CITATION LIST
Patent Literature

- Patent Literature 1: Japanese Laid-open Patent Publication No. 2012-170977
- Patent Literature 2: Japanese Patent No. 5541432
- Patent Literature 3: Japanese Patent No. 6015997

Non Patent Literature

- Non Patent Literature 1: Journal of the Japan Society for Technology of Plasticity vol. 59 No. 694 (2018), p. 203-208

SUMMARY
Technical Problem

The method described in Patent Literature 1 is a method of improving the out-of-roundness of the steel pipe after the pipe expanding step by associating the pressing position of the 3-point bending press with the pressing position of the pipe expanding tool. However, the steel pipe manufacturing process includes a plurality of steps including an end bending step, a press bending step, a seam gap reduction step, a welding step, and a pipe expanding step. Therefore, the method described in Patent Literature 1 has no consideration of the influence of the operational conditions of other steps on the out-of-roundness of the steel pipe after the pipe expanding step, making it difficult to always ensure improvement of the out-of-roundness of the steel pipe after the pipe expanding step.

Similarly to the method described in Patent Literature 1, the method described in Patent Literature 2 is a method in which the curvature radius is set such that the curvature radius of the outer circumferential surface of the punch used in the 3-point bending press as the operational condition of the press bending step and the curvature radius of the outer circumferential surface of the pipe expanding tool as the operational condition of the pipe expanding step satisfy a predetermined relational expression, thereby improving the out-of-roundness of the steel pipe after the pipe expanding step. However, in the method described in Patent Literature 2, similarly to the method described in Patent Literature 1, there is a problem of incapability of considering the influence of steps other than the press bending step, such as the seam gap reduction step.

The method described in Patent Literature 3 is a method of improving the out-of-roundness of the steel pipe after the pipe expanding step by changing the working condition of the 3-point bending press in the press bending step depending on the position of the steel sheet and setting the working condition to a condition associated with the forming condition in the seam gap reduction step. However, the method described in Patent Literature 3 has a problem that presence of variations in the sheet thickness or the material of the steel sheet as a material leads to variations in the out-of-roundness of the steel pipe after the pipe expanding step varies even under the same forming condition.

On the other hand, since the manufacturing process of the steel pipe includes a plurality of steps as described above, there is a problem of occurrence of a long lead time until completion of manufacturing of the steel sheet and an increase in the manufacturing cost. To handle this, there is a suggestion to improve the efficiency of the steel pipe manufacturing process by omitting some steps. Specifically, the seam gap reduction step can be omitted to provide a steel pipe manufacturing process that only includes the end bending step, the press bending step, the welding step, and the pipe expanding step. However, omitting the seam gap reduction step is assumed to deteriorate the out-of-roundness of the steel pipe after the pipe expanding step. In such a case, it is necessary to improve the out-of-roundness of the steel pipe after the pipe expanding step by appropriately combining the operational conditions of the plurality of steps.

On the other hand, as described in the method in Non Patent Literature 1, the influence of the operational parameter of the pipe expanding step on the out-of-roundness can be quantitatively predicted by analyzing the pipe expanding step using the finite element method as offline calculation. However, the method described in Non Patent Literature 1 has also a problem of incapability of considering the influence of the operational conditions of other steps on the out-of-roundness. Furthermore, such numerical analysis execution also includes a problem of the long time required for the calculation, making it difficult to predict the out-of-roundness online.

The present invention has been made to solve the above problems, and its one object is to provide a steel pipe out-of-roundness prediction model generation method capable of generating an out-of-roundness prediction model that accurately and promptly predicts an out-of-roundness of a steel pipe after a pipe expanding step in a steel pipe manufacturing process including a plurality of steps. Another object of the present invention is to provide a steel pipe out-of-roundness prediction method and a steel pipe out-of-roundness prediction device capable of accurately and promptly predicting out-of-roundness of the steel pipe after the pipe expanding step in the steel pipe manufacturing process including a plurality of steps. Another object of the present invention is to provide a steel pipe out-of-roundness control method capable of accurately controlling the out-of-roundness of the steel pipe after the pipe expanding step in the steel pipe manufacturing process including a plurality of steps. Another object of the present invention is to provide a steel pipe manufacturing method capable of manufacturing a steel pipe having a desired out-of-roundness with good yield.

Solution to Problem

To solve the problem and achieve the object, a steel pipe out-of-roundness prediction model generation method according to the present invention is a method of generating an out-of-roundness prediction model that predicts out-of-roundness of a steel pipe after a pipe expanding step in a steel pipe manufacturing process, the steel pipe manufacturing process including: an end bending step of applying end bending processing on an end of a steel sheet in a width direction; a press bending step of performing forming processing on a steel sheet that has undergone the end bending processing to form the steel sheet into an open pipe by a plurality of times of pressing using a punch; and the pipe expanding step being a step of performing forming processing by pipe expansion on the steel pipe obtained by joining ends of the open pipe. The method includes: a basic data acquisition step of executing a numerical computation in which an input data is an operational condition dataset including one operational parameter or two or more operational parameters selected from operational parameters of the end bending step and one operational parameter or two or more operational parameters selected from operational parameters of the press bending step, and an output data is a steel pipe out-of-roundness after the pipe expanding step, the execution of the numerical computation conducted a plurality of times while changing the operational condition dataset, and generating, by this numerical computation, a plurality of pairs of data of the operational condition data set and the steel pipe out-of-roundness data after the pipe expanding step corresponding to the operational condition dataset, offline as training data; and an out-of-roundness prediction model generation step of generating an out-of-roundness prediction model for which an input data is the operational condition dataset, and an output data is the out-of-roundness of the steel pipe after the pipe expanding step, the generation of the out-of-roundness prediction model performed offline by machine learning using the plurality of pairs of training data generated in the basic data acquisition step.

Moreover, the basic data acquisition step may include a step of calculating, by using a finite element method, the out-of-roundness of the steel pipe after the pipe expanding step from the operational condition dataset.

Moreover, the out-of-roundness prediction model may include, as the input data, one parameter or two or more parameters selected from attribute information of the steel sheet.

Moreover, the out-of-roundness prediction model may include a pipe expansion rate selected from the operational parameters of the pipe expanding step, as the input data.

Moreover, the operational parameters of the end bending step may include one parameter or two or more parameters of an end bending processing width, a C-press force, and a clamp gripping force.

Moreover, the operational parameter of the press bending step may include: press position information and a press depression amount regarding an operation in which a punch used in the press bending step presses a steel sheet; and the number of times of pressing performed through the press bending step.

Moreover, the machine learning to be used may be a type of machine learning selected from a neural network, decision tree learning, random forest, Gaussian process regression, and support vector regression.

Moreover, a steel pipe out-of-roundness prediction method according to the present invention includes: an operational parameter acquisition step of acquiring online an operational condition dataset to be set as an operational condition of the steel pipe manufacturing process, as an input of the steel pipe out-of-roundness prediction model generated by the steel pipe out-of-roundness prediction model generation method according to the present invention; and an out-of-roundness prediction step of predicting out-of-roundness information of a steel pipe after a pipe expanding step by inputting the operational condition dataset acquired in the operational parameter acquisition step to the out-of-roundness prediction model.

Moreover, a steel pipe out-of-roundness control method according to the present invention includes a reconfiguring step of predicting steel pipe out-of-roundness information after the pipe expanding step using the steel pipe out-of-roundness prediction method according to the present invention, the prediction being performed before starting a reconfiguration target step which is selected from the end bending step, the press bending step, and the pipe expanding step included in the steel pipe manufacturing process, and reconfiguring one operational parameter or two or more operational parameters selected from at least operational parameters of the reconfiguration target step or one operational parameter or two or more operational parameters selected from operational parameters of a forming processing step on a downstream side of the reconfiguration target step, the reconfiguration performed based on the predicted steel pipe out-of-roundness information.

Moreover, a steel pipe manufacturing method according to the present invention includes a step of manufacturing a steel pipe using the steel pipe out-of-roundness control method according to the present invention.

Moreover, a steel pipe out-of-roundness prediction device according to the present invention is a device of predicting out-of-roundness of a steel pipe after a pipe expanding step in a steel pipe manufacturing process, the steel pipe manufacturing process including: an end bending step of applying end bending processing on an end of a steel sheet in a width direction; a press bending step of performing forming processing on a steel sheet that has undergone the end bending processing to form the steel sheet into an open pipe by a plurality of times of pressing using a punch; and the pipe expanding step being a step of performing forming processing by pipe expansion on the steel pipe obtained by joining ends of the open pipe. The steel pipe out-of-roundness prediction device includes: a basic data acquisition unit that executes a numerical computation in which an input data is an operational condition dataset including one operational parameter or two or more operational parameters selected from operational parameters of the end bending step and one operational parameter or two or more operational parameters selected from operational parameters of the press bending step, and an output data is a steel pipe out-of-roundness information after the pipe expanding step, the execution of the numerical computation conducted a plurality of times while changing the operational condition dataset, and generates, by this numerical computation, a plurality of pairs of data of the operational condition data set and the steel pipe out-of-roundness information data after the pipe expanding step corresponding to the operational condition dataset, as training data; an out-of-roundness prediction model generation unit that generates an out-of-roundness prediction model for which an input data is the operational condition dataset, and an output data is the out-of-roundness information of the steel pipe after the pipe expanding step, the generation of the out-of-roundness prediction model performed by machine learning using the plurality of pairs of training data generated by the basic data acquisition unit; an operational parameter acquisition unit that acquires online an operational condition dataset to be set as an operational condition of the steel pipe manufacturing process; and an out-of-roundness prediction unit that predicts online the steel pipe out-of-roundness information after the pipe expanding step corresponding to the operational condition dataset acquired by the operational parameter acquisition unit, using the out-of-roundness prediction model generated by the out-of-roundness prediction model generation unit.

Moreover, the steel pipe out-of-roundness prediction device may further include a terminal device including an input unit that acquires input information based on a user's operation, and a display unit that displays the out-of-roundness information, wherein the operational parameter acquisition unit may update a part or all of the operational condition dataset in the steel pipe manufacturing process based on the input information acquired by the input unit, and the display unit may display the steel pipe out-of-roundness information predicted by the out-of-roundness prediction unit by using the updated operational condition dataset.

Advantageous Effects of Invention

According to the steel pipe out-of-roundness prediction model generation method of the present invention, it is possible to generate an out-of-roundness prediction model that accurately and promptly predicts an out-of-roundness of a steel pipe after a pipe expanding step in a steel pipe manufacturing process including a plurality of steps. In addition, according to the steel pipe out-of-roundness prediction method and the steel pipe out-of-roundness prediction device of the present invention, it is possible to accurately and promptly predict out-of-roundness of the steel pipe after the pipe expanding step in the steel pipe manufacturing process including a plurality of steps. In addition, according to a steel pipe out-of-roundness control method according to the present invention, it is possible to accurately control the out-of-roundness of the steel pipe after the pipe expanding step in the steel pipe manufacturing process including a plurality of steps. Furthermore, according to the steel pipe manufacturing method according to the present invention, it is possible to manufacture a steel pipe having desired out-of-roundness with good yield.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a diagram illustrating a steel pipe manufacturing process according to an embodiment of the present invention.

