RU2804572C1 - Method for generation of steel pipe ovality prediction model, steel pipe ovality prediction method, steel pipe ovality control method, steel pipe manufacturing method and device for steel pipe ovality prediction - Google Patents

Abstract

FIELD: pressure treatment.

SUBSTANCE: invention can be used in the manufacture of steel pipes. Pipe manufacturing includes a steel sheet bending step, a welding gap reduction step, a welding step, and a pipe expansion step. The method for generating a pipe out-of-roundness prediction model includes performing a numerical operation having a set of operating conditions data as an input and an out-of-roundness of a steel pipe after the expansion step as an output. The operation is performed multiple times with varying sets of operating conditions. As a result, many sets of ovality data of the steel pipe after the expansion step are generated offline as training data. An ovality prediction model is generated, having a set of operating conditions data as input data and the ovality of the steel pipe after the expansion stage as output data. Model generation is performed offline using machine learning using training datasets.

EFFECT: it is possible to obtain a steel pipe with the required ovality.

12 cl, 17 dwg, 1 tbl, 3 ex

Description

Field of technology to which the invention relates

The present invention relates to a method for generating an ovality prediction model for a steel pipe, which is a method for generating an ovality prediction model that predicts the ovality of a steel pipe after the pipe expansion step in the steel pipe manufacturing step using a pressure bending method, a method for predicting the ovality of a steel pipe, a method for controlling the ovality of a steel pipe. pipe, a method for manufacturing a steel pipe and a device for predicting the ovality of a steel pipe.

State of the art

Methods for manufacturing steel pipes having a large diameter and large thickness used in pipelines and the like include the widely used method of manufacturing a steel pipe (called UOE teak steel pipe) by stamping a steel sheet having a predetermined length, width and thickness, to shape it into a U-shape, forming the sheet to make it into an O-shape, and welding the joint portion to shape the sheet into a pipe shape and expanding the diameter of the steel pipe (called pipe expansion) to improve the ovality. However, the manufacturing step of UOE type steel pipe requires a large pressure force in the step of stamping the steel sheet into U-shape and O-shape, which necessarily requires the use of large-scale pressing equipment.

To solve this problem, there is a proposed method for producing a steel pipe having a large diameter and a large wall thickness, which is a forming method with reduced forming pressure. Specifically, the method put into practical use is a method in which, after bending the ends of a steel sheet in the width direction (referred to as end bending), a three-point pressure bending step by indentation is performed many times to obtain a molded workpiece having a U-shaped cross-section (hereinafter referred to as U-shaped preform), followed by the step of reducing the weld gap in the weld gap portion of the U-shaped preform to obtain an open pipe, welding the joint portions to obtain a steel pipe, and finally inserting the device into the steel pipe for pipe expansion to expand the inner diameter of a steel pipe. The pipe expansion device used is a device that contains a plurality of pipe expansion tools, each of which has a curved surface obtained by dividing an arc into a plurality of portions, and brings the curved surface of the pipe expansion tool into contact with the inner surface of the steel pipe to expand the steel pipe. and fixing the shape of the steel pipe.

In the pressure bending stage, increasing the number of pressings when performing three-point bending reduces the ovality of the steel pipe after the pipe expansion stage, although it takes a long time to form the steel pipe to obtain a U-shaped section. On the other hand, reducing the number of presses when performing three-point bending would create a problem in that the cross-section of the steel pipe would be essentially polygonal in shape, making it difficult to obtain a circular cross-section. To solve this problem, the number of presses when performing three-point bending (for example, 5 - 13 presses for a steel pipe with a diameter of 1200 mm) is determined empirically during operation depending on the size of the steel pipe. Regarding the operating conditions of the pressure bending step for reducing the ovality of a steel pipe after the pipe expansion step, many proposals have been made to develop a method for performing it.

For example, Patent Literature 1 describes a method for performing three-point flexible forming in a minimum amount, and a method for expanding a pipe by bringing a plurality of pipe expansion tools located in the circumferential direction of the pipe expanding apparatus into contact with an undeformed portion that has not been deformed by three-point forming. flexible.

Patent Literature 2 describes a method for reducing the ovality of a steel pipe after the pipe expansion step by setting the radius of curvature of the outer peripheral surface of the punch used in three-point bending forming and the radius of curvature of the outer peripheral surface of the pipe expansion tool to satisfy a predetermined relative ratio.

Patent Literature 3 describes a method that is a manufacturing method that can efficiently produce a steel pipe with proper ovality without the need to apply excessive pressure in the pressure bending step, in which, when performing three-point bending forming, an easily machined portion is provided that has very little curvature compared to others areas, or an untreated area in which bending processing is absent in at least part of the steel sheet. Patent Literature 3 also describes an operation in the weld gap reduction step, which is the application of a pressure force to a portion at a predetermined distance from the center of an easy-to-work portion or a non-machinable portion, without holding the easy-to-work portion or a non-machinable portion. In this regard, the O-shape forming device is used in the weld gap reduction step after the pressure bending step.

In this regard, Patent Literature 4 describes another method for reducing the weld gap (hereinafter referred to as the "locking forming method") in which a non-circular workpiece (U-section molded workpiece) is formed by three-point forming, and then instead of an O-forming step A pushing force is applied to the non-circular workpiece from the outside using a pushing tool located at the upper portion and facing the lower support rollers, while the workpiece is supported by two support rollers, resulting in a reduced welding gap. This method is characterized in that applying a pushing force by using a pushing tool on the outside of a non-circular workpiece can simplify the structure of the device by eliminating the need to prepare a die according to the outer diameter of the steel pipe as a device for O-shaping the workpiece. Patent Literature 4 also describes a method of providing a specially provided relatively easy-to-form area for molding a three-point bend and applying a pushing force in said area to a molded blank having a U-shaped cross-section in a closure molding method in a weld gap reduction step.

On the other hand, Non-Patent Literature 1 describes a method of performing a calculation analysis that uses the finite element method and the influence of the operating conditions of the pipe expansion step on the ovality of a steel pipe after the pipe expansion step.

List of opposed documents

Patent literature

Patent Literature 1: JP 2012-170977 A

Patent Literature 2: JP 5541432 B1

Patent Literature 3: JP 6015997 B1

Patent Literature 4: JP 2012-250285 A

Non-patent literature

Non-patent literature 1: Journal of the Japan Society for Technology of Plasticity, vol.59, no. 694 (2018), p. 203-208

The essence of the invention

Technical problem

The method described in Patent Literature 1 is a method for reducing the ovality of a steel pipe after the pipe expansion step by the relationship between the forming position of the three-point bending press and the forming position of the pipe expansion tool. However, the steel pipe manufacturing step includes a number of steps including at least a pressure bending step, a weld gap reduction step, a welding step, and a pipe expansion step. Therefore, the method described in Patent Literature 1 does not take into account the influence of operating conditions in other steps on the ovality of the steel pipe after the pipe expansion step, making it difficult to continuously ensure that the ovality of the steel pipe after the pipe expansion step is reduced.

Similar to the method described in Patent Literature 1, the method described in Patent Literature 2 is a method in which the radius of curvature is set such that the radius of curvature of the outer peripheral surface of the punch used in three-point bending molding is the operating condition of the step bending by pressure, and the radius of curvature of the outer peripheral surface of the pipe expansion tool as the working condition of the pipe expansion step satisfies the predetermined ratio, as a result, the ovality of the steel pipe after the pipe expansion step is reduced. However, in the method described in Patent Literature 2, similar to the method described in Patent Literature 1, there is a problem of not being able to take into account steps other than the pressure bending step, namely, the weld gap reduction step.

The method described in Patent Literature 3 is a method for reducing the ovality of a steel pipe after the pipe expansion step by changing the operating state of the three-point bending press in the pressure bending step depending on the position of the steel sheet and setting the operating state related to the forming condition in the gap reduction step. welding However, the method described in Patent Literature 3 has a problem that variations in the thickness of the sheet or material of the steel sheet cause the ovality of the steel pipe to fluctuate after the pipe expansion step even under the same forming condition.

The method described in Patent Literature 4 is also a method for reducing the ovality of a steel pipe after the pipe expansion step by setting conditions such that the forming condition of a U-shaped workpiece in the pressure bending step and the forming condition in the welding gap reduction step are related to each other. However, in the method described in Patent Literature 4, there is also a problem that variations in the thickness of the sheet or material of the steel sheet cause the ovality of the steel pipe to fluctuate after the pipe expansion step even under the same forming condition.

On the other hand, as described in the method of Non-Patent Literature 1, the effect of the operating parameter of the pipe expansion step on the ovality can be predicted quantitatively by performing an analysis of the pipe expansion step using the finite element method for offline calculation. However, the method described in Non-Patent Literature 1 also has the problem that it is not possible to take into account the influence of the operating conditions of other steps on the ovality. Moreover, performing such numerical analysis also involves the problem of long computation time, which makes it difficult to predict ovality in real time.

The present invention has been developed to solve the above problems, and its first object is to provide a steel pipe ovality prediction model generation method capable of generating an ovality prediction model that accurately and quickly predicts the ovality of a steel pipe after the pipe expansion step in the steel pipe manufacturing step. , which includes a number of stages. Another object of the present invention is to provide a steel pipe ovality prediction method and a steel pipe ovality prediction apparatus capable of accurately and quickly predicting the ovality of a steel pipe after a pipe expansion step in a steel pipe manufacturing step including a number of steps. Another object of the present invention is to provide a method for controlling the ovality of a steel pipe capable of accurately controlling the ovality of a steel pipe after a pipe expansion step in a steel pipe manufacturing step including a number of steps. Another object of the present invention is to provide a steel pipe manufacturing method capable of producing a steel pipe having a desired ovality with good yield.

Solution

The steel pipe ovality prediction model generating method of the present invention generates an ovality prediction model that predicts the ovality of a steel pipe after a pipe expansion step by expanding the inner diameter of a steel pipe whose ends are connected to each other in a steel pipe manufacturing process, which includes: a step pressure bending for processing a steel sheet to form a molded blank having a U-shaped cross-section by means of multiple pressings with a punch; a step of reducing the weld gap by reducing the weld gap portion of the molded blank having a U-shaped cross-section for forming an open pipe; a welding step for joining the ends of the open pipe to each other; and a pipe expansion step, and includes: an input data collection step, in which a numerical operation is performed having as input data a set of operating conditions data including one or more parameters selected from attribute information of the steel sheet, one or more parameters, selected from the operating parameters of the pressure bending step, and one or more parameters selected from the operating parameters of the weld gap reduction step, and having as an output the ovality of the steel pipe after the pipe expansion step, where the calculation is performed multiple times, changing the set of operating data conditions, and through this calculation, a plurality of ovality data sets of the steel pipe after the pipe expansion step corresponding to the operating condition data sets are generated offline as training data; and an ovality prediction model generation step, in which an ovality prediction model having the specified operating condition data set as input is generated, and the ovality of the steel pipe after the pipe expansion step is generated as the output, the model generation is performed offline by machine learning using a specified set of training data sets obtained during the initial data collection stage.

The input data collection step may include a finite element calculation step of the ovality of the steel pipe after the pipe expansion step using a set of operating conditions data.

The ovality prediction model may include one or more parameters selected from the operating parameters of the pipe expansion step as a set of operating conditions.

The steel pipe manufacturing step may include an end bending step by bending the ends of the steel sheet in the width direction before the pressure bending step, and the ovality prediction model may include one or more parameters selected from the operating parameters of the end bending step as a set of operating condition data. .

The operating parameter of the bending step may include: information about a forming position and a forming indentation amount related to an operation in which the punch used in the bending step forms the steel sheet; and the number of presses performed during the pressure bending phase.

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

The method for predicting the ovality of a steel pipe according to the present invention includes: an operating parameter collection step, in which a set of data of operating conditions to be set as the operating conditions of steps in the steel pipe manufacturing process is collected in real time as input data of an ovality prediction model. a steel pipe generated by the steel pipe ovality prediction model generation method of the present invention; and an ovality prediction step of predicting the ovality information of the steel pipe after the pipe expansion step by inputting the operating condition data set obtained from the operating parameter acquisition step into the ovality prediction model.

The method for controlling the ovality of a steel pipe according to the first aspect of the present invention includes: the step of collecting a set of operating condition data including an actual value from attribute information of the steel sheet, a target value of the operating parameter of the pressure bending step, and a target value of the operating parameter of the welding gap reduction step; predicting the ovality of the steel pipe after the pipe expansion step by inputting the obtained operating condition data set into the ovality prediction model, wherein the prediction is performed before starting the pressure bending step by using the steel pipe ovality prediction method of the present invention, and changing the set value of the operating parameter of the pressure bending step and/ or a given value of the operating parameter of the weld gap reduction stage in such a way as to reduce the predicted ovality.

The method for adjusting the ovality of a steel pipe according to the second aspect of the present invention includes: a step of predicting ovality information of a steel pipe after a pipe expansion step, wherein the prediction is performed before starting a target step for making changes selected from an end bending step, a pressure bending step, a gap reducing step. welding and pipe expansion step, which are included in the steel pipe manufacturing step, wherein the prediction is performed using the method of predicting the ovality of a steel pipe according to claim 7, and changing one or more operating parameters selected from at least the operating parameters of the intended step for making changes, or one or more operating parameters selected from the operating parameters of the forming processing steps performed after the intended step for making changes, wherein the change is made based on predicted information about the ovality of the steel pipe.

A process is provided for manufacturing a steel pipe using the steel pipe ovality control method of the present invention.