FIG. 2 is a perspective view illustrating an overall configuration of a C-press device.

FIG. 3 is a cross-sectional view illustrating a configuration of a press mechanism.

FIG. 4 is a diagram illustrating an example of a step of forming a formed body having a U-shaped cross section using a press bending device.

FIG. 5 is a diagram illustrating an example of a step of forming a formed body having a U-shaped cross section using a press bending device.

FIG. 6 is a diagram illustrating a configuration example of a pipe expanding device.

FIG. 7 is a diagram illustrating a configuration example of a measuring device of an outer diameter shape of a steel pipe.

FIG. 8 is a block diagram illustrating a configuration of a steel pipe out-of-roundness prediction device according to an embodiment of the present invention.

FIG. 9 is a block diagram illustrating a configuration of an out-of-roundness offline calculation unit illustrated in FIG. 8.

FIG. 10 is a diagram illustrating an example of a change in a relationship between a press working amount and out-of-roundness of a steel pipe after a pipe expanding step according to a change in the operational condition of the press bending step.

FIG. 11 is a diagram illustrating an example of a press depression position and a press depression amount for each number of times of pressing.

FIG. 12 is a diagram illustrating a steel pipe out-of-roundness control method according to an embodiment of the present invention.

FIG. 13 is a diagram illustrating a configuration of a steel pipe out-of-roundness prediction device according to an embodiment of the present invention.

FIG. 14 is a diagram illustrating an example of a finite element model.

DESCRIPTION OF EMBODIMENTS

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[Steel Pipe Manufacturing Process]

FIG. 1 is a diagram illustrating a steel pipe manufacturing process according to an embodiment of the present invention. As illustrated in FIG. 1, the steel pipe manufacturing process according to an embodiment of the present invention uses a thick steel sheet manufactured by a thick sheet rolling step which is a preprocessing step of the steel pipe manufacturing process, as a steel sheet to be a material. Here, the thick steel sheet typically has a yield stress of 245 to 1050 MPa, a tensile strength of 415 to 1145 MPa, a sheet thickness of 6.4 to 50.8 mm, a sheet width of 1200 to 4500 mm, and a length of 10 to 18 m. In addition, the end of the thick steel sheet in the width direction is ground in advance into a chamfered shape referred to as a bevel. This step is done for preventing overheating of the outer surface corner portion of the end of the sheet in the width direction to stabilize the welding strength in the subsequent welding step. In addition, since the width of the thick steel sheet will influence the outer diameter after formation into the steel pipe, the width is adjusted to a predetermined range in consideration of the deformation history in the subsequent steps.

The steel pipe manufacturing process includes execution of an end bending step of applying bending to the end of the steel sheet in the width direction. The end bending step, which is performed by a C-press device, performs an end bending processing (also referred to as crimping work) on the end of the steel sheet in the width direction. The C-press device includes a pair of upper and lower molds and a pair of upper and lower clamps that hold the central portion of the steel sheet in the width direction. Since the length of the mold is shorter than the length of the steel sheet, the end bending processing is repeated while sequentially feeding the steel sheet in the longitudinal direction. Such end bending processing is applied on both ends of the steel sheet in the width direction. Since it is difficult to apply a bending moment to the end in the width direction in the 3-point bending press, the end bending step is performed to apply bending deformation in advance by the molds. This makes it possible to improve the out-of-roundness of the steel pipe as a final product. At this time, examples of the operational parameters for specifying the machining conditions include: an end bending processing width which is a length at which the mold comes into contact with the steel sheet from the end in the width direction of the steel sheet toward the central direction in the width direction; a gripping force of the clamp; a feeding amount of the mold when the end bending processing is repeated in the longitudinal direction of the steel sheet, a feeding direction, the number of times of feeding, and the like.

The subsequent press bending step is a step of processing the steel sheet into a formed body having a U-shaped cross section by performing 3-point bending press using a punch a plurality of times by a press bending device. Note that the manufacturing process often includes, after the press bending step, a seam gap reduction step of reducing a seam gap of a formed body having a U-shaped cross section using an O-press device so as to form an open pipe. However, the present embodiment omits the seam gap reduction step and perform the welding step on the formed body having a U-shaped cross section that has undergone the press bending step. Hereinafter, the formed body having a U-shaped cross section obtained by the press bending step is also referred to as an open pipe. The subsequent welding step is a step of restraining the seam gap portion formed at the end of the open pipe so as to allow the ends to be in contact with each other, and joining the ends to each other. With this step, the formed body is formed into a steel pipe having the ends joined to each other. The subsequent pipe expanding step is a step of using a pipe expanding device including a plurality of pipe expanding tools each having a curved surface obtained by dividing an arc into a plurality of portions to expand the steel pipe by bringing the curved surface of the pipe expanding tool into contact with the inner surface of the steel pipe. The steel pipe manufactured in this manner is subjected to the determination, in the inspection step, as to whether the quality such as the material property, the appearance, and the dimension satisfies a predetermined specification, and then shipped as a product. The inspection step includes an out-of-roundness measurement step of measuring the out-of-roundness of the steel pipe.

In the present embodiment, an end bending step, a press bending step, a seam gap reduction step, and a pipe expanding step, among a series of manufacturing steps including formation of a steel sheet into an open pipe and a pipe expanding step performed after welding, are referred to as “forming processing step”. These steps are common steps of controlling the dimensions and shape of the steel pipe by applying plastic deformation to the steel sheet. Hereinafter, individual steps included in the steel pipe manufacturing process will be described in detail with reference to the drawings.

A C-press device that conducts end bending processing will be described in detail with reference to FIGS. 2 and 3. FIG. 2 is a perspective view illustrating an overall configuration of the C-press device. As illustrated in FIG. 2, a C-press device 30 includes: a conveyance mechanism 31 that conveys a steel sheet S in a direction along the longitudinal direction as a conveyance direction; a press mechanism 32A that performs bending processing on one width direction end Sc of the sheet to a predetermined curvature with the downstream side of the steel sheet S in the conveyance direction as a front side; a press mechanism 32B that performs bending processing on the other end Sd in the width direction to a predetermined curvature; and an interval adjustment mechanism (not illustrated) that adjusts the interval between the press mechanisms 32A and 32B on left and right according to the width of the steel sheet S subjected to end bending processing. The conveyance mechanism 31 includes a plurality of rotationally driven conveyance rolls 31a each disposed in front of and behind the press mechanisms 32A and 32B. Note that a reference sign Sa in the drawing indicates a leading end (front end in the longitudinal direction) of the steel sheet S.

FIG. 3(a) illustrates a cross section in the width direction of the press mechanism 32A that conducts bending processing on the one end Sc in the width direction of the steel sheet S as viewed in the direction from the upstream side to the downstream side in the conveyance direction of the steel sheet S. The press mechanism 32A and the press mechanism 32B are bilaterally symmetrical and have an identical configuration. The press mechanisms 32A/32B include an upper mold 33 and a lower mold 34 as a pair of molds arranged to face each other in the vertical direction, and a hydraulic cylinder 36 as a mold moving means that pushes up the lower mold 34 together with the tool holder 35 (moves in a direction approaching the upper mold 33) and clamps the molds with a predetermined pressing force (C-press force). The press mechanisms 32A and 32B may include a clamp mechanism 37 that grips the steel sheet S on the inner side of the upper mold 33 and the lower mold 34 in the width direction. The length of the steel sheet S in the longitudinal direction of the upper mold 33 and the lower mold 34 is normally shorter than the length of the steel sheet S. In this case, the end bending processing is performed a plurality of times while intermittently feeding the steel sheet S in the longitudinal direction by the conveyance mechanism 31 (refer to FIG. 2).

In the end bending step, the lower mold 34 in contact with a surface on the outer side in the bending direction of the ends Sc and Sd in the width direction of the steel sheet S subjected to the end bending processing has a pressing surface 34a facing the upper mold 33. The upper mold 33 has a forming surface 33a having a convex curved surface facing the pressing surface 34a and having a curvature radius corresponding to the inner diameter of the steel pipe to be manufactured. The pressing surface 34a has a concave curved surface shape approaching the upper mold 33 toward the outside in the width direction. However, although the pressing surface 34a of the lower mold 34 has a concave curved surface shape, the pressing surface may be any surface that approaches the upper mold 33 as it goes outward in the width direction, and may be an inclined plane. The curved surface shape of the upper mold 33 and the lower mold 34 is appropriately designed according to the thickness, the outer diameter, and the like of the steel sheet S, and may be appropriately selected and used according to the material to be processed.

FIG. 3(b) is a cross section of the press mechanism 32A in the width direction at the same position as FIG. 3(a), illustrating a state in which the lower mold 34 is pushed up by the hydraulic cylinder 36 and clamped. The lower mold 34 is pushed up by the hydraulic cylinder 36, and the end Sc of the steel sheet S in the width direction has been subjected to bending processing so as to be formed into a shape along the arc-shaped forming surface 33a of the upper mold 33. The width to be subjected to end bending (end bending processing width) varies depending on the width of the steel sheet S, and is generally about 100 mm to 400 mm.

FIG. 4 is a diagram illustrating an example of a step of forming a formed body having a U-shaped cross section using a press bending device. In the drawing, reference numeral 1 denotes a die disposed in a conveyance path of the steel sheet S. The die 1 includes a rod-shaped members 1a and 1b, provided in a left-right pair that support the steel sheet S at two positions in the conveyance direction, and an interval ΔD of the members can be changed according to the size of the steel pipe to be formed. In addition, a reference numeral 2 denotes a punch movable in a direction approaching and separating from the die 1. The punch 2 includes: a punch leading end 2a having a downward convex processing surface that is in direct contact with the steel sheet S and presses the steel sheet S in a concave shape; and a punch support 2b that is connected to the back surface of the punch leading end 2a and supports the punch leading end 2a. The maximum width of the punch leading end 2a is normally equal to the width (thickness) of the punch support 2b.

When performing bending processing on the steel sheet S using the press bending device having the above-described configuration, the steel sheet S is placed on the die 1, and 3-point bending press is sequentially performed by the punch 2 from both ends of the steel sheet S in the width direction toward the central portion as illustrated in FIG. 5 while intermittently feeding the steel sheet S at a predetermined feeding amount. FIG. 5 is a diagram illustrating a step of forming a formed body S₁as illustrated in the right row diagram (j) in which a steel sheet S, preprocessed with end bending processing, is processed as in the top to the bottom of the left row (first half of processing (a) to (e)) then processed as in the top to the bottom of the center row (second half of processing (f) to (i)), by operations including bending processing and feeding of the steel sheet S. In FIG. 5, arrows assigned to the steel sheet S and the punch 2 indicate the moving directions of the steel sheet S and the punch 2 in individual steps. In the formed body S₁having a U-shaped cross section as a result of being processed in this step, the gap between the ends is referred to as a “seam gap”.

Here, examples of the operational parameters of determining operational conditions for the press bending step include the number of times of pressing, press position information, a press depression amount, a lower die interval, a punch curvature, and the like.

The number of times of pressing refers to the total number of times of pressing the steel sheet in the width direction by the 3-point bending press. The more the number of times of pressing, the smoother curved shape is obtained as a formed body having the U-shaped cross section, and the more improvement obtained in the out-of-roundness of the steel pipe after the pipe expanding step.

The press position information refers to a position in the width direction of the steel sheet to be pressed by the punch. Specifically, the position can be specified by a distance from one end of the steel sheet in the width direction or a distance based on the central portion of the steel sheet in the width direction. The press position information is preferably treated as data linked to the number of times of pressing (orders including the first to N-th times of pressing).

The press depression amount refers to a thrust amount of the punch at each press depression position. The press depression amount is defined as an amount by which the lower end surface of the punch leading end 2a protrudes downward from a line connecting points on the uppermost surfaces of the die 1 illustrated in FIG. 4. At this time, since the thrust amount of the punch leading end 2a can be set to a different value for each pressing, it is preferable to treat the number of times of pressing and the press depression amount as linked data. Therefore, when the number of times of pressing is N, the operational condition of the press bending step is specified by 1 to N datasets with the number of times of pressing, the press position information, and the press depression amount as one set of datasets.

The reason why these datasets are used is that, by partially changing the press position and the pushing amount of the punch in the press bending step, the entire cross-sectional shape the pipe changes in a state of an open pipe, which has an impact on the out-of-roundness of the steel pipe after the pipe expanding step. Still, there is no need to use all the N datasets as input variables of an out-of-roundness prediction model to be described below. It is allowable to use a method of selecting a condition having a large impact on the out-of-roundness of the steel pipe after the pipe expanding step and generating an out-of-roundness prediction model using press position information and the press depression amount at the start (first time) or the end (N-th time) of the press bending step.

The lower die interval is an interval between the rod-shaped members 1a and 1b, provided in a left-right pair as illustrated in FIG. 4, and is a parameter expressed by ΔD in the drawing. An increase in the lower die interval will change the local curvature of the steel sheet even for the same press depression amount, having an impact on the out-of-roundness of the steel pipe after the pipe expanding step. Therefore, it is preferable to use the lower die interval set according to the size of the steel pipe to be formed as the operational parameter of the press bending step. In addition, in a case where the lower die interval is changed every time the punch is pushed, data linked with the number of times of pressing may be used as the operational parameters.

The punch curvature refers to a curvature of a punch leading end used for the pressing. An increase in the punch curvature will also increase the local curvature applied to the steel sheet at the time of 3-point bending press, having an impact on the out-of-roundness of the steel pipe after the pipe expanding step. However, since it is difficult to change the punch curvature for each pressing when forming one steel sheet, it is preferable to use the punch curvature set according to the size of the steel pipe to be formed, as the operational parameter of the press bending step.

As in the present embodiment, omitting the seam gap reduction step using an O-press device or the like after the press bending step will likely to increase the seam gap of the formed body, leading to deterioration of out-of-roundness after the pipe expanding step. Therefore, as compared with the case of using the seam gap reduction step, the press depression amount at the central portion in the width direction of the steel sheet S is often set to be large. However, setting the press depression amount at the central portion in the width direction of the steel sheet S to an excessively large amount will cause the end of the formed body in the width direction to come into contact with the punch support 2b. Therefore, there is a possibility of occurrence of an upper limit in the press depression amount.

Thereafter, the formed body S₁having a U-shaped cross section and having undergone forming processing by the press bending step is formed into a steel pipe by butting end surfaces of seam gap portions against each other and welding the end surfaces by a welding machine (joining means). The welding machine (joining means) to be used is, for example, a device including three types of welding machines, that is, a tack welding machine, an inner surface welding machine, and an outer surface welding machine. In these welding machines, the tack welding machine brings the surfaces butted by the cage rolls into close contact with each other continuously in an appropriate positional relationship, and welds the close contact portion over the entire length in the pipe axis direction. Next, the tacked pipe is welded (submerged arc welding) from the inner surface of the butt portion by an inner surface welding machine, and further welded (submerged arc welding) from the outer surface of the butt portion by an outer surface welding machine.

The steel pipe having the seam gap portion welded is processed such that a pipe expanding device is inserted into the steel pipe to expand the diameter of the steel pipe (referred to as pipe expansion). FIGS. 6(a) to 6(c) are diagrams illustrating a configuration example of the pipe expanding device. As illustrated in FIG. 6(a), the pipe expanding device includes a plurality of pipe expanding dies 16 having curved surfaces obtained by dividing an arc into a plurality of arcs along the circumferential direction of a tapered outer circumferential surface 17. When expanding the steel pipe using the pipe expanding device, as illustrated in FIGS. 6(b) and 6(c), the pipe expanding die 16 is first aligned with a pipe expanding start position by moving the steel pipe P using a steel pipe moving device. Then, by retracting a pull rod 18 from the pipe expanding start position, first pipe expanding processing is performed.

This allows each of the pipe expanding dies 16 in sliding contact with the tapered outer circumferential surface 17 to be displaced in the radial direction by a wedge action, expanding the steel pipe P. This reduces the level of irregularity in the cross-sectional shape of the steel pipe P, bringing the cross-sectional shape of the steel pipe P closer to a perfect circular shape. Next, the pull rod 18 is advanced to the pipe expanding start position to return the pipe expanding dies 16 to the inside in the vertical axial direction by a release mechanism, and then the steel pipe P is further moved by an amount corresponding to the pitch (axial length) of the pipe expanding dies 16. The pipe expanding dies 16 are adjusted to new pipe expanding positions, and then the above operation is repeated. This makes it possible to perform the first pipe expanding processing over the entire length of the steel pipe P by the pitch of the pipe expanding dies 16.

At this time, examples of the operational parameters for determining the operational conditions of the pipe expanding step include the pipe expansion rate, the number of pipe expanding dies, and the diameter of the pipe expanding dies. The pipe expansion rate refers to a ratio of a difference between an outer diameter after pipe expansion and an outer diameter before pipe expansion to an outer diameter before pipe expansion. The outer diameter before and after pipe expansion can be calculated by measuring the circumferential length of the steel pipe. The pipe expansion rate can be adjusted by a stroke amount when the pipe expanding dies are expanded in a radial direction. The number of pipe expanding dies refers to the number of dies that abut on the steel pipe disposed in the circumferential direction when pipe expansion is performed. The pipe expanding die diameter refers to a curvature of a portion of each pipe expanding die that abuts on the steel pipe.

Among these, the operational parameter that can easily adjust the out-of-roundness after the pipe expanding step is the pipe expansion rate. An increase in the pipe expansion rate will apply the curvature of the region in contact with the pipe expanding die over the entire circumference uniformly onto the pipe according to the pipe expanding dies R, leading to improvement of the out-of-roundness. At this time, the greater the number of pipe expanding dies, the more the local variation in curvature in the circumferential direction of the steel pipe can be suppressed, making it possible to achieve satisfactory out-of-roundness of the steel pipe after the pipe expanding step. On the other hand, however, an excessive large pipe expansion rate might lower the compressive yield strength of the steel pipe product due to the Bauschinger effect. When a steel pipe is used as a line pipe or the like, high compressive stress acts in the pipe circumferential direction. Therefore, high compressive yield strength is required as a material of the steel pipe, and it is not appropriate to increase the pipe expansion rate more than necessary. Consequently, the pipe expansion rate in actual operation is set such that the out-of-roundness of the steel pipe falls within a predetermined value at a pipe expansion rate lower than the preset upper limit value of the pipe expansion rate.

<Out-of-Roundness Measurement Step>

In an inspection step as the last step in the manufacturing process of the steel pipe, the quality of the steel pipe is inspected, and the out-of-roundness of the steel pipe is measured. The out-of-roundness measured in the out-of-roundness measurement step is an index representing the degree of deviation of the outer diameter shape of the steel pipe from the perfect circle. Usually, the closer the out-of-roundness is to zero, the closer the cross-sectional shape of the steel pipe is to a perfect circle. The out-of-roundness is calculated based on the outer diameter information of the steel pipe measured by the out-of-roundness measurement device. For example, the tube is equally divided in the circumferential direction at a certain tube length position and the outer diameters at the opposing positions are measured. When the maximum diameter and the minimum diameter among them are defined as Dmax and Dmin, respectively, the out-of-roundness can be defined as Dmax−Dmin. At this time, as more the number of divisions into equal parts, the more the small irregularities in the steel pipe after the pipe expanding step can be a numerical index, which is preferable. Specifically, it is preferable to use information divided into 4 to 36000 equal parts. It is more preferable to use equal parts of 360 or more.

However, the out-of-roundness does not necessarily have to be represented by the difference between the maximum diameter and the minimum diameter. It is also allowable to use a method of calculating an equivalent temporary perfect circle (diameter) having the same area as the area inside the curve of a figure representing the outer diameter shape of the steel pipe in a continuous diagram and defining a region deviated from the outer diameter shape of the steel pipe as an image based on the temporary perfect circle. Note that the out-of-roundness of the steel pipe after the pipe expanding step in the present embodiment may be referred to as out-of-roundness information including the out-of-roundness represented by such an image. As a means of measuring the outer diameter shape of the steel pipe, for example, the following methods can be used.

(a) As illustrated in FIG. 7(a), using a device including an arm 20 rotatable by 360 degrees about a substantially central axis of the steel pipe P, displacement gauges 21a and 21b attached to the distal end of the arm 20, and a rotation angle detector 22 that detects the rotation angle of the rotation shaft of the arm 20, a distance between the rotation center of the arm 20 and a measurement point on the outer circumference of the steel pipe P is measured with the displacement gauges 21a and 21b for each unit of minute angle of rotation of the arm 20, and then the outer diameter shape of the steel pipe P is specified based on the measured value.

(b) As illustrated in FIG. 7(b), using a device including a rotating arm 25 that rotates about the central axis of the steel pipe P, a frame (not illustrated) that is provided on the end side 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 comes into contact with the outer surface and the inner surface of the end of the steel pipe P and rotates with the rotation of the rotating arm 25, and a pair of pressing air cylinders fixed to the frame that presses the pressing rollers 26a and 26b against the outer surface and the inner surface of the steel pipe P, the outer diameter shape of the steel pipe P is specified based on the amount of movement in the radial direction of the frame and the pressing positions of the pressing rollers 26a and 26b by the individual pressing air cylinders.