The steel pipe ovality predicting device of the present invention predicts the ovality of a steel pipe after a pipe expansion step by expanding the inner diameter of a steel pipe whose ends are connected to each other in a steel pipe manufacturing process, which includes: a pressure bending step for processing the steel sheet to form a molded blank having a U-shaped cross-section by means of a plurality of pressings with a punch; a step of reducing a weld gap by reducing a weld gap portion of a molded blank having a U-shaped cross-section for forming an open pipe; a welding step for joining the ends of the open pipe to each other; and a pipe expansion step, and contains: an input data collection unit for performing a numerical operation having as input a set of operating conditions data, including one or more parameters selected from attribute information about the steel sheet, one or more operating parameters selected of the operating parameters of the pressure bending step, and one or more operating parameters selected from the operating parameters of the weld gap reduction step, and having as output information about the ovality of the steel pipe after the pipe expansion step; wherein the calculation is performed a plurality of times with the operating condition data set changing, and by means of said calculation, a plurality of data sets are generated regarding the ovality information of the steel pipe after the pipe expansion step corresponding to the operating condition data sets as training data; an ovality prediction model generating unit configured to generate an ovality prediction model having as input a set of operating conditions data and having as output information about the ovality of a steel pipe after the pipe expansion step, wherein the model generation is performed using machine learning using a set of training data sets generated by the source data collection block; an operating parameter acquisition unit that obtains in real time a set of data of operating conditions to be set as operating conditions of steel pipe manufacturing steps; and an ovality prediction unit that predicts the ovality information of the steel pipe after the pipe expansion step corresponding to the operating condition data set obtained by the operating parameter acquisition unit, the prediction being performed in real time using the ovality prediction model generated by the ovality prediction model generating unit.

The steel pipe ovality prediction device may include a terminal device including an input unit that collects input information based on user operations, and an display unit that displays the ovality information, and an operating parameter collection unit may update part or all of the operating condition data during manufacturing steps. the steel pipe based on the input information received by the input unit, and the display unit can display the ovality information of the steel pipe predicted by the ovality prediction unit using the updated operating condition data set.

Technical results of the invention

According to the method of generating a steel pipe ovality prediction model of the present invention, it is possible to generate an ovality model of a steel pipe that accurately and quickly predicts the ovality of a steel pipe obtained after a pipe expansion step in a steel pipe manufacturing process including a number of steps. Moreover, according to the steel pipe ovality predicting method and the steel pipe ovality predicting apparatus of the present invention, it is possible to accurately and quickly predict the ovality of a steel pipe obtained after the pipe expansion step in a steel pipe manufacturing process including a number of steps. Moreover, according to the method for controlling the ovality of a steel pipe of the present invention, it is possible to accurately control the ovality of a steel pipe obtained after the pipe expansion step in a steel pipe manufacturing process including a number of steps. Moreover, according to the steel pipe manufacturing method of the present invention, it is possible to manufacture a steel pipe having the required ovality with good product yield.

Brief description of drawings

Fig. 1 is a diagram illustrating the manufacture of a steel pipe according to an embodiment of the present invention;

fig. 2 is a diagram illustrating an example of a step of forming a workpiece having a U-shaped cross-section using a pressure bending apparatus;

fig. 3 is a diagram illustrating an example of a step of forming a workpiece having a U-shaped cross-section using a pressure bending apparatus;

fig. 4 is a diagram illustrating an example of the design of a device for giving an O-shape to a workpiece;

fig. 5 is a diagram illustrating an example of a structure of a closure molding apparatus;

fig. 6 is a diagram illustrating a design example of a pipe expansion device;

fig. 7 is a diagram illustrating a design example of a device for measuring the shape of an outer diameter of a steel pipe;

fig. 8 is a block diagram illustrating the configuration of a steel pipe ovality prediction model generating apparatus according to an embodiment of the present invention;

fig. 9 is a block diagram illustrating the configuration of the offline ovality calculation apparatus shown in FIG. 8;

fig. 10 is a graph showing an example of the change in the relationship between the forming amount and the ovality of a steel pipe after the pipe expansion step according to the change in operating conditions in the pressure bending step;

fig. 11 is a table illustrating an example of a molding indentation position and a molding indentation amount for each press of a series of presses;

fig. 12 is a diagram illustrating the operation sequence of predicting the ovality of a steel pipe after pipe expansion, the prediction is performed before starting the pipe bending step;

fig. 13 is a diagram illustrating an example of a finite element model;

FIG. 14 is a perspective view illustrating the overall structure of the device for giving a C-shape to a workpiece; FIG.

fig. 15 is a sectional view illustrating the structure of the molding mechanism;

fig. 16 is a diagram illustrating a method for adjusting the ovality of a steel pipe according to an embodiment of the present invention;

fig. 17 is a diagram illustrating the configuration of a steel pipe ovality predicting apparatus according to an embodiment of the present invention.

Description of Embodiments of the Invention

Steel pipe manufacturing stage

In fig. 1 is a diagram illustrating a manufacturing step of a steel pipe according to an embodiment of the present invention. As shown in FIG. 1, in the steel pipe manufacturing step of the embodiment of the present invention, a thick steel sheet is used as the material, produced in the thick plate rolling step, which is a preprocessing step before the steel pipe manufacturing step. In this case, the thick steel sheet has a yield strength of 245 - 1050 MPa, a tensile strength of 415 - 1145 MPa, a thickness of 6.4 - 50.8 mm, a width of 1200 - 4500 mm and a length of 10 - 18 m. In addition, the end of the thick of the steel sheet in the width direction is pre-ground to obtain a beveled edge called bevel edge. This step is performed to prevent overheating of the corner portion of the outer surface of the sheet end 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 affect the outer diameter after forming the steel pipe, the width is adjusted to a predetermined range taking into account the change in deformation in subsequent steps.

The step of making a steel pipe sometimes includes a step of performing end bending by bending the ends of the steel sheet to width. The end bending step, carried out by the C-shape machine, processes the ends of the steel sheet by bending (also called hemming) along its width. The device for giving the workpiece a C-shape contains a pair of dies - upper and lower - and a pair of clamps - upper and lower - which hold the central section of the steel sheet along the width. Since the length of the die is less than the length of the steel sheet, the process of bending the ends is repeated while successively feeding the steel sheet in the longitudinal direction. This end bending process is performed on both ends of the steel sheet along the width. Since applying a bending moment to the widthwise ends in a three-point bending press is difficult, the end bending step is performed using the specified dies in advance. This makes it possible to reduce the ovality of the finished steel pipe. Here, examples of operating parameters for specifying machining conditions include: an end bending machining width, which is equal to the length at which the die comes into contact with the steel sheet from the end, along the width, of the steel sheet to the center, along the width, of the sheet; clamp gripping force; the amount of feeding of the steel sheet when the end bending process is repeated in the longitudinal direction of the steel sheet, the direction of feeding, the number of feedings, etc.

The subsequent pressure bending step is a processing step to obtain a U-section molded workpiece by performing three-point bending by repeatedly using the punch of the pressure bending device. The subsequent weld gap reducing step is a step of reducing the weld gap in a molded blank having a U-shaped cross-section by using a device for shaping the blank into an O-shape for forming an open pipe. It should be noted that the locking molding method described in Patent Literature 4 can be used instead of the apparatus for forming the workpiece into an O-shape. The subsequent welding step is the step of limiting the weld gap portion formed at the end of the open pipe so that the ends can contact each other, and connecting the ends to each other. At this stage, a steel pipe is made from the molded blank, having the ends connected to each other. The subsequent pipe expansion step is a step of using an expansion device containing a plurality of pipe expansion tools, each of which has a curved surface obtained by dividing an arc into a plurality of sections for expanding the steel pipe by bringing the curved surface of the pipe expansion tools into contact with the inner surface of the pipe. At the inspection stage, the steel pipe produced in this way is checked whether its quality, namely, material properties, appearance and dimensions, meets the established technical specifications, after which the steel pipe is shipped as a finished product. In the present embodiment, the inspection step includes a step of measuring the ovality of the steel pipe.

In the present embodiment of the invention, the end bending step, the pressure bending step, the weld gap reducing step, and the pipe expansion step of the sequence of manufacturing steps including forming the steel sheet into an open pipe and the pipe expansion step after welding are called “forming processing step.” These steps are general steps for adjusting the size and shape of a steel pipe through plastic deformation of the steel sheet. Below, with reference to the drawings, is a detailed description of the individual steps involved in the steel pipe manufacturing step.

End bending stage

Below with reference to FIG. 14 and 15 give a detailed description of a C-shape forming apparatus that performs end bending processing. In fig. 14 is a perspective view of the overall layout of the device for giving a C-shape to a workpiece. As shown in FIG. 14, the device 30 for giving the workpiece a C-shape includes: a transport mechanism 31 that transports the steel sheet S in the longitudinal direction, i.e. in the direction of transportation; a forming mechanism 32A that bends one end width Sc of the sheet to obtain a predetermined curvature, with the output side of the steel sheet S in the conveying direction being the front side; a forming mechanism 32B, which bends the other end, along the width, Sd of the sheet to obtain a predetermined curvature; and an internal adjustment mechanism (not shown) that adjusts the distance between the forming mechanisms 32A and 32B on the left and right of the width of the steel sheet S that is subjected to end bending processing. The conveying mechanism 31 includes a plurality of rotatable conveying rollers 31a located in front and behind the forming mechanisms 32A and 32B. It should be noted that the reference symbol Sa in the drawing denotes the leading end (front end in the longitudinal direction) of the steel sheet S.

In fig. 15(a) is a sectional view in the width direction of the forming mechanism 32A, which performs bending processing at one end Sc in the width direction of the steel sheet S, viewed from the transport direction of the steel sheet S from the entry (into the processing process) to the exit (from processing process). The forming mechanism 32A and the forming mechanism 32B are bilaterally symmetrical and have an identical layout. The forming mechanisms 32A/32B include an upper half-die 33 and a lower half-die 34 as a pair of half-dies facing each other in a vertical direction, and a hydraulic cylinder 36 as a half-die moving means that pushes up the lower half-die 34 together with the tool holder 35 (moves in direction of approaching the upper half-matrix 33) and clamps the half-matrix with a given compression force. The forming mechanisms 32A and 32B may include a clamping mechanism 37 that grips the steel sheet S from the inside, along the width of the sheet, of the upper half-die 33 and the lower half-die 34. The length of the steel sheet S in the longitudinal direction of the upper half-die 33 and the lower half-die 34 is less than the length of the steel sheet S. In this case, end bending processing is performed many times with periodic feeding of the steel sheet S in the longitudinal direction by the conveying mechanism 31 (see FIG. 14).

In the end bending step, the lower half-die 34 in contact with the surface of the steel sheet on the outer side, with respect to the bending direction, of the width ends Sc and Sd of the steel sheet to be processed by end-bending has a forming surface 34a facing the upper half-die 33. The upper half-die 33 has a forming surface 33a of a convex curved shape facing the forming surface 34a and a radius of curvature corresponding to the inner diameter of the steel pipe to be manufactured. The forming surface 34a has a concave curved surface approaching the upper half-die 33 from the outer side in width. However, although the forming surface 34a of the lower half-die 34 is in the shape of a concave curved surface, the forming surface may be any surface that approaches the upper half-die 33 as it extends outward in the width direction, and may be an inclined plane. The shape of the curved surface of the upper half-die 33 and the lower half-die 34 is properly calculated in accordance with the thickness, outer diameter of the steel pipe and the like. steel sheet S and can be selected and used according to the specified material.

In fig. 15(b) is a cross-sectional view along the width of the forming mechanism 32A in the same position as in FIG. 15(a) showing a state in which the lower die 34 is pushed upward by the hydraulic cylinder 36 and clamped. The lower half die 34 is pushed upward by the hydraulic cylinder 36, and the width end Sc of the steel sheet S is subjected to bending processing and formed according to the shape in the direction of the arcuate forming surface 33a of the upper half die 33. The width at which end bending is performed (end bending processing width ), varies depending on the width of the steel sheet S and is generally approximately 100 - 400 mm.

Pressure bending stage

In fig. 2 shows an example of a step of forming a moldable blank having a U-shaped cross-section using a pressure bending apparatus. In the drawing pos. 1 indicates a matrix located in the transportation path of the steel sheet S. The matrix 1 contains rod elements 1a and 1b, left and right, for supporting the steel sheet S at two places in the transportation direction, and the distance ΔD between these elements can vary depending on the size of the steel pipe to be formed. In addition, pos. 2 shows a punch that can move towards and away from the die 1. The punch 2 has: a punch front end 2a having a downward convex processing surface that directly contacts and applies pressure to the steel sheet S to give the steel sheet a concave shape; and a punch support 2b that is connected to the rear surface of the punch front end 2a and supports the punch front end 2a. The maximum width of the punch front end 2a is equal to the width (thickness) of the punch support 2b.

When performing the bending processing of a steel sheet S using a pressure bending apparatus having the above-described structure, the steel sheet S is placed on the die 1, and three-point bending is sequentially performed by the punch 2 on both sides of the steel sheet S along the width direction toward the center portion, as shown in Fig. 3, when periodically feeding the steel sheet S with a given feeding amount. In fig. 3 shows the forming step of the molding piece S ₁ shown in the right row (j), in which the steel sheet S, pre-processed by end bending, is processed as shown from top to bottom in the left row (first half of processing (a) - e)) and then is processed as shown from top to bottom in the center row (second half of processing (f) to i)) by operations including bending processing and feeding of the steel sheet S. In FIG. The 3 arrows next to the steel sheet S and punch 2 indicate the direction of movement of the steel sheet S and punch 2 at individual stages. The molded blank S ₁ having a U-shaped cross-section as a result of processing at this stage has a gap between the ends, called a “weld gap”.