Here, in the present embodiment, by comparing an out-of-roundness prediction result obtained by the out-of-roundness prediction model to be described below with the measured value of the out-of-roundness obtained in the inspection step described above, it is possible to verify the out-of-roundness prediction accuracy. Therefore, with respect to the prediction result of the out-of-roundness prediction model to be described below, it is also possible to improve the prediction accuracy by adding the performance value of the prediction error to the prediction result of the out-of-roundness prediction model.

[Steel Pipe Out-of-Roundness Prediction Device]

FIG. 8 is a block diagram illustrating a configuration of a steel pipe out-of-roundness prediction device according to an embodiment of the present invention. FIG. 9 is a block diagram illustrating a configuration of an out-of-roundness offline calculation unit 112 illustrated in FIG. 8.

As illustrated in FIG. 8, a steel pipe out-of-roundness prediction device 100 according to an embodiment of the present invention includes an information processing device such as a workstation, and includes a basic data acquisition unit 110, a database 120, and an out-of-roundness prediction model generation unit 130.

The basic data acquisition unit 110 includes: an operational condition dataset 111 obtained by quantifying factors having an impact on the out-of-roundness of the steel pipe through the end bending step, the press bending step, the welding step, and the pipe expanding step; and an out-of-roundness offline calculation unit 112 that outputs the out-of-roundness after the pipe expanding step using the operational condition dataset 111 as an input condition.

In the present embodiment, the operational condition dataset 111 includes at least operational parameters of the end bending step and operational parameters of the press bending step. This is because the information will have a large impact on the out-of-roundness of the steel pipe after the pipe expanding step, and are factors having an impact on the variation in the out-of-roundness. In addition, it is preferable to include attribute information of the steel sheet to be a material and operational parameters in the pipe expanding step. In addition, it is also allowable to include operational parameters of the welding step. Data used for the operational condition dataset 111 will be described below.

The basic data acquisition unit 110 executes numerical computation by the out-of-roundness offline calculation unit 112 while variously changing the parameters included in the operational condition dataset 111, thereby calculating the out-of-roundness of the steel pipe after the pipe expanding step corresponding to the plurality of operational condition datasets 111. The range of changing the parameter included in the operational condition dataset 111 is determined based on a range that can be changed as a normal operational condition according to the size of the steel pipe to be manufactured, the specification of the facility of each step, and the like.

The out-of-roundness offline calculation unit 112 calculates the shape of the steel pipe after the pipe expanding step by numerical analysis through a series of manufacturing steps from the end bending step to the pipe expanding step, and obtains the out-of-roundness of the steel pipe from the shape of the steel pipe after the pipe expanding step. Here, the series of manufacturing steps includes the end bending step, the press bending step, and the pipe expanding step. As illustrated in FIG. 9, the out-of-roundness offline calculation unit 112 includes: finite element model generation units 112a to 112c corresponding to the respective steps; and a finite element analysis solver 112d.

The finite element model generation unit in the end bending step performs element division inside the steel sheet based on attribute information of the steel sheet. The element division is automatically performed based on a preset element division condition. The finite element model of the end bending step that has undergone the element division is transferred to the finite element analysis solver 112d together with the calculation conditions in the end bending step. The calculation conditions in the end bending step include operational parameters of the end bending step, and also include all information necessary for executing finite element analysis specifying all boundary conditions such as physical property values of a workpiece, a tool, and the like, geometric boundary conditions, and mechanical boundary conditions. The shape of the steel sheet and stress-strain distribution obtained by the finite element analysis in the end bending step are transferred to the finite element model generation unit 112a in the press bending step as initial conditions related to the workpiece to be processed in the press bending step.

The finite element analysis solver 112d can be obtained as a variety of commercially available general-purpose analysis software, and thus it is possible to appropriately select and incorporate some of these for effective application. Alternatively, it is also allowable to use a mode in which the finite element analysis solver 112d is mounted on a computer separate from the out-of-roundness offline calculation unit 112, and input data including a finite element model and output data as a calculation result are transmitted to and received from the out-of-roundness offline calculation unit 112. This is because once a finite element model corresponding to each step is generated, numerical analysis can be performed by a single finite element analysis solver.

The finite element method is a type of approximate solution method of dividing a continuous body into finite elements. While being the approximate solution, the finite element method is a method to obtain a solution that satisfies the balance of forces and the continuity of displacement at the node points of the elements, making it possible to obtain a highly accurate solution even with non-uniform deformation. In the finite element method, stress, strain, and displacement in an element are independently defined for each element, and are associated with displacement (velocity) of a node so as to be formulated as a problem of solving simultaneous equations. In a widely used method, the strain (increment) or stress is evaluated by using displacement (speed) at a node of an element as an unknown value at this time.

In addition, the finite element method is characterized in that calculation is performed based on the principle of virtual work expressed in an integral form with respect to an equilibrium condition of stress in an element. The accuracy of the analysis result varies depending on conditions such as element division. In addition, the calculation time required for the analysis is usually long. Still, the finite element method is characterized by its capability of obtaining a solution even for a problem difficult to solve by other methods, as a solution that satisfies a basic equation of plastic mechanics in a node or an element. Therefore, even with a complicated forming history in the manufacturing process of the steel pipe, it is possible to obtain solutions of the displacement, stress field, and strain field of the workpiece close to the actual phenomenon.

A part of the finite element analysis solver may be replaced with various numerical analysis methods or approximate solutions such as a slip line field theory or an energy method. This makes it possible to shorten the overall calculation time. In addition, the finite element analysis used in the present embodiment executes elastoplastic analysis and does not include analysis of a temperature field such as heat conduction analysis. However, when the working speed is high and the temperature rise of the workpiece is large due to working heat generation, it is allowable to perform an analysis combining the heat conduction analysis and the elastoplastic analysis. In addition, the elastoplastic analysis of the present embodiment is a cross-sectional two-dimensional analysis in any case of the end bending step, the press bending step, and the pipe expanding step, and it is sufficient to perform numerical analysis on an end bending shape cross section, a cross section of the stationary portion in the longitudinal direction when the steel sheet is formed into an open pipe, and further into a steel pipe. For high accuracy prediction of the shape of the nonstationary portion such as the leading end and the trailing end of the steel pipe, it is preferable to have a finite element model generation unit that performs three-dimensional analysis including the leading end and the trailing end.

Regarding the steel sheet after the end bending processing, which is a workpiece to be processed in the press bending step, attribute information of the steel sheet is given as input data. At this time, the shape and stress/strain distribution of the steel sheet obtained as a result of performing finite element analysis in the end bending step will be initial conditions for the workpiece to be processed in the press bending step. Here, the finite element model generation unit 112b in the press bending step performs element division inside the steel sheet based on the dimension and shape of the steel sheet before the press bending step. The element division is automatically performed based on a preset element division condition. At this time, the distribution of stress and strain remaining inside may be allocated to each element based on the manufacturing history applied to the steel sheet in the preceding step. This is because, in the press bending step mainly including bending processing, the initial residual stress also has an impact on the shape of the U-shaped formed body of the processed steel sheet.

Together with the finite element model of the press bending step generated in this manner, the calculation condition in the press bending step is transferred as input data to the finite element analysis solver 112d. At this time, the calculation conditions in the press bending step are supposed to include operational parameters of the press bending step, and supposed to include all information necessary for executing finite element analysis specifying all boundary conditions such as physical property values of a workpiece, a tool, and the like, geometric boundary conditions, and mechanical boundary conditions.

The finite element analysis solver 112d performs numerical analysis under the calculation conditions given as above to obtain the shape of the open pipe and the distribution of stress and strain remaining inside after the press bending step. The result calculated in this manner is used for input data in the finite element model generation unit 112c of the next step, namely, the pipe expanding step. At this time, also in the welding step of welding the seam gap portion of the open pipe, the residual stress and strain generated in the steel pipe after welding may be obtained by numerical analysis of the welding process.

However, it is difficult, in many cases, to perform precise numerical analysis on the welding step due to the reasons such as the thermal conduction behavior accompanying the melting of the steel sheet at the time of welding, and the impact on the mechanical properties of the heat-affected zone. In addition, the heat-affected zone due to welding only affects the shape of a part of the steel pipe, and has a small impact on the shape of the entire steel pipe. Therefore, it is allowable to ignore the influence of the heat-affected zone created by welding on the out-of-roundness of the steel pipe after the pipe expanding step.

The welding step performs the welding while restraining the open pipe from the outside so as to reduce the seam gap of the open pipe, which changes the stress and strain distribution due to elastic deformation change in a portion other than the vicinity of the seam gap portion. Therefore, using the finite element analysis solver 112d, numerical analysis can be performed for the behavior of restraining the seam gap of the open pipe from the surroundings so as to set the seam gap of the open pipe to zero by the finite element method, and the result can be set to the stress-strain state after the welding step.

Meanwhile, when the step of reducing the seam gap in such a welding step corresponds to elastic deformation, the stress/strain distribution after the welding step may be obtained by superimposing an analytical solution of stress and strain on the curved beam according to the beam theory on the distribution of stress and strain in the open pipe calculated by finite element analysis. This makes it possible to shorten the calculation time.

Based on the shape of the steel pipe after the welding step obtained as described above, the finite element model generation unit 112c in the pipe expanding step performs element division inside the steel pipe. The element division is automatically performed based on a preset element division condition. At this time, it is preferable to assign the distribution of stress and strain calculated as described above to each element. The generated finite element model of the pipe expanding step is transferred to the finite element analysis solver 112d together with the calculation conditions in the pipe expanding step. The calculation conditions in the pipe expanding step are supposed to include operational parameters of the pipe expanding step according to the present embodiment, and supposed to also include all information necessary for executing finite element analysis specifying all boundary conditions such as physical property values of a workpiece, a tool, and the like, geometric boundary conditions, and mechanical boundary conditions.

The finite element analysis solver 112d performs numerical analysis under the calculation conditions given as above to obtain the shape of the steel pipe and the distribution of internal stress and strain after the pipe expanding step. The calculated shape of the steel pipe has a non-uniform curvature distribution in the circumferential direction, and the out-of-roundness of the steel pipe is obtained according to the definition of the out-of-roundness in the out-of-roundness measurement step. The numerical analysis using the finite element method by the out-of-roundness offline calculation unit 112 sometimes requires a calculation time of about 1 to 10 hours for one operational condition dataset (single case).

However, since the processing is executed offline, there is no constraint on the calculation time. In order to shorten the calculation time for a large number of operational condition datasets, numerical computation corresponding to a plurality of operational condition datasets may be executed in parallel using a plurality of computers. This makes it possible to build a database for generating an out-of-roundness prediction model in a short period of time. Furthermore, in recent years, the calculation time per case is about ½ to 1/10 of the conventional calculation time by calculation using a General-Purpose computing on Graphics Processing Unit (GPGPU), and such a computer tool may be used.