Herein, examples of operating parameters for setting the operating conditions in the forming step include the number of presses, information about the forming location, the amount of indentation when forming, the distance between die elements, the curvature of the punch, and the like.

The number of presses refers to the total number of presses on the steel sheet in the width direction during three-point bending forming. A greater number of pressings ensures a smoother curved shape of the molded workpiece having a U-shaped cross-section, and a greater reduction in the ovality of the steel pipe after the pipe expansion stage.

The forming position information refers to the position in the width direction of the steel sheet to be formed by the punch. Specifically, said position may be determined by a distance from one end of the steel sheet in the width direction or a distance based on the center portion of the steel sheet in the width direction. The molding location information is preferably processed in the form of data associated with the number of presses (a sequence including the number of presses from the first to the Nth).

The amount of indentation during molding is the amount of movement of the punch 2 at each indentation position during molding. The molding indentation amount is defined as the amount by which the lower end surface of the punch front end 2a protrudes downward from a line connecting the points of the uppermost surfaces of the die 1 shown in FIG. 2. In this case, since the amount of movement of the punch front end 2a can be set individually for each press, it is preferable that the number of presses and the molding indentation amount are processed as related data. Thus, when the number of presses is N, the operating state in the bending step is determined by data sets 1 to N, where the number of presses, the forming position information, and the forming indentation amount constitute one set of data sets.

The reason why these data sets are used is that by partially changing the forming position and the displacement amount of the punch in the pressure bending step, the entire cross-sectional shape of the pipe in the open pipe state is changed, which affects the ovality of the steel pipe after the pipe expansion step. However, it is not necessary to use all N data sets as input variables to the ovality prediction model that will be described below. It is possible to use a method for selecting the condition that has a great influence on the ovality of the steel pipe after the pipe expansion step, and generating an ovality prediction model using the information of the forming position and the amount of indentation when forming at the beginning (first time) or end (Nth time) of the pressure bending step .

The gap in the bottom die is the distance between the left and right core members 1a and 1b shown in FIG. 2, and is the parameter denoted by ΔD in the drawing. Increasing the gap in the bottom die will change the local curvature of the steel sheet even at the same amount of indentation during forming and will affect the ovality of the steel pipe after the pipe expansion step. Therefore, it is preferable to use the bottom die gap set according to the size of the steel pipe to be formed as the operating parameter in the pressure bending step. In addition, in the case where the gap in the lower die changes with each press of the punch, the data associated with the number of presses can be used as operating parameters.

Punch curvature refers to the curvature of the front end of the punch used for forming. Increasing the punch curvature also increases the local curvature of the steel sheet during three-point bending forming, affecting the ovality of the steel pipe after the pipe expansion step. However, since it is difficult to change the curvature for each pressing when forming a single steel sheet, it is preferable to use the punch curvature set according to the size of the steel pipe to be formed as the operating parameter in the pressure bending step.

Weld gap reduction stage

The weld gap reducing step is a step of reducing the weld gap of the U-shaped cross-section molding formed in the pressure bending step, and applying bending force and compressive force to bring the ends of the U-shaped cross-section molding closer to each other. In this case, despite applying a bending force or a compressive force to the U-shaped workpiece, the weld gap would increase due to elastic springback when the load is removed. Thus, taking into account the phenomenon of springback, a significant bending force or compressive force is applied to cause deformation and change of the U-shaped section of the molded workpiece as a whole in the longitudinal direction.

In fig. Figure 4 shows an example of the design of a device for giving the workpiece an O-shape, used at the stage of reducing the welding gap. As shown in FIG. 4(a), through the use of the upper half-die 3 and the lower half-die 4, the O-shape device provides compressive deformation in the longitudinal direction of the molded workpiece S ₁ having a U-shaped cross-section. In this case, the surfaces of the upper half-matrix 3 and the lower half-matrix 4, which are in contact with the molded workpiece S ₁ having a U-shaped cross-section, are processed so that they have the shape of a curved surface. Due to the proximity of the upper half-matrix 3 and the lower half-matrix 4 to each other, the lower section of the molded blank S ₁ having a U-shaped cross-section is held along the curved surface of the lower half-matrix 4. The upper section of the molded blank S ₁ containing the ends absorbs the bending force and the compressive force, created by the upper semi-matrix 3, to bring the ends closer to each other along the curved surface of the upper semi-matrix 3.

This temporarily reduces the weld gap between the ends facing each other in the circumferential direction. Subsequently, after the pressure of the matrix is removed, the gap for welding expands as a result of springing, and the final value of the clearance in the area G of the gap for welding in the open pipe S ₂ is determined, as shown in Fig. 4(b). In this case, the O-forming indentation amount is a value that is obtained by subtracting from the specified outer diameter of the steel pipe the distance between the highest point of the inscribed surface of the upper half-die 3 and the lowest point of the inscribed surface of the lower half-die 4 at the time of compression of the die. This index is also referred to as O-forming weld gap reduction index, which uses the ratio of the O-forming indentation amount to the outer diameter of the steel pipe.

Here, examples of operating parameters for specifying operating conditions in the weld gap reduction step other than the O-forming indentation amount include the O-forming indentation position, R of the O-forming die, and the like.

The indentation position in O-forming is the angle formed by the vertical line and the line connecting the end of the weld gap portion of the molded blank S ₁ having a U-shaped cross-section and the position at the center of the molded blank S ₁ in width. In addition, R of the O-forming die refers to the curvature of the region of the upper half-die 3 and the lower half-die 4 contacting the molded workpiece S ₁ . The greater the amount of O-forming indentation in the O-shape apparatus, the greater the curvature of the formed blank S ₁ around the 3 o'clock and 9 o'clock positions, which reduces the ovality of the finished steel pipe.

On the other hand, when using the end-forming method instead of the O-shape forming device for the workpiece, the end-forming device shown in FIG. 5 is used as a device for forming an open pipe S ₂ from a molded blank S ₁ . As shown in FIG. 5, the end-forming apparatus includes lower tools 10a and 10b. The lower tools 10a and 10b are located at intervals from each other, and each of them contains a drive mechanism capable of reversing the direction of rotation. In addition, the lower tools 10a and 10b are supported by spring means 11a and 11b and the like. The upper tool 13, containing the punch 12, is positioned so that it faces the lower tools 10a and 10b. From the outside, using a punch 12, a pushing force is applied to the molded workpiece S ₁ having a U-shaped cross-section.

A molded blank S ₁ having a U-shaped cross-section is formed to form an open pipe S ₂ in two steps. In the first step, the forming position of the molded workpiece S ₁ is positioned by the rotary tools 10a and 10b so that the region R1, which is on the right side of the weld gap portion G and is the predetermined area for applying bending deformation, is positioned near the three o'clock position. , as schematically shown with a dash-dotted line. Subsequently, a pushing force is applied using the punch 12, and after applying the pushing force, the load of the punch 12 is removed. Then, in a similar manner to the first stage, the operation is performed at the second stage, in which the molding place is located in such a way that the area R2, which is on the left side from the weld gap section G and is a predetermined area for applying bending deformation, positioned near the nine o'clock position. Thereafter, a pushing force is applied by the punch 12, and after applying the pushing force, the punch is released, so that an open pipe S ₂ is obtained from the molded blank S ₁ .

The forming position in the first and second stages is indicated by the angle formed by the line connecting the center of the weld gap portion G and the center width position of the steel sheet (the angle formed by the dash-dotted line in Fig. 5 and the vertical line). The pushing force in the first and second stages refers to the forming force applied to the molded workpiece S ₁ by the punch 12.

Welding stage

Subsequently, the end surfaces of the section with a gap for welding the open pipe S ₂ are joined to each other and welded in a welding machine (connecting means) to obtain a steel pipe. The welding machine (connecting means) used is, for example, a device containing three types of welding machines, i.e. a welding machine for tack welding pipes, a welding machine for welding the inner surface of pipes and a machine for welding the outer surface of pipes. In these welding machines, the pipe tack welding machine continuously brings the surfaces to be joined by the forming rolls into close contact with each other at an appropriate relative position, and welds the entire length of the contact portion in the direction of the pipe axis. Next, the tack welded pipe is welded (submerged arc welding) on the inner surface side of the joint section using a welding machine for welding the inner surface of pipes, and then welded (submerged arc welding) on the outer surface of the joint section using a welding machine for welding outer surface of pipes.

Pipe expansion stage

A steel pipe having a welded weld gap portion is processed such that a pipe expansion device is inserted into the steel pipe to expand the diameter of the steel pipe (called pipe expansion). In fig. 6(a) - 6(c) show an example of the design of a pipe expansion device. As shown in FIG. 6(a), the pipe expansion device includes a plurality of pipe expansion segments 16 having curved surfaces obtained by dividing an arc into a plurality of arcs in the circumferential direction of the conical outer peripheral surface 17. When expanding a steel pipe, using the pipe expansion device as shown in fig. 6(b) and 6(c), the pipe expansion segment 16 is first aligned with the pipe expansion start position by moving the steel pipe P using the steel pipe moving device. Next, by retracting the draw rod 18 from the pipe expansion starting position, the first pipe expansion processing is performed.

This allows each of the pipe expansion segments 16, in sliding contact with the conical outer peripheral surface 17, to move radially by wedge action, expanding the steel pipe P. This reduces the level of irregularity in the section shape of the steel pipe P, making the section shape of the steel pipe closer to an ideal circular shape. form. Next, the draw rod 18 is extended to the pipe expansion starting position to return the pipe expansion segments 16 inward in the vertical axial direction using the release mechanism, and thereafter, the steel pipe P is further moved a distance corresponding to the pitch (axial length) of the pipe expansion segments 16 . The pipe expansion segments 16 are adjusted according to the new pipe expansion position, and then the above-described operation is repeated. This allows the first pipe expansion processing to be carried out along the entire length of the steel pipe P in accordance with the pitch of the pipe expansion segments 16.

Examples of operating parameters for determining the operating conditions of the pipe expansion step include the pipe expansion ratio, the number of pipe expansion segments, and the diameter of the pipe expansion segments. The pipe expansion ratio is the ratio of the difference between the outer diameter after pipe expansion and the outer diameter before pipe expansion to the outer diameter before pipe expansion. The outer diameter before pipe expansion and after pipe expansion can be calculated by measuring the circumference length of the steel pipe. The expansion ratio of the pipe can be adjusted by the stroke amount when the pipe expansion segments are expanded in the radial direction. The number of segments for pipe expansion is the number of segments that abut the steel pipe located in the circumferential direction when the pipe is expanded. The diameter of the pipe expansion segment refers to the curvature of the portion of each pipe expansion segment that abuts the steel pipe.

Along with these parameters, the pipe expansion ratio is a working parameter that can easily adjust the ovality after the pipe expansion stage. Increasing the pipe expansion ratio ensures the curvature of the contact area with the pipe expansion segment evenly over the entire circumference of the pipe according to the R of the pipe expansion segments, allowing for a reduction in ovality. The larger the number of pipe expansion segments, the larger the local curvature variation in the circumferential direction of the steel pipe can be eliminated, allowing the steel pipe to achieve satisfactory ovality after the pipe expansion step.

However, on the other hand, an excessively high pipe expansion ratio could reduce the compression yield strength of the finished steel pipe due to the Bauschinger effect. When a steel pipe is used as a pipeline pipe, high compressive stresses are applied in the circumferential direction of the pipe. Therefore, the steel pipe material must have a high compressive yield strength, and increasing the expansion ratio of the pipe more than necessary is unacceptable. Accordingly, the pipe expansion ratio under actual conditions is set so that the ovality of the steel pipe is within a predetermined value when the pipe expansion ratio is lower than the specified upper limit value of the pipe expansion ratio.

Ovality measurement step

In the inspection stage, which is the last step in the production of steel pipe, the quality of the steel pipe is checked and the ovality of the steel pipe is measured. Ovality, measured in the ovality measurement step, is an indicator representing the degree of deviation of the shape of the outer diameter of a steel pipe from an ideal circle. As a rule, the closer the ovality is to zero, the closer the cross-sectional shape of the steel pipe is to an ideal circle. Ovality is calculated based on the outer diameter of the steel pipe measured by an ovality measuring device. For example, a pipe is divided into equal parts in the circumferential direction at a certain location along the length of the pipe, and then the outer diameters are measured at opposite positions. When, among the specified diameters, the maximum diameter and the minimum diameter are defined as Dmax and Dmin, respectively, the ovality can be defined as Dmax - Dmin. The larger the number of divisions into equal parts, the smaller the irregularities in the shape of the steel pipe after the expansion step can be a numerical indicator, which is preferable. In particular, it is preferable to use the information obtained by dividing into equal parts in an amount of 4 - 36,000. It is more preferable to use division into 360 or more equal parts.

However, ovality need not be represented by the difference between the maximum diameter and the minimum diameter. It is also permissible to use a method of calculating an equivalent temporary ideal circle (diameter) having the same area as the area inside the curved figure representing the shape of the outer diameter of a steel pipe on a continuous diagram, and determining, based on the temporary ideal circle, the area deviating from the shape of the outer diameter of the steel pipe. pipes, in the form of an image. As a means of measuring the shape of the outer diameter of a steel pipe, for example, the following methods can be used.

(a) As shown in FIG. 7(a), by means of a device comprising a rod 20 that can be rotated 360 degrees about substantially the central axis of the steel pipe P, movement probes 21a and 21b attached to the distal end of the rod 20, and a rotation angle sensor 22 that determines the rotation angle of the rotary shaft of the rod 20, the distance between the center of rotation of the rod 20 and the measurement point on the outer periphery of the steel pipe P is measured using the displacement probes 21a and 21b for each rotation angle of the rod 20 in minutes, and then the shape of the outer diameter of the steel pipe P is determined based on the measured value.

(b) As shown in FIG. 7(b), by means of a device comprising a rotating rod 25 which rotates about the central axis of the steel pipe P, a frame (not shown) which is located on the end side of the rotating rod 25 and can be moved in the radial direction of the steel pipe P, a pair of pressure rollers 26a and 26b, which come into contact with the outer surface and the inner surface of the end of the steel pipe P and rotate when the rotating rod 25 is rotated, and a pair of pneumatic pressure cylinders attached to the frame, which press the pressure rollers 26a and 26b against the outer surface and the inner surface of the steel pipe P, the shape of the outer diameter of the steel pipe P is determined based on the amount of movement in the radial direction of the frame and the positions of the pressure rollers 26a and 26b pressed by the individual pneumatic cylinders.

In the present embodiment, by comparing the ovality prediction result obtained by the ovality prediction model described below with the measured ovality value obtained in the verification step described above, the accuracy of the ovality prediction can be confirmed. Thus, regarding the prediction result of the ovality prediction model described below, it is also possible to improve the prediction accuracy by adding the actual prediction error amount to the prediction result of the ovality prediction model.

Device for generating steel pipe ovality prediction model

In fig. 8 is a diagram showing the configuration of an apparatus for generating a steel pipe ovality prediction model according to an embodiment of the present invention. In fig. 9 is a configuration diagram of the offline pipe ovality calculator 112 of FIG. 8.

As shown in FIG. 8, the apparatus 100 for generating an ovality prediction model of a steel pipe according to an embodiment of the present invention includes an information processing device such as a workstation, and includes a raw data collection unit 110, a database 120, and an ovality prediction model generating unit 130.

The raw data collection unit 110 contains a working condition data set 111 obtained by determining the number of factors affecting the ovality of the steel pipe in the pressure bending step, the weld gap reduction step, the welding step, and the pipe expansion step; and an offline pipe ovality calculator 112 that outputs ovality data after the pipe expansion step using the operating condition data set 111 as the input condition.

In the present embodiment, the operating condition data set 111 includes at least attribute information related to the steel sheet used as a material, an operating parameter of the pressure bending step, and an operating parameter of the weld gap reducing step. This is because the above information has a great influence on the ovality of the steel pipe after the pipe expansion stage and influences the ovality fluctuation. However, other operating parameters of the welding step and the pipe expansion step can be taken into account. The data used for the operating conditions data set 111 will be described below.

The raw data collection unit 110 performs numerical operations with the help of the pipe ovality calculation unit 112 offline, variously changing the parameters included in the operating condition data set 111, thereby calculating the ovality of the steel pipe after the pipe expansion step corresponding to the multiple data set 111 for working conditions. The variation range of the parameter included in the operating condition data set 111 is determined based on the range that may vary as a normal operating condition according to the size of the steel pipe to be manufactured, equipment specifications for each step, and the like.

The pipe ovality calculation unit 112 offline calculates the shape of the steel pipe after the pipe expansion step by numerical analysis for a group of manufacturing steps from the pressure bending step to the pipe expansion step, and obtains the ovality of the steel pipe based on the shape after the pipe expansion step. In this case, the manufacturing steps include a pressure bending step, a weld gap reduction step, and a pipe expansion step. As shown in FIG. 9, the offline pipe ovality calculation block 112 contains: finite element model generation blocks 112a - 112c for the corresponding stages; and a finite element analysis solver 112d. It should be noted that the off-line pipe ovality calculation block 112 may contain a finite element model generation block corresponding to the end bending stage.

In the case where the finite element model generation block corresponding to the end bending step is included in the offline pipe ovality calculation block 112, the finite element model generation block of the end bending step performs division into elements inside the steel sheet based on attribute information about the steel leaf. Element division is performed automatically based on the specified element division condition. The finite element model of the end bending step that has undergone elementalization is passed to the finite element analysis solver 112d along with the end bending step calculation conditions. The design conditions for the end bending step include the operating parameters of the end bending step and also include all the information necessary to perform a finite element analysis that accurately determines all boundary conditions, such as physical property values of the workpiece, tool, etc. , geometric boundary conditions and mechanical boundary conditions. The shape of the steel sheet and the stress-strain distribution obtained by the finite element analysis in the end bending step are transmitted to the end bending step finite element model generation unit 112a as the initial conditions related to the workpiece in the end bending step.

The finite element analysis solver 112d may be constructed as a number of commercially available generic analysis software modules, and thus it is possible to appropriately select and implement some of them for effective use. Alternatively, it is also possible to use a mode in which the finite element analysis solver 112d is installed in a computer separately from the pipe ovality calculation unit 112 in an offline mode, and the input data includes the finite element model, and the output data is the result of the calculation are transmitted to block 112 for calculating pipe ovality in offline mode and are received from it. This is because after generating the finite element model corresponding to each step, the numerical analysis can be performed using a single finite element solver.

The finite element method is a method of approximate solution by dividing a continuous body into finite elements. Although the finite element method is an approximate solution, it is a method of obtaining a solution that satisfies the equilibrium of forces and the continuity of movement at the nodal points of the elements, allowing a highly accurate solution to be obtained under non-uniform deformation. In the finite element method, the stress, strain, and displacement in an element are determined independently for each element and are related to the displacement (velocity) of the node and are defined as a problem solving a system of equations. In a widely used method, the strain (increment) or stress is estimated by using the increment (velocity) at an element node as the unknown quantity.

In addition, the finite element method is characterized by the fact that the calculation is performed on the basis of the principle of possible displacements, expressed in integral form of the relative equilibrium state of stress in the element. The accuracy of the analysis result varies depending on conditions such as partitioning. In addition, the computation time required to perform the analysis is typically long. In addition, the finite element method is characterized by the ability to obtain a solution to a problem that is difficult to provide by other methods, and the specified solution satisfies the basic equation of plastic mechanics in the node or element. Thus, even in the case of a complicated forming process at the steel pipe manufacturing stage, it is possible to obtain solutions in relation to the displacement, mechanical stress field and deformation field of the workpiece that are close to reality.

The finite element analysis part of the solver can be replaced by various numerical analysis methods or approximate solutions, such as slip line field theory or the energy method. This allows you to reduce the overall calculation time. Moreover, the finite element analysis used in the present embodiment performs elasto-plastic analysis and does not include temperature field analysis such as thermal conductivity analysis. However, when the processing speed is high and the temperature of the workpiece increases greatly due to heat generation during processing, it is permissible to perform analysis by combining thermal conductivity analysis and elasto-plastic analysis. Moreover, the elasto-plastic analysis in the present embodiment is a two-dimensional transverse analysis at any of the bending steps, the weld gap reduction step, and the pipe expansion step, and is sufficient to perform a numerical analysis of the U-section, section of the fixed portion in the longitudinal direction when the steel sheet is formed to form an open pipe, and the steel pipe. To predict the high accuracy of the shape of a moving portion such as the leading end and the trailing end of a steel pipe, it is preferable to have a finite element model generation unit that performs three-dimensional analysis including the leading and trailing ends.

Attribute information about the steel sheet, which is the workpiece being processed, is provided as input data at the pressure bending stage. If an end bending step is provided as a step prior to the pressure bending step, the shape of the steel sheet and its stress/strain distribution obtained from the finite element analysis of the end bending step will be the initial conditions for the workpiece processed in the pressure bending step. Here, the finite element model generation unit 112a in the bending step performs element division within the steel sheet based on the size and shape of the steel sheet before the bending step. Element division is performed automatically based on the specified element division condition. The distribution of stress and strain within the workpiece can be assigned to each element based on the manufacturing history related to the steel sheet at the previous stage. This is because in the forming step, which mainly includes forming, the original residual stresses also affect the shape of the U-shaped workpiece of the steel sheet being processed.

Together with the finite element model of the bending step generated in this manner, the calculation condition of the bending step is provided as input to the finite element analysis solver 112d. It is intended that the design conditions for the bending stage include the operating parameters of the bending stage and all information necessary to perform a finite element analysis that accurately determines all boundary conditions, such as the values of the physical properties of the workpiece, tool, etc., geometric boundary conditions and mechanical boundary conditions.

The finite element analysis solver 112d performs numerical analysis according to the calculation conditions as specified above to obtain the shape of the U-shaped molded workpiece and the distribution of stress and strain remaining within the workpiece after the pressure bending step. The result obtained by such calculation is used as input for the next step of reducing the weld gap in the offline pipe ovality calculation block 112. Based on the calculated shape after the pressure bending step, the finite element model generation unit 112b of the weld gap reduction step performs element division in the U-shaped workpiece. Element division is performed automatically based on the specified element division condition. Preferably, the stress and strain distribution calculated for the previous step on each element should be specified. The reason for this task is similar to the reason described above.

Together with the finite element model of the weld gap reduction step generated in this manner, the calculation conditions for the weld gap reduction step are provided as input to the finite element analysis solver 112d. It is intended that the design conditions for the weld gap reduction stage include the operating parameters of the weld gap reduction stage and all information necessary to perform a finite element analysis that accurately determines all boundary conditions, such as the physical properties of the workpiece, tool, etc. etc., geometric boundary conditions and mechanical boundary conditions.

The finite element analysis solver 112d performs numerical analysis according to the calculation conditions as specified above to obtain the shape of the open pipe and the distribution of residual stress and strain within the workpiece after the weld gap reduction step. The result obtained through such calculation is used for the input data in the block. 112c generating a finite element model of the next stage, namely the pipe expansion stage. In the welding stage of the open pipe welding gap section, the residual stress and strain generated in the steel pipe after welding can be obtained by numerical analysis in the welding stage.

However, in many cases, performing accurate numerical analysis during the welding stage is difficult due to reasons such as the nature of the change in thermal conductivity due to the melting of the steel sheet during welding, and the effect on the mechanical properties of the heat-affected zone. In addition, the heat-affected zone resulting from welding only affects part of the shape of the steel pipe and has little effect on the remaining areas of the steel pipe. Thus, it is possible to ignore the influence of the heat-affected zone of the steel sheet at the welding stage on the ovality of the steel pipe after the pipe expansion stage.

In the welding step, welding is performed by holding the open pipe from the outside to reduce the gap in the open pipe, which changes the stress and strain distribution due to the change in elastic deformation in a region that is not located near the weld gap region. Thus, using the finite element analysis solver 112d, numerical analysis can be performed on the weld gap holding characteristics of the open pipe from the outside to set the weld gap of the open pipe to zero using the finite element method, and the result can correspond to the stress - deformed state after the welding stage.

However, when the weld gap reduction step in such a welding step corresponds to elastic deformation, the stress/strain distribution after the welding step can be obtained by combining the analytical solution of stress and strain with a beam with a curved axis according to beam theory in the stress and strain distribution in an open pipe, calculated using finite element analysis. This allows you to reduce 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 expansion step performs element division within the steel pipe. Element division is performed automatically based on the specified element division condition. Preferably, the stress and strain distribution calculated as described above should be specified for each element. The generated finite element model of the pipe expansion step is passed to the finite element analysis solver 112d along with the calculation conditions for the pipe expansion step. It is intended that the design conditions for the pipe expansion step include the operating parameters of the pipe expansion step of the present embodiment and also include all information necessary to perform a finite element analysis that accurately determines all boundary conditions, such as the physical properties of the workpiece, tool, etc., geometric boundary conditions and mechanical boundary conditions.

The finite element analysis solver 112d performs numerical analysis according to the calculation as described above to obtain the shape of the steel pipe and the distribution of internal stresses and strains after the pipe expansion step. The calculated shape of the steel pipe has an uneven distribution of curvature in the circumferential direction, and the ovality of the steel pipe is obtained according to the ovality determination in the ovality measurement step. Numerical analysis by the off-line pipe ovality block 112 using the finite element method sometimes requires a computation time of approximately 1 to 10 hours for one set of operating conditions data (a single case).

However, since the processing will be performed offline, there is no computation time limit. To reduce computation time for a large number of operating condition data sets, numerical operations corresponding to multiple operating condition data sets can be performed in parallel by multiple computers. This allows a database to be generated to generate an ovality prediction model within a short period of time. Moreover, in recent years, the computation time for a single case using general purpose graphics processing units (GPGPUs) is approximately 1/2 to 1/10 of the normal computation time, and such computing means can be used.

With reference to FIG. 8, a data base 120 stores a data set 111 of operating conditions and corresponding data related to the ovality of the steel pipe after the pipe expansion step. The data stored in the database 120 can be retrieved offline. Unlike a database collected as actual values of actual operations, an operating condition data set can be a flexible data set, making the database suitable for machine learning with less chance of statistical error occurring in the operating condition of the data set. In addition, since the calculation results obtained through precise numerical analysis are accumulated without storing training data that fluctuates over time, so the more data is collected, the more useful the database can be obtained.

The ovality prediction model generation unit 130 generates an ovality prediction model M trained by machine learning to obtain the ovality of the steel pipe after the pipe expansion step for the operating condition input data set 111 based on the relationship between the plurality of data from the operating condition data sets 111 and the ovality data. steel pipe stored in a database of 120 data. In particular, the relationship between the operating conditions in each step and the ovality of the steel pipe after the pipe expansion step may tend to exhibit complex nonlinearity. Thus, while modeling that assumes linearity would result in low-accuracy predictions, a machine learning method that uses a function that has non-linearity, a method such as a neural network, can perform high-accuracy predictions. Here, modeling means replacing the input data - output data relationship for a numerical operation with an equivalent functional form.