The description returns to FIG. 8. The database 120 stores the operational condition dataset 111 and its corresponding data related to the out-of-roundness of the steel pipe after the pipe expanding step. The data stored in the database 120 can be obtained offline. Unlike the database accumulated as the performance value of the actual operation, the operational condition dataset can be flexibly set, making it a database suitable for machine learning with less likelihood of occurrence of statistical bias in the operational condition of the operational condition dataset. In addition, since calculation results obtained by precise numerical analysis are accumulated, with no accumulation of training data fluctuating over time, the more the data is accumulated, the more the useful database can be obtained.

The out-of-roundness prediction model generation unit 130 generates an out-of-roundness prediction model M trained by machine learning to obtain the out-of-roundness of a steel pipe after the pipe expanding step for the input operational condition dataset 111 based on a relationship between the plurality of pairs of operational condition datasets 111 and the out-of-roundness of the steel pipe stored in the database 120. Incidentally, the relationship between the operational conditions in each step and the out-of-roundness of the steel pipe after the pipe expanding step may exhibit complicated nonlinearity. Therefore, while modeling assuming a linearity would result in prediction with low accuracy, the machine learning method using a function having nonlinearity, such as a neural network, will be able to perform prediction with high accuracy. Here, modeling means replacing an input/output relationship in numerical computation with an equivalent functional form.

The number of databases required to generate the out-of-roundness prediction model M is desirably 500 or more pieces of data, although it varies depending on conditions such as the size of the steel pipe to be manufactured. Preferably, 2000 or more pieces of data and more preferably 5000 or more pieces of data are to be used. The method of machine learning may be a known learning method. The machine learning may be implemented by using, for example, a known machine learning method such as a neural network. Examples of other methods include decision tree learning, random forest, Gaussian process regression, support vector regression, and k-nearest neighbor algorithm. While the out-of-roundness prediction model M is generated offline, the out-of-roundness prediction model generation unit 130 may be incorporated in an online control system, and the out-of-roundness prediction model may be periodically updated using a database calculated and accumulated offline as needed.

The out-of-roundness prediction model M for the steel pipe after the pipe expanding step generated as described above has the following characteristics.

First, the end bending step applies bending deformation by a mold to an end of a steel sheet to be a material in the width direction, having an impact on the out-of-roundness of the steel pipe after the pipe expanding step in the vicinity of the welded portion of the steel pipe. The reason is as follows. When bending deformation is applied to the steel sheet by 3-point bending press as in the press bending step, it is difficult to apply a bending moment to the end in the width direction, leading to a difficulty in reducing the curvature in the vicinity of the end of the steel sheet in the width direction. On the other hand, the press bending step is a step of applying a plurality of operations of bending deformation in the width direction of the steel sheet, having an impact on the curvature distribution in the circumferential direction occurring in the open pipe. This will have an impact on the entire circumferential direction of the steel pipe regarding the out-of-roundness of the steel pipe after the pipe expanding step. In this manner, the end bending step and the press bending step have different application position of bending deformation in the width direction of the steel sheet. Therefore, it is preferable to predict the out-of-roundness of the steel pipe after the pipe expanding step by combining the operational conditions of the two steps.

Meanwhile, when the curvature to be applied to the steel sheet in the end bending step is small, the deformation of the end in the width direction is to be small. Therefore, unless relatively large bending deformation is applied in the press bending step, the seam gap of the open pipe would not be reduced, leading to deterioration of the out-of-roundness of the steel pipe after the pipe expanding step. Conversely, when the curvature applied to the steel sheet in the end bending step is large, the seam gap of the open pipe would be too small unless the bending deformation in the press bending step is suppressed. This case also leads to deterioration of the out-of-roundness of the steel pipe after the pipe expanding step. Accordingly, combining the operational conditions in the end bending step and the operational conditions in the press bending step will be the way to achieve satisfactory out-of-roundness of the steel pipe after the pipe expanding step, and therefore, the out-of-roundness prediction model M is created in consideration of such factors.

That is, as the attribute information of the steel sheet to be a material, for example, the yield stress, the sheet thickness, and the like has a certain level of variation in manufacturing the steel sheet, and this will have an impact on the curvature of the steel sheet after unloading of the C-press device in the end bending step and the curvature of the steel sheet at the time of punch thrusting in the 3-point bending press in the press bending step or the curvature after unloading. Therefore, by selecting the attribute information of these steel sheets as input parameters of the out-of-roundness prediction model M generated offline, it is possible to predict the influence of the attribute information such as the yield stress and the sheet thickness of the material on the out-of-roundness of the steel pipe after the pipe expanding step.

For example, FIG. 10 illustrates a result of measuring the out-of-roundness of a steel pipe after a pipe expanding step (with same operational conditions of the pipe expanding step) by changing the press depression amount under the pressing pressure in the first pass in the press bending step when the end bending processing width in the end bending step is set to 180 mm, 200 mm, and 220 mm under the condition that the number of times of pressing is set to nine in the press bending step in manufacturing a steel pipe having an outer diameter of 30 inches and a pipe thickness of 44.5 mm. FIG. 10 illustrates a result of changing the depression amount (first pass depression amount) at the time of the initial (first) pressing while using the same settings in the other operational conditions in the press bending step.

As illustrated in FIG. 10, the out-of-roundness of the steel pipe after the pipe expanding step varies depending on the end bending processing width which is an operational parameter in the end bending processing and the first pass depression amount which is an operational parameter in the press bending step. At this time, when the out-of-roundness of the steel pipe after the pipe expanding step is to be controlled to achieve the same target (for example, the out-of-roundness of 0.68% is set as a target value), it is necessary to appropriately change the first pass depression amount in the press bending step depending on the end bending processing width in the end bending step. This means that there is a case where attribute information of the steel sheet varies, and the deformation state (curvature) of the steel sheet after the end bending step might vary even with the same operational conditions of the end bending step, and that the out-of-roundness of the steel pipe after the pipe expanding step varies as a result without appropriately controlling the operational conditions of the press bending step. In this manner, in order to appropriately control the out-of-roundness of the steel pipe after the pipe expanding step, it is necessary to change the operational conditions of the press bending step in accordance with the operational conditions of the end bending step, and it is difficult to set an appropriate operational condition merely by treating the operational condition of each of the end bending step and the press bending step as an independent parameter.

In this regard, the out-of-roundness prediction model of the present embodiment can consider the influence of such operational parameters of a plurality of manufacturing steps on the out-of-roundness of the steel pipe after the pipe expanding step, making it possible to predict the out-of-roundness with high accuracy. In addition, since the out-of-roundness prediction model trained by machine learning is generated, it is possible to immediately calculate the out-of-roundness to be output even when the variable to be input condition is changed, which lead to a feature of capability to immediately set and correct the operational conditions even in the case of using online. Hereinafter, each parameter used for input for the out-of-roundness prediction model will be described.

Applicable attribute information of the steel sheet to be a material, in a case where the attribute information is to be used for the input to the out-of-roundness prediction model, can be any parameter having an impact on the out-of-roundness of the steel pipe after the pipe expanding step, such as yield stress of the steel sheet, tensile strength, modulus of longitudinal elasticity, sheet thickness, sheet thickness distribution in the sheet surface, yield stress distribution in the sheet thickness direction of the steel sheet, the degree of the Bauschinger effect, and surface roughness. In particular, it is preferable to use, as an index, a factor having an impact on the spring back of the steel sheet at the end in the width direction in the end bending step, or a factor having an impact on the deformation state or the spring back of the steel sheet by the 3-point bending press in the press bending step.

The yield stress of the steel sheet, the distribution of the yield stress in the sheet thickness direction of the steel sheet, and the sheet thickness have a direct impact on the state of stress and strain in the 3-point bending press. The tensile strength is a parameter reflecting the state of work hardening in bending processing, and has an impact on the stress state during bending deformation. The Bauschinger effect has an impact on the yield stress and the subsequent work hardening behavior when the loading due to bending deformation is reversed, and impact on the stress state during bending deformation. In addition, the modulus of longitudinal elasticity of the steel sheet has an impact on the spring back behavior after the bending processing. Furthermore, the sheet thickness distribution in the sheet surface generates the distribution of the bending curvature in the press bending step, leading to an impact on the out-of-roundness of the steel pipe after the pipe expanding step.

Among these pieces of attribute information, it is particularly preferable to use the yield stress, the representative sheet thickness, the sheet thickness distribution information, and the representative sheet width. These items are information measured in a quality inspection step of a thick sheet rolling step, which is a step of manufacturing a steel sheet to be a material, and have an impact on deformation behavior in an end bending step and a press bending step and have an impact on out-of-roundness of a steel pipe after a pipe expanding step. Therefore, these items are to be preferably used as attribute information of the steel sheet in the basic data acquisition unit 110.

The yield stress is information that can be obtained from a tensile test of a small test piece for quality assurance collected from a thick steel sheet to be a material, and may be a representative value in a plane of the steel sheet to be a material. In addition, the representative sheet thickness is a sheet thickness representing a sheet thickness in a plane of a steel sheet to be a material, and may be a sheet thickness of a central portion in a width direction of the steel sheet at an arbitrary position in a longitudinal direction of the steel sheet may be used, or an average value of sheet thicknesses in the longitudinal direction. Furthermore, an average value of the sheet thicknesses of the entire in-plane sheet of the steel sheet may be obtained and used as the representative sheet thickness.

In addition, the sheet thickness distribution information refers to information representing the sheet thickness distribution in the width direction of the steel sheet. A typical example is a crown occurring in a steel sheet. The crown represents a difference in the sheet thickness between a central portion of the steel sheet in the width direction, and a position away from the end of the steel sheet in the width direction by a predetermined distance (for example, 100 mm, 150 mm, or the like). The representative sheet width is a representative value regarding the width of the steel sheet as a material. There is a case where there is a variation in the width of the thick steel sheet to be a material or the width of the steel sheet varies when the end of the steel sheet in the width direction is ground by beveling, which will influence the variation in the outer diameter accuracy of the steel pipe to be a product.

The attribute information of the steel sheet described above is information used to set operational conditions in the manufacturing process of the steel pipe, as information collected by the host computer in the online operation. The basic data acquisition unit 110 may select, from among the pieces of information, appropriate information that matches the attribute information of the steel sheets collected by the online host computer.

Regarding the operational parameters of the end bending step, a parameter specifying the shape formed by the forming surface 33a of the upper mold 33 and the shape formed by the pressing surface 34a of the lower mold 34 used in the C-press device 30 can be used as the operational parameters. In addition, it is also allowable to use an end bending processing width (width subjected to end bending forming), a push-up force (C-press force), and a gripping force of the clamp mechanism 37 in the end bending step, as operational parameters. This is because these items are factors having an impact on the deformation of the end of the steel sheet in the width direction in the end bending step. When three-dimensional deformation analysis is performed for the end bending step, the feed amount, the feed direction, and the number of times of feeding of the steel sheet may be used as the operational parameters for the end bending step.