It is desirable that the number of databases required to generate the ovality prediction model M be 500 or more data blocks, although it varies depending on conditions such as the size of the steel pipe to be manufactured. Preferably, 2000 or more data blocks should be used, and more preferably 5000 or more data blocks. The machine learning method may be a known learning method. Machine learning is implemented by using, for example, a known machine learning method such as a neural network. Examples of other methods include decision tree, random forest, Gaussian process regression, support vector regression, and k-nearest neighbors. Although the ovality prediction model M is generated offline, the ovality prediction model generation unit 130 can be implemented into the control system in real time, and the ovality prediction model can be periodically updated using the offline calculated and accumulated database as needed. .

The steel pipe ovality prediction model M after the pipe expansion step generated as described above has the following characteristics. As attribute information about the steel sheet, for example, yield strength, sheet thickness, etc. have a certain level of variation in the production of steel sheet as a material, which affects the curvature of the steel sheet and the curvature after releasing the load during punch pushing in three-point bending forming in the pressure bending stage. Thus, by selecting the attribute information of the specified steel sheets as input parameters to the off-line generated ovality prediction model M, the influence of attribute information such as yield strength and sheet thickness on the ovality of the steel pipe after the pipe expansion step can be predicted. Moreover, the step of reducing the weld gap is a step of applying bending force and compressive force using a die or the like, as well as changing the curvature of the steel sheet after removing the load depending on the yield strength, sheet thickness, and the like. and thus the said information should be used as input parameters of the ovality prediction model M.

In addition, since the pressure bending step is a step of performing continuous processing by applying curvature many times in the width direction of the steel sheet, local distribution of the curvature occurs in the width direction of the steel sheet. After this, when the combined compressive and bending deformation is applied in the weld gap reduction stage, the bending moment applied in the weld gap reduction stage is distributed locally according to the local curvature distribution of the steel sheet applied in the pressure bending stage, similar to the case when the bending the moment acting on a beam, called a "curved beam", varies depending on the curvature of the beam before deformation. This causes the operating conditions of the pressure bending step to affect the curvature distribution in the width direction of the steel sheet after the welding gap reduction step. In this regard, it is important to use both the operating parameters of the pressure bending step and the weld gap reduction step as input parameters for the ovality prediction model M.

For example, in FIG. Figure 10 shows the results of measuring the ovality of a steel pipe after the pipe expansion stage (under the same operating conditions of the pipe expansion stage), when changing the weld gap reduction index in O-forming when using a device to make the workpiece O-shaped in the weld gap reduction stage welding, provided that the number of pressings during forming during the bending stage is set to nine when producing a steel pipe with an outer diameter of 30 inches and a pipe wall thickness of 44.5 mm. In fig. Figure 10 shows the results of changing the indentation amount (indentation amount at the final pass) during the last (ninth) pressing at three levels, while using the same settings for other operating conditions in the pressure bending stage.

As shown in FIG. 10, the O-forming weld gap reduction index, which is the forming amount in the weld gap reduction stage, has an optimal value for reducing the ovality of steel pipe after the pipe expansion stage, and it can be seen that the optimal value changes depending on the final indentation amount in the pressure bending stage, which is the working condition of the pressure bending stage. In other words, it is clear that in order to reduce the ovality of the steel pipe after the pipe expansion step, there is a need to change the operating conditions of the weld gap reduction step depending on the operating conditions of the pressure bending step, and it is not always possible to set the corresponding operating conditions only by focusing on that the operating conditions of the pressure bending stage and the weld gap reduction stage influence the ovality of the steel pipe after the pipe expansion stage as independent parameters.

In this regard, the ovality prediction model of the present embodiment can take into account the influence of such operating parameters of multiple manufacturing steps on the ovality of the steel pipe after the pipe expansion step, allowing the ovality to be predicted with high accuracy. In addition, since an ovality prediction model trained by machine learning is generated, the output ovality can be immediately calculated even when the variable that should be the input condition changes, allowing operating conditions to be immediately set and adjusted even in real-time use. Below is a description of each parameter used as input to the ovality prediction model.

Attribute information about steel sheet

The required attribute information of the steel sheet used as a material can be any parameter that affects the ovality of the steel pipe after the pipe expansion step, such as the yield strength of the steel sheet, tensile strength, tensile modulus, sheet thickness, thickness distribution in the sheet plane, yield stress distribution in the thickness direction of the steel sheet, degree of Bauschinger effect and surface roughness. In particular, it is desirable to use as an indicator the factor influencing the deformation state and springback of the steel sheet during three-point bending forming in the pressure bending step, and the factor influencing the deformation state and springback of the steel sheet during compression/bending processing in the gap reduction step. for welding.

The yield strength of the steel sheet, the distribution of the yield strength in the thickness direction of the steel sheet, and the thickness of the sheet have a direct effect on the stress and strain state in three-point bending forming. Tensile strength is a parameter reflecting the strain hardening state during bending processing, and influences the stress state during bending deformation. The Bauschinger effect influences the yield strength and subsequent strain hardening characteristics during load reversal due to bending deformation and influences the stress state due to bending deformation. In addition, the tensile modulus of the steel sheet influences the springback characteristics after bending processing. In addition, the thickness distribution in the plane of the sheet generates the bending curvature distribution in the pressure bending stage, and the surface roughness affects the friction state between the die and the steel sheet in the welding gap reduction stage, which affects the ovality of the steel pipe after the pipe expansion stage.

Of these attribute information blocks, it is especially preferred to use yield strength, representative sheet thickness, sheet thickness distribution information, and representative sheet width. These parameters, which are information measured in the quality control stage of thick sheet rolling, which is the production stage of the steel sheet used as a material, affect the deformation characteristics of the pressure bending stage and the weld gap reduction stage, and also affect the ovality steel pipe after the pipe expansion step, and thus should preferably be used as attribute information of the steel sheet in the raw data collection unit 110.

The yield strength is information that can be obtained from a tensile test of a small quality assurance specimen taken from a thick steel sheet used as a material, and may be a representative value in the plane of the steel sheet used as a material. In addition, the representative sheet thickness is a sheet thickness representing the sheet thickness in the plane of the steel sheet used as the material, and may be the thickness of the central portion in the width direction of the steel sheet at a random location in the longitudinal direction or the average value of the sheet thickness in the longitudinal direction. In addition, the average value of the sheet thickness over the entire plane of the steel sheet can be obtained and used as a 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 the thickening found in a steel sheet. The thickening represents the difference in sheet thickness between a central portion of the steel sheet in the width direction and a location away from the end of the steel sheet in the width direction at a predetermined distance (for example, 100 mm, 150 mm, etc.). The representative sheet width is a representative value related to the width of the steel sheet used as the material. The variation in the width of the thick steel sheet used as the material should be taken into account, or the width of the steel sheet varies when the end of the sheet is ground to obtain a bevel, which will affect the variation in the accuracy of the outer diameter of the steel pipe product.

Attribute information about the above-described steel sheet is information used to set operating conditions in the steel pipe manufacturing step as information collected by the host computer during real-time operation. The raw data collection unit 110 can select relevant information from the information blocks, which is combined with the attribute information of the steel sheets collected by the host computer in real time.

Operating parameters of the end bending stage

When the operating parameters of the end bending step are used as input to the ovality prediction model, a parameter for determining the shape formed by the forming surface 33a of the upper half-die 33 and the shape formed by the forming surface 34a of the lower half-die 34, which is used in the device 30 for giving the workpiece a C-shape , can be used as an operating parameter. In addition, it is also possible to use the end bending processing width (the width that is formed when the ends are bent), the pushing force (the C-forming force), and the gripping force of the clamping mechanism 37 in the end bending step as operating parameters. This is due to the fact that these parameters are factors that influence the deformation of the ends of the steel sheet along the width at the stage of bending the ends. When spatial deformation analysis is performed in the end bending step, the feed amount, feed direction and feed number of the steel sheet can be used as the operating parameters in the end bending step.

Consideration should be given to cases where the shape formed by the forming surface 33a of the upper half-matrix 33 is a shape having arcs with multiple radii of curvature in a continuous shape, or the shape is an involute or the like, allowing a parameter to be used to determine the geometric shape of the section. For example, when the section shape is formed by a parabolic shape, the section shape can be determined by using the coefficients of the first order term and the second order term of the quadratic expression representing the parabola passing through the origin, and thus the coefficient can be used as an operating parameter for the end bending step .

On the other hand, in the case where multiple dies are provided to obtain the shape formed by the forming surface 33a of the upper half-die 33, and the dies are suitably replaced and used according to conditions such as outer diameter, wall thickness and type of steel pipe to be manufactured, die number to determine the matrix used in the end bending step, can be used as the operating parameter of the end bending step.

Operating Parameters of Pressure Bending Stage

In the present embodiment, the operating parameter for the bending step is used as an input to the ovality prediction model. The operating parameters for the pressure bending stage can be various parameters that influence the amount of local bending curvature of the steel sheet and its distribution along the width of the steel sheet, such as those described above: the number of presses when forming a three-point bender, information about the forming location, the amount of indentation during forming, the gap in the lower die and the curvature of the punch. In particular, it is preferable to use information including all information about the forming position and the amount of indentation when forming during the operation of the punch forming the steel sheet and the number of pressings in the pressure bending step. The method presented in Fig. 11 can be given as an example of a case including all of these pieces of information.

In fig. 11(a) and 11(b) show an example of the forming indentation position and the forming indentation amount when the punch performs 16 indentations/10 indentations in steel sheets having the same width. The forming indentation position is information indicating the width distance from the base end of the steel sheet, and is used as the forming indentation position information. In addition, the molding indentation amount corresponds to each molding indentation position, and the “number of indentations”, the “molding indentation position” and the “molding indentation amount” can be combined into a data set. In the examples shown in FIGS. 11(a) and 11(b), the operating parameters of the bending step are determined by 16 sets of data and 10 sets of data at the number of presses of 16 and 10, respectively.

In the present embodiment, such a data set is used as input to an ovality prediction model in the following form. For example, the forming indentation position and the forming indentation amount can be used as input data of the ovality prediction model when performing the forming indentation at the closest position to the end at one end of the steel sheet, and the forming indentation position and the forming indentation amount can be used at closest position to the end at the other end of the steel sheet.

When the forming indentation amount at one end of the steel sheet increases during three-point bending forming, the amount of curvature at the sections of the steel pipe, in FIG. 2 corresponding to approximately the one o'clock position and a portion corresponding to approximately the eleven o'clock position are enlarged, resulting in a molded blank having a U-shaped cross-section, with the overall shape having a greater length in the horizontal direction. In addition, the closer the forming indentation position is to the end of the steel sheet, the lower the position of the weld gap portion, resulting in a molded piece having a U-shaped cross-section, with the overall shape having a longer length in the horizontal direction. As a result, the steel pipe formed into an open pipe and subjected to the welding step and the pipe expansion step will also have an overall longer length in the horizontal direction, which affects the ovality. In addition, the curvature of the punch during molding indentation, the total number of molding indentations, and the gap between the lower die elements during molding indentation also influence the ovality.

However, by using all the information of the molding indentation position and the data of the molding indentation amount together with the number of pressings as input to the ovality prediction model, the prediction accuracy of the ovality prediction model can be further improved. For example, taking into account the estimated maximum number of pressings, data about the molding indentation position and the molding indentation amount are stored in accordance with the number of pressings in the case of molding. The forming indentation position and the forming indentation amount in post-processing without forming are set to zero. For example, in the estimated case where the maximum estimated number of clicks in the examples in FIG. 11(a) and 11(b) are 16, and when the number of clicks is 10, the 11th to 16th click data, which is set to zero, will be the input to the ovality prediction model.

The above-described operating parameters of the bending step are information used as operating conditions set by the host computer in real-time operation. The input data collection unit 110 preferably selects a parameter to be used as input to the ovality prediction model from the operating parameters of the bending step collected in this manner by the host computer in real time operation.

Operating parameters of the welding gap reduction stage

In the present embodiment, the operating parameters of the weld gap reduction step are used as input to the ovality prediction model. When the O-shaping device is used in the weld gap reduction step, the O-forming indentation amount, the O-forming indentation position, and the O-forming die R can be used as the operating parameters. In another case where the locking molding method is used, in each of the above-described steps, the pressing position for locking molding and the pushing force for locking molding are used as operating parameters. In particular, when an apparatus for O-shaping a workpiece is used, it is preferable to use an O-forming indentation amount. This is because increasing the indentation value in O-forming would lead to a state in which the area between the position where the steel pipe receives the forming holding force from the upper half-die and the position where the steel pipe is held by the lower half-die is mainly near areas “three o’clock” and “nine o’clock” steel pipes would not be limited and where bending and compression deformations would be concentrated. This area will have increased curvature, which will affect the final ovality.

The above-described operating parameters of the weld gap reduction step are information used as operating conditions set by the host computer in real-time operation. The input data collection unit 110 preferably selects a parameter to be used as input data for the ovality prediction model from the operating parameters of the weld gap reduction step collected in this manner by the host computer in real time operation.

Operating parameters of the pipe expansion stage

In addition to the above-described operating parameters, when the operating parameters of the pipe expansion step are used as input to the ovality prediction model, the pipe expansion ratio can be used as the operating parameter of the pipe expansion step. Although the larger the pipe expansion ratio, the more the ovality of the steel pipe is reduced after the pipe expansion step, the upper limit value of the pipe expansion ratio is limited in terms of the compression yield strength of the finished steel pipe, and therefore the calculation condition in the block 110 raw data collection must be determined using a value within a range. Since the pipe expansion ratio is information necessary for adjusting the pipe expansion device, the pipe expansion ratio can be determined by a set value set by the high order solving device. As an operating parameter of the pipe expansion step, in addition to the pipe expansion ratio, the number of pipe expansion segments and the diameter of the pipe expansion segments can be used.

Method for predicting ovality

In the present embodiment, the ovality prediction model M generated offline by the above-described ovality prediction model generating unit 130 is used to perform real-time ovality prediction of a steel pipe after the pipe expansion step. During the initial procedure of predicting the ovality of the steel pipe after the pipe expansion step, a set of operating conditions data to be set as the operating conditions of the steel pipe manufacturing step is collected in real time (operating parameter acquisition step). This is the step of obtaining the necessary data from the host computer that performs the overall control of the steel pipe manufacturing step, or from the individual control computer in the forming processing step, as a set of operating conditions data that corresponds to the input data for the ovality prediction model generated as described above. Here, the term "real time" means the duration of a series of manufacturing steps from the time before the steel pipe manufacturing step begins until the pipe expansion step is completed. Thus, the processing does not necessarily involve performing any of the molding processing steps. The waiting period during transport of the steel sheet to the next stage between every two production stages is also referred to as “real time”. In addition, the period before starting the steel pipe manufacturing step and after completing the thick sheet rolling step in steel sheet manufacturing can also be referred to as “real time”. This is because upon completion of the thick plate rolling step in the production of the steel sheet used as the material, a set of operating condition data can be obtained to be input into the ovality prediction model of the present invention. The model used in real time is an ovality prediction model M trained by machine learning, and after setting the operating parameter to be the input condition, the ovality to be the output parameter can be immediately calculated and the operating conditions can be quickly changed and etc.

Prediction of the ovality of the steel pipe after the pipe expansion step can be performed at any time before or during the steel sheet manufacturing step. A data set of operating conditions to be input into the ovality prediction model M is appropriately generated according to the prediction execution time. In other words, when predicting the ovality of a steel pipe after pipe expansion is performed before performing the pressure bending step, it is possible to use the actual values (actual measurement values) for the attribute information of the steel sheet used as the material and the operating condition set value preset in the host computer , is used as an operating parameter for the next manufacturing step, which includes the pressure bending step.

In addition, when the pressure bending step ends and the prediction of the ovality of the steel pipe after pipe expansion is performed before starting the welding gap reduction step, the actual values (actual measurement values) for the attribute information of the steel sheet used as the material and the actual working values the conditions preset by the host computer shall be used as operating parameters for the next manufacturing step, which includes the weld gap reduction step. It should be noted that the preset operating condition values are predetermined values that have been set based on previous operating characteristics and are stored in the host computer.

As described above, the present embodiment of the invention uses a group of operating condition data sets obtained according to the ovality prediction time point of the steel pipe after the pipe expansion for the input data to the ovality prediction model, and the ovality of the steel pipe after the pipe expansion step, which is the output parameter. forecast in real time. This makes it possible to change the operating conditions of the subsequent manufacturing step according to the predicted ovality of the steel pipe, which leads to a further reduction in the ovality of the steel pipe after the pipe expansion step.

Ovality control method

Below will be a description of a method for adjusting the ovality of a steel pipe after the pipe expansion step of the present embodiment. In fig. 12 shows a flowchart for predicting the ovality of a steel pipe after pipe expansion before starting the pipe bending step. As shown in FIG. 12, this sequence involves obtaining data including: actual data related to attribute information of the steel sheet used as a material; predetermined values (predetermined operating values), which are preset as operating conditions of the pressure bending step; and predetermined values (predetermined operating quantities) that are preset as operating conditions of the weld gap reduction step by the host computer 140, and collecting these pieces of information as an operating condition data set 111. In addition, the target ovality value, which is preset as the target ovality of the steel pipe after the pressure bending step, is transmitted from the main computer 140 to the operating condition changing unit 150.

The ovality of the steel pipe after pipe expansion is predicted using the operating condition data set 111 obtained in this manner as input to the ovality prediction model M. Subsequently, the predicted ovality (predicted ovality value) is compared with the target ovality (predicted ovality value), and when the predicted ovality value is less than the target ovality value, the steel pipe is produced without changing the target values for the operating conditions of the pressure bending stage, the weld gap reduction stage and the pipe expansion. In contrast, when the predicted ovality is greater than the predetermined ovality value, a change in the operating conditions of the pressure bending step or the weld gap reduction step is performed.

Specifically, the change is performed one or two times or more to increase the number of presses in the pressure bending step with a smaller distance between pressing positions for forming. This reduces the ovality of the steel pipe after pipe expansion. In addition, the set value of the operating condition of the bending step, changed in this way, can again be used as input to the ovality prediction model M to perform the ovality prediction again, and confirmation can be made whether the predicted ovality is smaller than the set ovality value. , in order to determine the magnitude of the change in the operating conditions of the pressure bending stage.

Next, the changed operating condition of the pressure bending stage is transmitted to the operating condition control unit of the pressure bending stage, followed by determination of the operating conditions of the pressure bending stage. By repeatedly determining the ovality in the operating condition changing unit 150, a proper operating condition of the bending step can be set even when the ovality target value is set as a strict value, which allows the steel pipe to be manufactured by further reducing the ovality.

However, when the operating parameter of the weld gap reduction step is to be changed, the O-forming indentation amount changes. For example, a plurality of O-forming indentation amount conditions are input into the ovality prediction model M, and among the conditions to be specified, an O-forming indentation amount condition to obtain the smallest ovality is selected. Thus, the changed operating conditions of the weld gap reduction step are transmitted to the operating condition control unit of the weld gap reduction step, and the operating conditions of the weld gap reduction step are determined. In particular, changing the operating conditions can be performed both for the operating parameters of the pressure bending step and for the operating parameters of the weld gap reduction step. The method for changing the operating conditions is similar to the method described above. The greater the number of operating parameters to be changed, the more the ovality adjustment range of the steel pipe after expansion increases, which further reduces the ovality.

As described above, the ovality control method of an embodiment of the present invention uses an ovality prediction model capable of simultaneously considering the change in attribute information of the steel sheet used as a material and the effect on ovality due to the relationship of the pressure bending step and the weld gap reduction step. This makes it possible to set the proper operating conditions to reduce the ovality of the steel pipe after pipe expansion, and quickly change the operating conditions in real time, ensuring the production of a steel pipe having satisfactory ovality.

Below with reference to the table. 1 and fig. 16 is a description of an ovality control method according to an embodiment of the present invention, which is an ovality control method for a case involving a steel sheet end bending step before a pressure bending step.

In the present embodiment, a target step for making changes is first selected from a series of molding processing steps constituting a steel pipe manufacturing step. Next, before starting the intended modification stage, the ovality of the steel pipe after the pipe expansion stage is predicted using the ovality prediction model M. Then, a change is made to one or more operating parameters selected from at least the operating parameters of the target change step or one or more operating parameters selected from the operating parameters of the molding processing step that is performed after said target change step, such that to reduce the ovality of the steel pipe obtained after pipe expansion.

Here, the series of forming processing steps constituting the steel pipe manufacturing step includes an end bending step, a pressure bending step, a weld gap reduction step, and a pipe expansion step, which are steps of plastically deforming a steel sheet to form it into a predetermined shape. Regarding the intended stage for making changes, a specific stage is selected from these molding processing stages. Before performing the forming processing at the selected step scheduled for modification, the ovality of the steel pipe after the pipe expansion step is predicted using the ovality prediction model M for the steel pipe. At this time, the forming of the steel sheet is completed in the forming step before the target step for making changes, and thus, when the operating parameters of this processing step through forming from the input side are to be used, the actual data of the operating parameters can be used to input data into the M model ovality prediction. In contrast, the actual operating data cannot be collected in subsequent forming steps including the target step for making changes, and thus the ovality prediction model M for input data for the steel pipe uses a target value preset in the host computer or the like Thus, for a given material, the ovality of a steel pipe after the pipe expansion stage can be predicted.

Subsequently, a determination is made as to whether the ovality predicted as the ovality of the steel pipe after the pipe expansion step corresponds to the ovality acceptable for the finished product. With this operation, when the obtained ovality of the steel pipe after the pipe expansion step is less than the predicted value, it is possible to carry out a change in the operating conditions at the target change step and at the forming step performed after the target change step. Here, the operating parameter to be changed may be an operating parameter at the target step for making the change or an operating parameter at the molding steps performed after the target step for making the change. The operating parameter of the forming step suitable for changing the ovality of the steel pipe after pipe expansion will preferably be selected according to the difference between the predicted ovality and the ovality acceptable for the finished product. In addition, it is possible to change both the operating parameters at the target change step and the operating parameters at a certain molding step performed after the target change step. This is because the ovality of a steel pipe after the pipe expansion step can effectively change when there is a large difference between the predicted ovality and the ovality acceptable for the finished product.

In table 1 shows examples of a molding processing step selected as the target step for making changes and a molding processing step in which operating parameters can be changed accordingly. In case 1, the end bending step is selected as the target step for making changes in the steel pipe manufacturing stage, which includes the end bending step. Before starting the end bending step, the ovality of the steel pipe after pipe expansion is predicted, and the ovality is predicted using the specified operating parameter values in the forming processing step including the pressure bending step and the weld gap reduction step. When the predicted ovality is large, it is possible to change an arbitrary operating parameter at each forming step, namely, the end processing step, the pressure bending step, the weld gap reduction step, and the pipe expansion step. The operating parameters to be changed are not limited to the operating parameters of the end bending step, and they may also be the operating parameters of other forming processing steps. When attribute information of the steel sheet is included in the input data to the ovality prediction model M, actual data including measured values and the like related to the attribute information of the steel sheet can be used as input data before starting the end bending step that is targeted stage for making changes.

In Cases 2 and 3, the target step for making changes and the operating parameters to be changed can be selected according to a similar idea to Case 1. Case 4 is a case where the pipe expansion step is set as the target step for making changes. In this case, before the start of the pipe expansion stage, the ovality of the steel pipe after the pipe expansion stage is predicted using the ovality prediction model M. In this case, at least the actual operating data in the pressure bending step and the weld gap reduction step can be used as input to the ovality prediction model M. In addition, it is also possible to use the actual data from the attribute information of the steel sheet or the actual working data in the end bending step. Thus, the predicted ovality of the steel pipe after the pipe expansion stage is compared with the ovality acceptable for the finished product, and as the ovality decreases, the operating parameter at the pipe expansion stage changes. It is preferable to use the pipe expansion ratio as the operating parameter of the pipe expansion step to be changed. It should be noted that the amount of change from the original specified value of the pipe expansion ratio to be changed can be set based on acquired experience. However, when the input data of the ovality prediction model M includes the pipe expansion factor in the pipe expansion step, the ovality of the steel pipe after the pipe expansion step can be predicted again using the re-set value of the pipe expansion factor as the input to the ovality prediction model M, and can be the compliance of the conditions for making changes has been determined.

Table 1 Example Target stage for making changes End bending stage Pressure bending stage Welding gap reduction stage Pipe expansion stage 1 End bending stage 2 Pressure bending stage - 3 Weld gap reduction stage - - 4 Pipe expansion stage - - - - a processing step by molding for which the operating parameter is changed

Below with reference to FIG. 16 is a description of a method for adjusting the ovality of a steel pipe according to an embodiment of the present invention. The example shown in FIG. 16 is a case where the weld gap reduction step was selected as the target step for making changes, the pressure bending step was completed, and the U-shape molded workpiece was transferred to the weld gap reduction step. At this time, the actual operating data corresponding to the pressure bending step is transmitted to the operating condition changing unit 150. The actual operating data can be transmitted over a network from a control computer provided at each control step of each molding processing step. However, data may be immediately transmitted from the control computer at each forming processing step to the main computer 140, which performs overall control of the steel pipe manufacturing step, after which the data is transmitted from the main computer 140 to the operating condition changing unit 150. In addition, if necessary, actual data related to attribute information of the steel sheet is transmitted from the main computer 140 to the operating condition changing unit 150. In addition, the actual operating data in the end bending step can also be transferred if necessary. The set operating parameter values of the weld gap reduction step and the pipe expansion step, which are molding processing steps carried out after the target change step, and the set operating parameter values of the target change step are transmitted from the control computer for each step to the change unit 150 working conditions. However, when the set values of the operating parameters in the weld gap reduction step and the pipe expansion step are stored in the main computer 140, the set values can be transmitted from the main computer 140 to the operating condition changing unit 150. It should be noted that the predetermined ovality value determined according to the specifications for the finished steel pipe is transmitted from the main computer 140 to the operating condition changing unit 150.

The operating condition changing unit 150 predicts the ovality of the steel pipe after the pipe expansion step based on the specified information elements using the real-time ovality prediction model M, and compares the predicted ovality (predicted ovality value) with the target ovality (predetermined ovality value). When the predicted ovality value is smaller than the predetermined ovality value, the operating condition changing unit 150 leaves the set operating conditions of the remaining forming steps of the pressure bending step, the weld gap reduction step, and the pipe expansion step unchanged, and produces the steel pipe. Conversely, when the predicted ovality value is larger than the predetermined ovality value, the operating condition changing unit 150 changes at least the operating condition of the weld gap reduction step or the operating condition of the pipe expansion step. In particular, the amount of indentation in O-forming and the like can be changed. at the stage of reducing the welding gap. In addition, the pipe expansion ratio can be changed during the pipe expansion step. In addition, the O-forming indentation amount and the pipe expansion ratio can be changed.