Here, there are cases where the shape formed by the forming surface 33a of the upper mold 33 is given as a shape having arcs of a plurality of curvature radii provided in a continuous shape or the shape is given by an involute curve or the like, allowing the use of a parameter for specifying a geometric cross-sectional shape. For example, when the cross-sectional shape is formed by a parabolic shape, the cross-sectional shape can be specified by using coefficients of a first-order term and a second-order term of a quadratic expression representing a parabola passing through an origin, and thus such a coefficient can be used as an operational parameter for the end bending step.

In another case where a plurality of molds are held and appropriately exchanged to be used for forming the shape formed by a forming surface 33a of the upper mold 33 according to conditions such as the outer diameter, the wall thickness, and the type of the steel pipe to be manufactured, a mold management number for specifying a mold to be used in the end bending step may be used as the operational parameter for the end bending step.

In the present embodiment, the operational parameter for the press bending step is used for the input to the out-of-roundness prediction model. The operational parameters for the press bending step may be various parameters having an impact on the local bending curvatures of the steel sheet and the distribution thereof in the sheet width direction, such as the number of times of pressing of the 3-point bending press, press position information, press depression amount, lower die interval, and punch curvature described above. In particular, it is preferable to use information including all of press position information and the press depression amount in an operation of the punch pressing the steel sheet, and the number of times of pressing performed through the press bending step. The method illustrated in FIG. 11 can be exemplified as a case of including all of these pieces of information.

FIGS. 11(a) and 11(b) each illustrate an example of the press depression position and the press depression amount when the punches are pressed with the number of times of pressing of 16 times/10 times with respect to the steel sheets having the same sheet width. At this time, the press depression position is information indicating a distance from a reference end of the steel sheet in the width direction, and is used as the press depression position information. In addition, the press depression amount is described corresponding to each press depression position, and such “number of times of depression”, “press depression position”, and “press depression amount” can be integrated as a set of data. In the examples illustrated in FIGS. 11(a) and 11(b), the operational parameters of the press bending step are specified by 16 sets of data and 10 sets of data at the number of times of pressing of 16 times and 10 times, respectively.

In the present embodiment, such a dataset is used as an input of an out-of-roundness prediction model in the following form. For example, as inputs of the out-of-roundness prediction model, it is possible to use a press depression position and a press depression amount when performing press depression at a position closest to the end at one end of the steel sheet, and possible to use a press depression position and a press depression amount when performing press depression at a position closest to the end at the other end of the steel sheet.

When the press depression amount at one end of the steel sheet is increased in the 3-point bending press, curvatures at portions of the steel pipe in FIG. 4 corresponding to approximately one o'clock and a portion corresponding to approximately eleven o'clock are increased, leading to formation of the formed body having a U-shaped cross section with a horizontally long shape as a whole. In addition, the closer the press depression position is to the end of the steel sheet, the lower the position of the seam gap portion, leading to formation of the formed body having a U-shaped cross section with a horizontally long shape as a whole. As a result, the steel pipe formed into the open pipe and subjected to the welding step and the pipe expanding step will also have a horizontally long shape as a whole, having an impact on the out-of-roundness. Furthermore, the punch curvature at the time of press depression, the total number of times of press depression, and the interval between the lower dies at the time of press depression also has an impact on the out-of-roundness.

Meanwhile, by using, as an input of the out-of-roundness prediction model, all the press depression position information and the data of the press depression amount together with the number of times of pressing, it is possible to further improve the prediction accuracy of the out-of-roundness prediction model. For example, based on the assumed maximum number of times of pressing, data of the press depression position and the press depression amount is stored according to the number of times of pressing in the case of performing pressing. The press depression position and the press depression amount in the subsequent press working with no pressing are set to zero. For example, in an assumed case where the maximum assumable number of times of pressing is 16 times in the examples illustrated in FIGS. 11(a) and 11(b), and when the number of times of pressing is 10 times, the data of the 11 to 16th depression, which is set to zero, will be input to the out-of-roundness prediction model.

The operational parameters of the press bending step described above are information used as operational conditions set by the host computer in online operation. The basic data acquisition unit 110 preferably selects a parameter to be used for the input to the out-of-roundness prediction model from among the operational parameters of the press bending step collected by the online host computer in this manner.

In addition to the operational parameters described above, when the operational parameters of the pipe expanding step are used for the input to the out-of-roundness prediction model, the pipe expansion rate can be used as the operational parameters of the pipe expanding step. Although the greater the pipe expansion rate, the more the out-of-roundness of the steel pipe after the pipe expanding step is improved, the upper limit value of the pipe expansion rate is limited from the viewpoint of the compressive yield strength as a steel pipe product, and therefore, the calculation condition of the basic data acquisition unit 110 is to be determined using a value within the range. At this time, since the pipe expansion rate is information necessary for controlling the pipe expanding device, the pipe expansion rate can be specified by the set value set by a host computer 140. As the operational parameter of the pipe expanding step, the number of pipe expanding dies and the pipe expanding die diameter may be used in addition to the pipe expansion rate.

[Out-of-Roundness Prediction Method]

In the present embodiment, the out-of-roundness prediction model M generated offline by the out-of-roundness prediction model generation unit 130 as described above is used to perform online prediction of the out-of-roundness of the steel pipe after the pipe expanding step. An initial procedure of the out-of-roundness prediction of the steel pipe after the pipe expanding step, an operational condition dataset to be set as an operational condition of the steel pipe manufacturing process is acquired online (operational parameter acquisition step). This is a step of acquiring necessary data from the host computer that performs overall control of the steel pipe manufacturing process or from an individual forming processing step control computer, as the operational condition dataset to be the input to the out-of-roundness prediction model generated as described above. Here, the “online” means duration of a series of manufacturing steps from the timing before the start of the manufacturing process of the steel pipe to the completion of the pipe expanding step. Therefore, the processing or working does not necessarily have to be in execution in any of the forming processing steps. The period of waiting for conveying the steel sheet to the next step between each forming processing step is also included in “online”. In addition, the period before the start of the manufacturing process of the steel pipe and after the completion of the thick sheet rolling step of manufacturing the steel sheet as a material can also be included in “online”. This is because when the thick sheet rolling step of manufacturing a steel sheet to be a material is completed, an operational condition dataset to be an input to the out-of-roundness prediction model of the present embodiment can be acquired. The model used online is the out-of-roundness prediction model M trained in machine learning, and after setting the operational parameter to be the input condition, it is possible to immediately calculate the out-of-roundness to be output and promptly perform reconfiguration of operational conditions, and the like.

The steel pipe out-of-roundness prediction after the pipe expanding step can be performed at any timing before or during the manufacturing process of the steel sheet. The operational condition dataset to be the input to the out-of-roundness prediction model M is appropriately generated according to the timing of performing the prediction. That is, when the steel pipe out-of-roundness prediction after the pipe expanding step is performed before the end bending step, it is possible to use performance values (actual measurement values) for the attribute information of the steel sheet to be a material, and the set value of the operational condition preset in the host computer is used as the operational parameter of the subsequent manufacturing step including the end bending step.

In addition, when the end bending step is finished and the out-of-roundness prediction of the steel pipe after the pipe expanding step is performed before the start of the press bending step, the performance values (actual measurement values) for the attribute information of the steel sheet to be the material and the performance values of the operational parameters of the end bending step are to be used, and the set values of the operational condition preset by the host computer is to be used as the operational parameters of the subsequent manufacturing step including the press bending step. Note that the preset set values of the operational conditions are set values that have been set based on the past operational performance and stored in the host computer in advance.

As described above, the present embodiment uses a set of operational condition datasets, acquired according to the timing of predicting the out-of-roundness of the steel pipe after the pipe expanding step, for the input to the out-of-roundness prediction model, and the out-of-roundness of the steel pipe after the pipe expanding step, which is the output, is predicted online. This makes it possible to perform reconfiguration of the operational conditions of the subsequent manufacturing step according to the predicted out-of-roundness of the steel pipe, leading to further reduction of the out-of-roundness of the steel pipe after the pipe expanding step.

[Out-of-Roundness Control Method]

Next, an out-of-roundness control method according to an embodiment of the present invention will be described with reference to Table 1 and FIG. 12.

In the present embodiment, first, a reconfiguration target step is selected from a plurality of forming processing steps constituting the manufacturing process of the steel pipe. Then, before the start of the reconfiguration target step, the out-of-roundness of the steel pipe after the pipe expanding step is predicted using the out-of-roundness prediction model M. Subsequently, reconfiguration is performed on one or more operational parameters selected from at least the operational parameters of the reconfiguration target step or one or more operational parameters selected from the operational parameters of the forming processing step on the downstream side of the reconfiguration target step so as to reduce the out-of-roundness of the steel pipe after the pipe expanding step.

Here, the plurality of forming processing steps constituting the steel pipe manufacturing process refers to the end bending step, the press bending step, and the pipe expanding step, which are the steps of applying plastic deformation onto the steel sheet to form the steel pipe into a predetermined shape. For the reconfiguration target step, a certain step is selected from these forming processing steps. Before execution of the forming processing in the selected reconfiguration target step, the out-of-roundness of the steel pipe after the pipe expanding step is predicted using the out-of-roundness prediction model M for the steel pipe. At this time, the forming of the steel sheet has been completed in the forming step on the upstream side of the reconfiguration target step, and thus, when the operational parameters of the forming processing step on the upstream side are to be used, the performance data on operational parameters can be used for the input to the out-of-roundness prediction model M. In contrast, the operational performance data cannot be collected in the downstream forming processing steps including the reconfiguration target step, and thus, the set value preset in the host computer or the like is used for the input to the out-of-roundness prediction model M for the steel pipe. In this manner, the out-of-roundness of the steel pipe after the pipe expanding step for the target material can be predicted.

Subsequently, determination is made as to whether the out-of-roundness predicted as the out-of-roundness of the steel pipe after the pipe expanding step falls within the out-of-roundness allowed as a product. With this operation, when making the out-of-roundness of the steel pipe after the pipe expanding step smaller than the predicted value, it is possible to reconfigure the operational conditions in the reconfiguration target step and the forming processing step on the downstream side of the reconfiguration target step. Here, the operational parameter to be reconfigured may be an operational parameter in the reconfiguration target step or an operational parameter in the forming processing steps downstream of the reconfiguration target step. An operational parameter of the forming processing step suitable for changing the out-of-roundness of the steel pipe after the pipe expanding step will preferably be selected according to a difference between the predicted out-of-roundness and the out-of-roundness allowed as a product. In addition, it is allowable to reconfigure both the operational parameters in the reconfiguration target step and the operational parameters in a certain forming processing step downstream of the reconfiguration target step. This is because the out-of-roundness of the steel pipe after the pipe expanding step can be effectively changed when there is a large difference between the predicted out-of-roundness and the out-of-roundness allowed as a product.