The operating condition changing unit 150 can repeatedly perform the ovality prediction using the thus modified operating parameters as input data to the ovality prediction model M, confirm whether the predicted ovality is smaller than the predetermined ovality value, and determine the changed operating condition amount of the weld gap reduction step. and the pipe expansion stage. The modified operating conditions of the weld gap reduction step and the pipe expansion step are transmitted to separate control computers to be used as the operating conditions of the weld gap reduction step and the pipe expansion step. By repeatedly performing the ovality determination in the operating condition changing unit 150, the corresponding operating conditions of the weld gap reduction step and the pipe expansion step can be set even when a strict target ovality value is set, allowing the steel pipe to be manufactured with a further reduced ovality. In addition, it is also possible to carry out processes in which the adjustment of the ovality of the steel pipe after the pipe expansion step is carried out in such a way that the weld gap reduction step is determined as the target step for making changes; and again performed on the steel pipe formed and welded into an open pipe, adjusting the ovality of the steel pipe after the pipe expansion step, with the pipe expansion step being determined as the target step for making changes. This is because the accuracy of predicting the ovality of the steel pipe is further improved due to the fact that the actual operating data of the welding gap reduction stage has been obtained.

As described above, the method for controlling the ovality of a steel pipe according to an embodiment of the present invention uses the ovality prediction model M, which takes into account the influence on the ovality due to the relationship between the pressure bending step and the weld gap reduction step, so that proper operating conditions can be set for reducing ovality of the steel pipe after the pipe expansion step, making it possible to produce a steel pipe having acceptable ovality. In addition, it is possible to adjust the ovality with high precision, taking into account changes in the attribute information of the steel sheet used as the material.

Steel Pipe Ovality Prediction Device

Below with reference to FIG. 17 is a description of a steel pipe ovality prediction apparatus according to an embodiment of the present invention.

In fig. 17 is a configuration diagram of a steel pipe ovality predicting apparatus according to an embodiment of the present invention. As shown in FIG. 17, the steel pipe ovality prediction apparatus 160 according to an embodiment of the present invention includes an operating parameter acquisition unit 161, a data storage unit 162, an ovality prediction unit 163, and an output unit 164.

The operating parameter collection unit 161 includes a defined interface capable of collecting data from the ovality prediction model M generated by the machine learning unit from the ovality prediction model generating unit 130 . For example, the operating parameter collection unit 161 preferably includes a communication interface for collecting data of the ovality prediction model M from the ovality prediction model generating unit 130. In this case, the operating parameter acquisition unit 161 may receive the ovality prediction model M from the machine learning unit 100b using a preset communication protocol. In addition, the operating parameter collecting unit 161 collects operating conditions for the molding equipment (equipment for performing the molding processing step) from the control computer or host computer installed in the equipment used in each molding processing step. For example, the operating parameter collecting unit 161 preferably includes a communication interface for collecting operating conditions. In addition, the operating parameter collecting unit 161 can collect input information based on the user's operation. In this case, the steel pipe ovality predicting apparatus 160 also includes an input block containing one or more input interfaces that detect user input and collect input information based on the user's operation. Examples of the input unit include, but are not limited to, a hardware key, a capacitive key, a touch panel integrated with an output unit display, a microphone for voice input, and the like. For example, the input block receives operating condition input data for the ovality prediction model M obtained from the ovality prediction model generating block 130 using the operating parameter acquisition block 161 .

The data 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 devices. The data storage unit 162 functions, for example, as a main storage device, an auxiliary storage device, or a random access memory device. The data storage unit 162 stores any kind of information used to operate the apparatus 160 for predicting the ovality of a steel pipe. The data storage unit 162 stores, for example, the ovality prediction model M obtained from the ovality prediction model generation unit 130 using the operating parameter collection unit 161, the operating conditions obtained from the host computer using the operating parameter collection unit 161, and information about the ovality predicted a device 160 for predicting the ovality of a steel pipe. The data storage unit 162 may store a system program, an application program, or the like.

The ovality prediction unit 163 includes one or more processors. In the present embodiment, the processor includes, without limitation, a general purpose processor or a special purpose processor adapted for specific processing. The ovality prediction unit 163 is communicatively coupled to the individual components constituting the steel pipe ovality predicting apparatus 160 and controls the operation of the entire steel pipe ovality predicting apparatus 160 . The ovality prediction unit 163 may be any general-purpose electronic device such as a personal computer (PC) or a smartphone. The ovality prediction unit 163 is not limited to this, and may be one server device or a plurality of server devices capable of communicating with each other, or may be another electronic device assigned to the steel pipe ovality prediction device 160. The ovality prediction unit 163 calculates the predicted amount of ovality information of the steel pipe using the operating conditions obtained by the operating parameter acquisition unit 161 and the ovality prediction model M obtained from the ovality prediction model generation unit 130 .

The output unit 164 outputs a prediction value regarding the ovality information of the steel pipe calculated by the ovality prediction unit 163 to the device for setting the operating conditions of the forming processing equipment. Output block 164 may include one or more output interfaces that provide information and notification to the user. The output interface is, for example, a display. Examples of the display include an LCD display or an organic electroluminescent display. The output unit 164 outputs data obtained by operating the apparatus 160 for predicting the ovality of a steel pipe. The output unit 164 may be connected to the steel pipe ovality predicting device 160 as an external output device instead of being located in the steel pipe ovality predicting device 160. The connection method can be any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark). Examples of the output unit 164 include, but are not limited to, a display that outputs video information, a speaker that outputs audio information, and the like. For example, the output block 164 provides the user with a predicted value regarding the ovality information calculated by the ovality prediction block 163 . The user can appropriately set the operating conditions for the forming processing equipment based on the predicted ovality value output by the output unit 164.

A more preferable form of the apparatus 160 for predicting the ovality of a steel pipe after the pipe expansion step as described above is a terminal device such as a flatbed terminal comprising: an input unit 165 that collects input information based on a user's operation; and a display unit 166 that displays a prediction value regarding the ovality information calculated by the ovality prediction unit 163 . This type of apparatus has a function of receiving input information based on user operation from the input unit 165 and updating part or all of the operating parameters of the forming processing equipment already input in the apparatus 160 for predicting the ovality of the steel pipe using the received input information. In other words, when the ovality information of the steel pipe is predicted by the ovality prediction unit 163 for the steel sheet being processed in the forming equipment, a function may be provided to receive the work performed by the operator using the terminal device and perform an adjustment of part of the operating parameters of the forming equipment already entered to block 161 for collecting operating parameters. The operating parameter acquisition unit 161 stores the original input data for the operating parameters for which correction input data has not been received from the terminal device along with the operating parameters of the molding equipment, and changes only the operating parameter for which the correction input data has been input. With this configuration, new input data for the ovality prediction model M is generated in the operating parameter collecting unit 161, and the ovality prediction unit 163 for the steel sheet calculates a prediction value regarding the ovality information based on the input data. In addition, the calculated predicted value regarding the ovality information is displayed on the terminal device display unit 166 by the output unit 164. With this procedure, an employee such as a molding equipment operator and a production manager can immediately clarify the predicted value regarding the ovality information when the operating parameter of the molding equipment has been changed, and can promptly change the operating state to the corresponding operating state.

Examples

Example 1

In this example, a pipeline pipe steel sheet (API grade X60) with a thickness of 38.0 - 38.4 mm and a width of 2700 - 2720 mm was used, and the ovality prediction model was generated offline after the pipe expansion step according to the steel fabrication conditions. a pipe having a diameter of 36 inches after the pipe expansion step; When manufacturing the pipe, the following stages were performed: pressure bending stage, welding gap reduction stage, welding stage and pipe expansion stage. In fig. Figure 13 shows an example of a finite element model generated by the finite element model generation block at the stage of reducing the weld gap in relation to the present example. Abaqus 2019 was used as a solver for finite element analysis; the computation time per case was approximately three hours. The number of datasets accumulated in the database was 300, and Gaussian process regression using radial basis function as the basis function was used as the machine learning model.

Representative sheet thickness (average sheet thickness over the entire plane of the steel sheet), sheet width, and yield strength of the steel sheet were selected as attribute information of the steel sheet. The range of variation as the operating condition was determined from the manufacturing performance record, and the input data for the calculation varied within this range. Regarding the operating parameters of the pressure bending step, the number of molding indentations and the molding indentation positions were selected. In this case, the number of molding indentations varied from 7 to 15, with 11 indentations as the initial condition. Regarding the molding indentation position, pressings were performed at equal intervals across the width of the sheet depending on the number of indentations, and the molding indentation position was determined according to the number of indentations. The bend angle at one stroke was set to 30°C based on the amount of indentation during molding, determined by the amount at which the front end of the punch reaches a position located 15.8 mm from the line connecting the uppermost portions of the core elements.

Next, the steel sheet was placed on the die with the core spacing set to 450 mm, and pressing by molding was started from a position 1120 mm from the center portion in the direction of the width of the steel sheet set as the starting position using a punch, having a processing surface with a radius of 308 mm. When the pressing number of molding was 11, pressing by molding was performed five times on the right side as viewed in FIG. 2, in the direction of the central section along the width of the sheet, while the sheet feed pitch was 224 mm. After this, the end of the sheet on the left side, when looking at FIG. 2 was moved close to the core member, and indentation by molding was performed six times on the left half of the steel sheet starting from a position of 1120 mm from the end, while the sheet feeding pitch was 224 mm.

At the stage of reducing the welding gap, a device was used to give the workpiece an O-shape. The O-forming weld gap reduction rate was selected as the operating parameter of the weld gap reduction step, and the operating conditions were varied in the range of 1.0 - 3.0%. Another operating condition that had to be specified was that the upper half die had an arcuate surface with a radius R of 457.2 mm and a central angle θc of 60°, and a flat surface connected to the arcuate surface at an angle θd equal to 30°, and the lower half-matrix had a concave arc-shaped surface with a radius R equal to 502.9 mm. On the other hand, a constant value of 1.0% was used as the pipe expansion ratio, which was the operating parameter of the pipe expansion step.

In the present example, the analysis conditions as described above were set in the offline pipe ovality calculation unit, the analysis conditions were changed within the range of the above operating conditions, and the ovality calculation results after the pipe expansion step obtained by analysis were accumulated in the database. Next, based on the accumulated database, a model for predicting ovality was generated. In the present example, the ovality prediction model generated in this manner was used in real time. The ovality in the present example was defined as ovality = Dmax - Dmin when the outer diameters at positions facing each other were selected by dividing the pipe into 360 equal parts in the circumferential direction, and the maximum diameter and minimum diameter were determined as Dmax and Dmin, respectively.

In the real-time stage, before the start of the pressure bending stage, the representative sheet thickness and width of the steel sheet were obtained from the host computer as the actual attribute information data of the steel sheet used as the material. In addition, yield test data were obtained during the inspection phase of the thick plate rolling stage. At the same time, the specified values of the operating conditions for the pressure bending stage and the welding gap reduction stage were received from the main computer. In the steel pipe manufacturing step of the present example, the set value of the operating condition preset by the host computer was selected so that the number of presses in the pressure bending step was 11, the location at a distance of 1120 mm from the center portion in the width direction of the steel sheet was set to as the first molding indentation location, and the molding indentation locations were set at a pitch of 224 mm in the width direction of the steel sheet. In addition, the molding indentation amount at each molding indentation location was a predetermined value of 15.8 mm. However, in the weld gap reduction step using the O-shape forming device, the weld gap reduction rate in O-forming was set to 2% as the operating condition target value preset by the host computer.

In the present example, before starting the pressure bending step, the ovality of the steel pipe after the pressure bending step is predicted using the specified target values together with the representative sheet thickness and sheet width, which are the actual attribute information data of the steel sheet, as input data to the ovality prediction model. On the other hand, in the host computer, the target ovality value was set to 10 mm, the predicted ovality of the steel pipe (predicted ovality value) was compared with the target ovality value, and when the predicted ovality exceeded the target ovality value, the operating conditions of the pressure bending step were changed. The number of clicks was chosen as the operating condition to be changed. As a result, in the example of the invention, it was confirmed that the average ovality value was 4.0 mm, and the acceptance rate was 100%. In comparison, in the case where, as a comparative example, manufacturing was carried out with the operating condition of the bending step according to the target value preset by the host computer, the average ovality value was 11.2 mm, and the acceptance rate was 80%.

Example 2

In the present example, a pipeline pipe steel sheet (API grade X60) with a thickness of 50.0 - 50.4 mm and a width of 4450 - 4460 mm was used, and the ovality prediction model was generated offline after the pipe expansion step for the case of steel pipe manufacturing, having a diameter of 56 inches after the pipe expansion step, with a pressure bending step, a weld gap reduction step, a welding step, and a pipe expansion step. In this case, the parameters to be input into the steel pipe ovality prediction model, namely, the steel sheet attribute information parameter, the operating parameter of the pressure bending step, and the operating parameter of the welding gap reduction step, were selected so that they were the same parameters as in example 1. However, the ranges of these operating parameters were different from the ranges of example 1. The set of data operating conditions was specified in the range of representative sheet thickness of 50.0 - 50.4 mm and sheet width of 4450 - 4460 mm, which are the parameters attribute information about sheet thickness.