Table 1 specifically illustrates a case of the forming processing step selected as the reconfiguration target step and the forming processing step in which the operational parameters can be reconfigured correspondingly. Case 1 selects the end bending step as the reconfiguration target step in the steel pipe manufacturing process including the end bending step. At this time, before the start of the end bending step, the out-of-roundness of the steel pipe after the pipe expanding step is predicted using the set values of the operational parameters in the forming processing step including the press bending step. When the predicted out-of-roundness is large, it is possible to reconfigure certain operational parameters in each forming processing step of the end bending step, the press bending step, and the pipe expanding step. The operational parameters to be reconfigured are not limited to the operational parameters of the end bending step and may also be the operational parameters of other forming processing steps. When the attribute information of the steel sheet is included as the input to the out-of-roundness prediction model M, the performance data including the measured value and the like related to the attribute information of the steel sheet can be used for the input before starting the end bending step being the reconfiguration target step.

In Case 2, it is possible to select the reconfiguration target step and the operational parameters to be reconfigured according to the concept similar to Case 1. Case 3 is a case where the pipe expanding step is set as the reconfiguration target step. At this time, before the start of the pipe expanding step, the out-of-roundness of the steel pipe after the pipe expanding step is predicted using the out-of-roundness prediction model M. In this case, at least operational performance data in the end bending step and the press bending step can be used for the input to the out-of-roundness prediction model M. In addition, it is also allowable to use performance data of the attribute information of the steel sheet. In this manner, the predicted out-of-roundness of the steel pipe after the pipe expanding step is compared with the out-of-roundness allowed as a product, and when reducing the out-of-roundness, the operational parameter in the pipe expanding step is reconfigured. It is preferable to use the pipe expansion rate as the operational parameter of the pipe expanding step to be reconfigured. Note that the amount of change from the initially set value of the pipe expansion rate to be reconfigured may be set based on findings from experience. However, when the input of the out-of-roundness prediction model M includes the pipe expansion rate of the pipe expanding step, the out-of-roundness of the steel pipe after the pipe expanding step may be predicted again using the reconfigured value of the pipe expansion rate as the input to the out-of-roundness prediction model M, and the appropriateness of the conditions for reconfiguration may be determined.

TABLE 1

End
Press
Pipe

Reconfiguration
Bending
Bending
Expanding

Case
Target Step
Step
Step
Step

1
End Bending Step
◯
◯
◯

2
Press Bending Step
—
◯
◯

3
Pipe Expanding Step
—
—
◯

◯: Forming processing step in which it is possible to reconfigure operational parameters

Here, the steel pipe out-of-roundness control method according to an embodiment of the present invention will be described with reference to FIG. 12. The example illustrated in FIG. 12 is a case where the press bending step is selected as the reconfiguration target step, in which an end C-shaped formed body is transferred to the press bending step after completion of the end bending step. At this time, the operational performance data in the end bending step is transferred to an operational condition reconfiguration unit 150. The operational performance data may be transferred from a control computer provided in each step of controlling each forming processing step via a network. However, the data may be once transferred from the control computer of each forming processing step to the host computer 140 that performs overall control of the steel pipe manufacturing process, and then transferred from the host computer 140 to the operational condition reconfiguration unit 150. In addition, the performance data regarding the attribute information of the steel sheet is transferred from the host computer 140 to the operational condition reconfiguration unit 150 as necessary. The set values of the operational parameters of the press bending step and the pipe expanding step, which are forming processing steps downstream of the reconfiguration target step and the reconfiguration target step, are transferred from the control computer for each step to the operational condition reconfiguration unit 150. However, when the set values of the operational parameters in the press bending step and the pipe expanding step are stored in the host computer 140, the set values may be transferred from the host computer 140 to the operational condition reconfiguration unit 150. Note that the out-of-roundness target value determined according to the specification of the steel pipe as a product is transferred from the host computer 140 to the operational condition reconfiguration unit 150.

The operational condition reconfiguration unit 150 predicts the out-of-roundness of the steel pipe after the pipe expanding step based on these pieces of information by using the out-of-roundness prediction model M online, and compares the predicted out-of-roundness (out-of-roundness predicted value) with the target out-of-roundness (out-of-roundness target value). When the out-of-roundness predicted value is smaller than the out-of-roundness target value, the operational condition reconfiguration unit 150 determines the operational conditions of the remaining forming processing steps without changing the set values of the operational conditions of the press bending step, and the pipe expanding step, and then manufactures the steel pipe. On the contrary, when the predicted out-of-roundness is larger than the out-of-roundness target value, the operational condition reconfiguration unit 150 reconfigures the operational condition of the press bending step or the operational condition of the pipe expanding step. Specifically, it is possible to reconfigure the press depression amount, the number of times of pressing, and the like in the press bending step. The number of times of pressing in the press bending step may be increased once or twice or more, and a lower die interval ΔD may be reconfigured. In addition, the pipe expansion rate in the pipe expanding step can be reconfigured. Furthermore, both the press depression amount and the pipe expansion rate in the press bending step can be reconfigured.

The operational condition reconfiguration unit 150 may perform the out-of-roundness prediction again using the operational parameters reconfigured in this manner as the input data to the out-of-roundness prediction model M, confirm whether the predicted out-of-roundness is smaller than the out-of-roundness target value, and determine the reconfigured value of the operational conditions of the press bending step and the pipe expanding step. The reconfigured operational conditions of the press bending step and the pipe expanding step are transmitted to the individual control computers, so as to be defined as the operational conditions of the press bending step and the pipe expanding step. By repeatedly performing the out-of-roundness determination in the operational condition reconfiguration unit 150 a plurality of times, appropriate operational conditions of the press bending step and the pipe expanding step can be set even when a strict out-of-roundness target value is set, making it possible to manufacture a steel pipe with further improvement in the out-of-roundness. Furthermore, it is also allowable to perform steps in which the out-of-roundness control of the steel pipe after the pipe expanding step having the press bending step defined as the reconfiguration target step is executed in this manner, the out-of-roundness control of the steel pipe after the pipe expanding step having the pipe expanding step defined as the reconfiguration target step is executed again for the steel pipe that has undergone forming processing and be welded into the open pipe. This is because the steel pipe out-of-roundness prediction accuracy is further improved due to the state where the operational performance data of the press bending step has been obtained.

As described above, the steel pipe out-of-roundness control method according to an embodiment of the present invention uses the out-of-roundness prediction model M considering the impact on the out-of-roundness due to the interaction between the end bending step and the press bending step, making it possible to set an appropriate operational condition for improving the out-of-roundness of the steel pipe after the pipe expanding step, leading to achievement of manufacture of a steel pipe having satisfactory out-of-roundness. In addition, it is possible to realize highly accurate out-of-roundness control reflecting variation in attribute information of a steel sheet to be a material.

Next, a steel pipe out-of-roundness prediction device according to an embodiment of the present invention will be described with reference to FIG. 13.

FIG. 13 is a diagram illustrating a configuration of a steel pipe out-of-roundness prediction device according to an embodiment of the present invention. As illustrated in FIG. 13, a steel pipe out-of-roundness prediction device 160 according to an embodiment of the present invention includes an operational parameter acquisition unit 161, a storage unit 162, an out-of-roundness prediction unit 163, and an output unit 164.

The operational parameter acquisition unit 161 includes a certain interface capable of acquiring the out-of-roundness prediction model M generated by the machine learning unit from the out-of-roundness prediction model generation unit 130, for example. For example, the operational parameter acquisition unit 161 preferably includes a communication interface for acquiring the out-of-roundness prediction model M from the out-of-roundness prediction model generation unit 130. In this case, the operational parameter acquisition unit 161 may receive the out-of-roundness prediction model M from a machine learning unit 100b using a predetermined communication protocol. In addition, the operational parameter acquisition unit 161 acquires operational conditions for a forming processing facility (facility for executing the forming processing step) from a control computer or a host computer installed in the facility used in each forming processing step. For example, the operational parameter acquisition unit 161 preferably includes a communication interface for acquiring operational conditions. In addition, the operational parameter acquisition unit 161 may acquire input information based on a user's operation. In this case, the steel pipe out-of-roundness prediction device 160 further includes an input unit including one or more input interfaces that detect a user input and acquire input information based on the user's operation. Examples of the input unit include, but are not limited to, a physical key, a capacitive key, a touch screen integrated with a display of the output unit, a microphone for voice input, and the like. For example, the input unit receives an input of the operational condition for the out-of-roundness prediction model M acquired from the out-of-roundness prediction model generation unit 130 by the operational parameter acquisition unit 161.

The storage unit 162 includes at least one semiconductor memory device, at least one magnetic memory device, at least one optical memory device, or a combination of at least two of these. The storage unit 162 functions as, for example, a main storage device, an auxiliary storage device, or cache memory. The storage unit 162 stores any kind of information used for the operation of the steel pipe out-of-roundness prediction device 160. The storage unit 162 stores, for example, the out-of-roundness prediction model M acquired from the out-of-roundness prediction model generation unit 130 by the operational parameter acquisition unit 161, the operational condition acquired from the host computer by the operational parameter acquisition unit 161, and the out-of-roundness information predicted by the steel pipe out-of-roundness prediction device 160. The storage unit 162 may store a system program, an application program, and the like.

The out-of-roundness prediction unit 163 includes one or more processors. In the present embodiment, the processor includes, but not limited to, a general-purpose processor or a dedicated processor specialized for specific processing. The out-of-roundness prediction unit 163 is communicably connected to individual components constituting the steel pipe out-of-roundness prediction device 160, and controls the operation of the entire steel pipe out-of-roundness prediction device 160. The out-of-roundness prediction unit 163 may be any general-purpose electronic device such as a personal computer (PC) or a smartphone. The out-of-roundness prediction unit 163 is not limited thereto, and may be one server device or a plurality of server devices capable of communicating with each other, or may be another electronic device dedicated to the steel pipe out-of-roundness prediction device 160. The out-of-roundness prediction unit 163 calculates a predicted value of the steel pipe out-of-roundness information using the operational conditions acquired through the operational parameter acquisition unit 161 and the out-of-roundness prediction model M acquired from the out-of-roundness prediction model generation unit 130.

The output unit 164 outputs the predicted value of the out-of-roundness information of the steel pipe calculated by the out-of-roundness prediction unit 163 to a device for setting the operational conditions of the forming processing facility. The output unit 164 may include one or more output interfaces that output information and notify the user. The output interface is a display, for example. Examples of the display include an LCD or an organic EL display. The output unit 164 outputs data obtained by the operation of the steel pipe out-of-roundness prediction device 160. The output unit 164 may be connected to the steel pipe out-of-roundness prediction device 160 as an external output device instead of being provided in the steel pipe out-of-roundness prediction device 160. As the connection method, for example, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used. Examples of the output unit 164 include, but are not limited to, a display that outputs information by video, a speaker that outputs information by audio, and the like. For example, the output unit 164 presents the predicted value of the out-of-roundness information calculated by the out-of-roundness prediction unit 163 to the user. The user can appropriately set the operational conditions of the forming processing facility based on the predicted value of the out-of-roundness presented by the output unit 164.

A more preferable form of the steel pipe out-of-roundness prediction device 160 after the pipe expanding step as described above is a terminal device such as a tablet terminal including: an input unit 165 that acquires input information based on a user's operation; and a display unit 166 that displays a predicted value of the out-of-roundness information calculated by the out-of-roundness prediction unit 163. This type of device has a function of acquiring input information based on a user's operation from the input unit 165, and updating a part or all of the operational parameters of the forming processing facility already input to the steel pipe out-of-roundness prediction device 160, by using the acquired input information. That is, when the out-of-roundness information of the steel pipe has been predicted by the out-of-roundness prediction unit 163 for the steel sheet being processed in the forming processing facility, it is possible to provide a function of receiving an operation, made by an operator using the terminal device, of performing a correction input of a part of the operational parameters of the forming processing facility already input to the operational parameter acquisition unit 161. At this time, the operational parameter acquisition unit 161 holds the initial input data for the operational parameters to which the correction input has not been made from the terminal device among the operational parameters of the forming processing facility, and changes only the operational parameter to which the correction input has been made. With this configuration, new input data to the out-of-roundness prediction model M is generated in the operational parameter acquisition unit 161, and a predicted value of the out-of-roundness information based on the input data is calculated by the out-of-roundness prediction unit 163. Furthermore, the calculated predicted value of the out-of-roundness information is displayed on the display unit 166 of the terminal device through the output unit 164. With this procedure, the person such as an operator of the forming processing facility and a factory manager can immediately confirm the predicted value of the out-of-roundness information when the operational parameter of the forming processing facility has been changed, and can promptly change the operational condition to an appropriate operational condition.

EXAMPLE
Example 1

The present Example used a line pipe steel sheet (API grade X60) having a sheet thickness of 38.0 mm to 38.4 mm and a sheet width of 2700 mm to 2720 mm and generated an out-of-roundness prediction model after an offline pipe expanding step, in accordance with manufacturing conditions for manufacturing a steel pipe having a diameter of 36 inches after a pipe expanding step through an end bending step, a press bending step, a welding step, and a pipe expanding step. An example of a finite element model generated by a finite element model generation unit in the end bending step used in the present Example is illustrated in FIG. 14. The finite element analysis solver used was Abaqus 2019, and the calculation time per case was approximately three hours. The number of datasets accumulated in the database was 300, and Gaussian process regression using a radial basis function as a basis function was used as a machine learning model.

The attribute information of the steel sheet is selected as items of a representative sheet thickness (in-plane average sheet thickness), sheet width, and yield stress of the steel sheet. The range of variation as the operational condition was specified from the manufacturing performance record, and the input data of the calculation was changed within the range. As an operational parameter of the end bending step, the end bending processing width was selected. As operational conditions for the end bending step, upper and lower molds were used, specifically, an upper mold having a curvature radius of the forming surface of R300 mm and a lower mold having a curvature radius of the pressing surface of R300 mm. As an operational parameter of the end bending step in the operational condition dataset, the end bending processing width was changed in a range of 180 mm to 240 mm. As the operational parameters of the press bending step, the number of times of press depression and the press depression position were selected. At this time, the number of times of press depression was changed within a range of 7 to 15 times with 11 times as a reference condition. Regarding the press depression position, the press depression was performed at equal intervals in the sheet width direction according to the number of times of pressing, and the press depression position was determined according to the number of times of pressing. Based on the press depression amount, defined as an amount by which the punch leading end reaches a position of 15.8 mm from a line connecting the uppermost portions of the rod-shaped member, a bending angle for one press depression was set to 30°.

Next, the steel sheet was placed on a die having an interval between the rod-shaped members set to 450 mm, and press depression was started with a position 1120 mm away from the central portion in the width direction of the steel sheet set as a reference, by a punch having a processing surface with a radius of 308 mm. When the number of times of press depression was 11, press depression was performed five times from the right side of FIG. 4 toward the central portion in the width direction under the condition of a sheet feed pitch of 224 mm. Thereafter, the end of the sheet on the left side of FIG. 4 was moved to the vicinity of the rod-shaped member, and press depression was performed six times for the left half of the steel sheet beginning at the position of 1120 mm from the end under the condition of a sheet feed pitch of 224 mm. In addition, a constant value of 1.0% was used as the pipe expansion rate which is an operational parameter of the pipe expanding step.

In the present Example, the analysis conditions as described above were set in the out-of-roundness offline calculation unit, the analysis conditions were changed within the range of the above operational conditions, and a calculation result of the out-of-roundness after the pipe expanding step obtained by the analysis was accumulated in a database. Next, an out-of-roundness prediction model was generated based on the accumulated database. In the present Example, the out-of-roundness prediction model generated in this manner is applied online. The out-of-roundness in the present Example was defined such as out-of-roundness=Dmax−Dmin when outer diameters at positions facing each other were selected by dividing the tube into 3600 equal parts in the circumferential direction, and a maximum diameter and a minimum diameter among them were defined as Dmax and Dmin, respectively.

In the online step, before the start of the end bending step, a representative sheet thickness and a sheet width of the steel sheet were acquired from the host computer as performance data of attribute information of the steel sheet to be a material. In addition, test data of the yield stress obtained in the inspection step of the thick sheet rolling step was acquired. Along with this, set values of the operational conditions of the end bending step and the press bending step were acquired from the host computer. In the manufacturing process of the steel pipe targeted in the present Example, the end bending processing width in the end bending step was 200 mm as the set value of the operational condition that has been preset by the host computer. On the other hand, the number of times of pressing in the press bending step was eleven, and the press depression position was set at a 224 mm pitch in the width direction of the steel sheet with a position 1120 mm away from the central portion in the width direction of the steel sheet defined as a first press position. In addition, the press depression amount at each press depression position was a preset value under the condition of 15.8 mm.

In the present Example, before the start of the end bending step, the out-of-roundness of the steel pipe after the pipe expanding step was predicted using these set values together with the representative sheet thickness and the sheet width, which are the performance data of the attribute information of the steel sheet, as an input to the out-of-roundness prediction model. On the other hand, in the host computer, the out-of-roundness target value was set to 10 mm, the predicted steel pipe out-of-roundness (out-of-roundness predicted value) is compared with the out-of-roundness target value, and when the predicted out-of-roundness exceeded the out-of-roundness target value, the operational condition of the press bending step was reconfigured. The number of times of pressing was selected as the operational condition to be reconfigured. As a result, in the invention example, it was confirmed that the average value of the out-of-roundness was 4.0 mm, and the acceptance rate was 100%. In comparison, in a case where, as a comparative example, manufacture was conducted while the operational condition of the press bending step was set to the set value preset by the host computer, the average value of the out-of-roundness was 11.2 mm, and the acceptance rate was 80%.

INDUSTRIAL APPLICABILITY

According to the present invention, it is possible to provide a steel pipe out-of-roundness prediction model generation method capable of generating an out-of-roundness prediction model that accurately and promptly predicts the out-of-roundness of a steel pipe after a pipe expanding step in a steel pipe manufacturing process including a plurality of steps. In addition, according to the present invention, it is possible to provide a steel pipe out-of-roundness prediction method and a steel pipe out-of-roundness prediction device capable of accurately and promptly predicting the out-of-roundness of the steel pipe after the pipe expanding step in the steel pipe manufacturing process including a plurality of steps. In addition, according to the present invention, it is possible to provide a steel pipe out-of-roundness control method capable of accurately controlling the out-of-roundness of the steel pipe after the pipe expanding step in the steel pipe manufacturing process including a plurality of steps. Furthermore, according to the present invention, it is possible to provide a steel pipe manufacturing method capable of manufacturing a steel pipe having a desired out-of-roundness with good yield.

REFERENCE SIGNS LIST

- 1 DIE
- 1
  a, 1b ROD-SHAPED MEMBER
- 2 PUNCH
- 2
  a PUNCH LEADING END
- 2
  b PUNCH SUPPORT
- 16 PIPE EXPANDING DIE
- 17 TAPERED OUTER CIRCUMFERENTIAL SURFACE
- 18 PULL ROD
- 20 ARM
- 21
  a, 21b DISPLACEMENT METER
- 22 ROTATION ANGLE DETECTOR
- 25 ROTATING ARM
- 26
  a, 26b PRESSING ROLLER
- 30 C-PRESS DEVICE
- 31 CONVEYANCE MECHANISM
- 31
  a CONVEYANCE ROLL
- 32A, 32B PRESS MECHANISM
- 33 UPPER MOLD
- 33
  a FORMING SURFACE
- 34 LOWER MOLD
- 34
  a PRESSING SURFACE
- 36 HYDRAULIC CYLINDER
- 37 CLAMP MECHANISM
- 110 BASIC DATA ACQUISITION UNIT
- 111 OPERATIONAL CONDITION DATASET
- 112 OUT-OF-ROUNDNESS OFFLINE CALCULATION UNIT
- 112
  a FINITE ELEMENT MODEL GENERATION UNIT IN END BENDING STEP
- 112
  b FINITE ELEMENT MODEL GENERATION UNIT IN PRESS BENDING STEP
- 112
  c FINITE ELEMENT MODEL GENERATION UNIT IN PIPE EXPANDING STEP
- 112
  d FINITE ELEMENT ANALYSIS SOLVER
- 120 DATABASE
- 130 OUT-OF-ROUNDNESS PREDICTION MODEL GENERATION UNIT
- 140 HOST COMPUTER
- 150 OPERATIONAL CONDITION RECONFIGURATION UNIT
- 160 STEEL PIPE OUT-OF-ROUNDNESS PREDICTION DEVICE
- 161 OPERATIONAL PARAMETER ACQUISITION UNIT
- 162 STORAGE UNIT
- 163 OUT-OF-ROUNDNESS PREDICTION UNIT
- 164 OUTPUT UNIT
- 165 INPUT UNIT
- 166 DISPLAY UNIT
- G SEAM GAP PORTION
- M OUT-OF-ROUNDNESS PREDICTION MODEL
- P STEEL PIPE
- R1, R2 REGION
- S STEEL SHEET
- S₁FORMED BODY

STEEL PIPE OUT-OF-ROUNDNESS PREDICTION MODEL GENERATION METHOD, STEEL PIPE OUT-OF-ROUNDNESS PREDICTION METHOD, STEEL PIPE OUT-OF-ROUNDNESS CONTROL METHOD, STEEL PIPE MANUFACTURING METHOD, AND STEEL PIPE OUT-OF-ROUNDNESS PREDICTION DEVICE

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

Priority Claims (1)

PCT Information