As the operating condition of the pressure bending step, the distance between the bottom members of the pressure bending equipment was set to 620 mm, and the starting position of pressing by forming with a punch whose front end had a radius of curvature of 478 mm was set to be at a distance of 1824 mm from the central portion of the steel sheet in the width direction. The operating condition data set was set in a range in which the number of indentation by molding in the bending step was from 7 to 15. The number of indentation by molding preset in the host computer was 11. In this case, the operating condition was set so that performing press-forming six times, specifically, performing press-forming five times with a predetermined feed pitch of 365 mm from one press-forming starting point for the steel sheet to the center of the steel sheet in the width direction, and performing press-forming six times with a predetermined feed pitch of 365 mm from the other indentation starting point by forming the steel sheet to the center of the steel sheet in the width direction. In addition, the amount of molding indentation was 33.8 mm at any molding indentation. When the number of molding indentations was different, only the feed pitch was changed, while the initial position of the molding indentation and the amount of molding indentation were set as constant set values.

At the stage of reducing the welding gap, a device was used to give the workpiece an O-shape. The upper half-die of the device for giving the workpiece an O-shape had an arcuate surface with a radius R of 704.0 mm and a central angle θc of 60°, and a flat surface connected to the arcuate surface at an angle θd of 30°. The lower half-matrix had a concave arc-shaped surface with a radius R equal to 704.0 mm. The operating parameter of the weld gap reduction step was the weld gap reduction rate of O-forming, and the operating condition data set was set in the range of 1.0 - 3.0%. The O-forming weld gap reduction rate preset in the host computer was 2%. In addition, the specified value of the pipe expansion ratio as the operating parameter in the pipe expansion step was 0.9%.

Many sets of operating conditions data were prepared, in which the manufacturing conditions in the steel pipe forming processing step as described above were set, and the representative sheet thickness and width, which are the attribute information parameters of the steel sheet, were changed, the number of indentations by forming, which is the working parameter in the pressure bending stage, and the weld gap reduction index in O-forming, which is the operating parameter in the weld gap reduction stage. Using the specified data sets, a finite element analysis was performed in the source data collection block, and information about the ovality of the steel pipe after the pipe expansion stage was accumulated in the database. Using 300 datasets accumulated in the database as output, the ovality prediction model of an ovality steel pipe after the pipe expansion step was generated offline by Gaussian process regression using the radial basis function as the basis function. in relation to a machine learning model.

The ovality prediction model generated as described above is transmitted to the operating condition changing unit and is configured to receive an operating condition database received from the host computer as input and configured to output a predicted ovality value of the steel pipe after the pipe expansion step. In the present example, the bending step is selected as the target step for making changes, and before the bending step starts, the representative sheet thickness and sheet width, which are the actual attribute information data of the steel sheet, and the operating target values of the bending step and the gap reduction step welding are retrieved from the host computer to compile a data set of operating conditions, and the ovality of the steel pipe after the pipe expansion step was predicted using the ovality prediction model. Here, the target ovality value set by the host computer for a given steel pipe was 14.2 mm, and the ovality is determined by comparing the target ovality value and the predicted ovality value. When the predicted ovality value exceeded the target ovality value, the operating condition of the weld gap reducing step, which is a forming processing step performed after the target step for making changes, had to be changed. The operating conditions that had to be changed were the reduction in weld gap in O-forming.

This ovality control method was used to produce 100 steel pipes. As a result, the average ovality value of the steel pipe after the pipe expansion stage was 10.0 mm in relation to the specified ovality value of 14.2 mm with an acceptance rate of 90%. In comparison, in the case where a steel pipe was produced without using the ovality control method of the present example and without changing the operating condition set in the host computer, the average ovality value was 14.4 mm with an acceptance rate of 60%.

Example 3

In the present example, pipeline pipe steel sheet (API grade X60) with a thickness of 38.0 - 38.4 mm and a width of 2700 - 2720 mm was used, and the ovality prediction model was generated offline after the pipe expansion step for the case of steel pipe manufacturing, having a diameter of 36 inches after the pipe expansion step, with an end bending step, a pressure bending step, a weld gap reduction step, a welding step, and a pipe expansion step. In this case, the parameters to be input into the steel pipe ovality prediction model, namely, the steel sheet attribute information parameter, the operating parameter of the pressure bending step, and the operating parameter of the welding gap reduction step, were the same parameters as in Example 1 In addition, the range of each operating parameter constituting the operating condition data set was also the same as the range in Example 1.

On the other hand, in the end bending step, the same pair of upper and lower half-dies mounted in the C-shape machine was used unchanged for each steel sheet, and the end bending width was selected to be as the operating parameter of the end bending step, and the data set of operating conditions in the raw data acquisition unit was changed so that the range changed from 180 mm to 240 mm. Moreover, in the present example, the pipe expansion ratio was selected as the operating parameter of the pipe expansion step, and the operating condition data set in the raw data acquisition unit was changed so that the pipe expansion ratio varied in the range of 0.6 - 1.4%,

Under the above conditions, Abaqus 2019 was used as the finite element analysis solver, and the finite element analysis was performed using the operating condition data set specified in the above range. The ovality values after the pipe expansion step, obtained through analysis, were accumulated in a database in relation to a data set of operating conditions. The number of data sets accumulated in the dataset was 600. The machine learning model used was Gaussian process regression using the radial basis function as the basis function, and an ovality prediction model having the ovality of the steel pipe after the pipe expansion stage as the output. , was generated offline.

The ovality prediction model generated as described above is implemented in an ovality prediction device, which performs real-time ovality prediction of a steel pipe. Using the operating parameter in the steel pipe manufacturing step received in real time from the host computer as input, the ovality prediction device outputs a predicted value for the ovality of the steel pipe after the pipe expansion step. The ovality predictor used in the present example was a tablet terminal capable of receiving input information from an input unit based on operator performance and capable of updating, by using the received input information, some or all of the operating parameters of the molding processing equipment that were input to the predictor. ovality. The specified tablet terminal has a function for recognizing an operating parameter, for which corrections are entered through operator action and output to the display unit of the predicted value from the ovality information displaying the correction input.

In the real-time manufacturing step of the present example, as the first step before starting the end bending step, the operating parameter acquisition unit obtains the actual attribute information data of the steel sheet and the preset operating values of the pressure bending step, weld gap reduction step, and pipe expansion step from main computer. Next, the ovality prediction unit outputs the predicted ovality value of the steel pipe after the pipe expansion step to the display unit of the terminal device, using the obtained operating condition data set as input data. In response, the operator who controls the steel pipe manufacturing process confirms the displayed predicted ovality value and compares it with the value (7.0 mm in this case) set as the target ovality value for the desired steel pipe. If the predicted ovality is large, the working condition in terms of width when processing ends by bending can be adjusted by inputting data from the control panel of the device to shape the workpiece into a C-shape.

In this example, the operator adjusts the width value when processing flexible ends, which is the operating parameter of the end bending stage and is displayed on the display unit of the terminal device, in the range of 180 - 240 mm. The ovality prediction device of the present example updates the bending end processing width adjusted by the operator to the adjusted value as a parameter to be an input value of the ovality prediction model. The ovality predictor displays the predicted ovality value in the terminal device, storing the value already obtained by the operating parameter acquisition unit in relation to other input data. The operator confirms the displayed predicted ovality value, determines the condition for the bending end processing width, which is the operating parameter of the end bending step, and sets a new target value from the control panel of the device to C-shape the workpiece.

Thus, using the ovality prediction device of the present example, 50 steel pipes were produced by operator action by changing the operating conditions in the steel pipe forming processing step. As a result, compared with the target ovality value of 7.0 mm, the ovality of the steel pipe as an average value after the pipe expansion step was 4.2 mm, and the acceptance rate was 100%. In comparison, in the case where the steel pipe was produced without using the ovality predictor of the present example and without changing the operating condition preset in the host computer, the average ovality value was 9.1 mm with an acceptance rate of 35%. In other words, it was confirmed that the ovality prediction device of the present example effectively assists the operator in the steel pipe manufacturing process.

Industrial applicability

The present invention provides a method for generating an ovality prediction model for a steel pipe, capable of generating an ovality prediction model that accurately and quickly predicts the ovality of a steel pipe after a pipe expansion step in a steel pipe manufacturing process including a plurality of steps. In addition, the present invention provides a steel pipe ovality prediction method and a steel pipe ovality prediction device capable of accurately and quickly predicting the ovality of a steel pipe after a pipe expansion step in a steel pipe manufacturing process including a plurality of steps. In addition, the present invention provides a method for controlling the ovality of a steel pipe, capable of accurately controlling the ovality of a steel pipe after a pipe expansion step in a steel pipe manufacturing process including a plurality of steps. In addition, the present invention provides a steel pipe manufacturing method capable of producing a steel pipe having a desired ovality with good product yield.

List of item numbers

1 - matrix

1a, 1b - rod element

2 - punch

2a - front end of the punch

2b - punch support

3 - upper semi-matrix

4 - lower half-matrix

10a, 10b - lower tool

11a, 11b - spring means

12 - punch

13 - upper tool

16 - segment for pipe expansion

17 - conical outer peripheral surface

18 - traction rod

20 - rod

21a, 21b - displacement probe

22 - rotation angle sensor

25 - rotary rod

26a, 26b - pressure roller

30 - devices for giving the workpiece a C-shape

31 - transport mechanism

31a - transport roller

32A, 32B - forming mechanism

33 - upper semi-matrix

33a - forming surface

34 - lower half matrix

34a - forming surface

36 - hydraulic cylinder

37 - clamping mechanism

110 - source data collection block

111 - operating conditions data set

112 - unit for calculating pipe ovality in offline mode

112a - block for generating a finite element model at the end bending stage

112b - block for generating a finite element model at the stage of reducing the welding gap

112c - block for generating a finite element model at the pipe expansion stage

112d - solver for finite element analysis

120 - database

130 - block for generating an ovality prediction model

140 - main computer

150 - block for changing operating conditions

160 - steel pipe ovality prediction device

161 - operating parameters collection block

162 - data storage unit

163 - ovality prediction block

164 - output block

165 - input block

166 - display block

G - welding gap area

M - ovality prediction model

P - steel pipe

R1, R2 - area

S - steel sheet

S ₁ - molded blank

S ₂ - open pipe

Claims (31)

1. A method for generating a steel pipe ovality prediction model, characterized in that the ovality prediction model is for predicting the ovality of a steel pipe after a pipe expansion step in which the inner diameter of a steel pipe whose ends are connected to each other is expanded in a steel pipe manufacturing process, which includes into yourself: a pressure bending step of processing the steel sheet through multiple pressings with a punch to obtain a molded workpiece having a U-shaped cross-section; a weld gap reducing step of reducing a weld gap portion of the molded blank having a U-shaped cross-section to obtain an open pipe; a welding step that connects the ends of the open pipe to each other; and the pipe expansion stage; the specified model generation method includes: an input data collection step in which a numerical operation is performed having as input a set of operating conditions data including: one or more parameters selected from the attribute information of the steel sheet, one or more parameters selected from the operating parameters of the bending step, and one or more parameters selected from the operating parameters of the weld gap reduction step; and having as output the ovality of the steel pipe after the pipe expansion stage; wherein the calculation is performed multiple times, changing the specified set of operating conditions; and generating, by said calculation, a plurality of ovality data sets of the steel pipe after the pipe expansion step corresponding to said operating condition data sets, offline as training data; And an ovality prediction model generation step of generating an ovality prediction model containing as input data the specified set of operating conditions data and as output data the specified ovality of the steel pipe after the pipe expansion step, wherein the model generation is performed offline using machine learning, using the specified set of training data sets obtained during the initial data collection stage. 2. The method for generating a steel pipe ovality prediction model according to claim 1, wherein the initial data collection step comprises the step of calculating the ovality of the steel pipe after the pipe expansion step, which is performed using the finite element method based on the specified set of operating conditions data. 3. The method for generating an ovality prediction model for a steel pipe according to claim 1 or 2, wherein the ovality prediction model includes, as a set of operating conditions data, one or more parameters selected from the operating parameters of the pipe expansion step. 4. A method for generating a model for predicting the ovality of a steel pipe according to any one of claims. 1 to 3, in which the steel pipe manufacturing step includes an end bending step by bending the ends of the steel sheet in the width direction before the pressure bending step, and as a set of operating conditions data, the ovality prediction model includes one or more parameters selected from the operating conditions parameters of the end bending stage. 5. A method for generating a model for predicting the ovality of a steel pipe according to any one of claims. 1 to 4, wherein the operating parameters of the bending step include: information about a forming position and a forming indentation amount related to an operation in which the punch used in the bending step forms the steel sheet; and the number of presses performed during the pressure bending phase. 6. A method for generating a model for predicting the ovality of a steel pipe according to any one of claims. 1-5, wherein said machine learning is a type of machine learning selected from the following algorithms: neural network, decision tree, random forest, Gaussian process regression, and support vector regression. 7. A method for predicting the ovality of a steel pipe, including an operating parameter collection step in which a set of operating condition data is collected in real time, which is to be set as the operating conditions of the steel pipe manufacturing step, and which is the input data of the steel pipe ovality prediction model generated by the steel pipe ovality prediction model generation method according to any one of paragraphs. 1–6, and an ovality prediction step of obtaining predictive information regarding the ovality of the steel pipe after the pipe expansion step by inputting the operating condition data set obtained in the operating parameter acquisition step as input to the ovality prediction model. 8. A method for regulating the ovality of a steel pipe, characterized in that performing a step of collecting a data set of operating conditions containing an actual value from the attribute information of the steel sheet, a specified operating parameter value of the pressure bending step, and a given operating parameter value of the welding gap reduction step, predicting the ovality of the steel pipe after the pipe expansion step by inputting the resulting set of operating conditions data into the ovality prediction model, wherein the prediction is performed before the start of the pressure bending step and using the method for predicting the ovality of the steel pipe according to claim 7, and changing the specified value of the operating parameter of the pressure bending stage and/or the specified value of the operating parameter of the welding gap reduction stage in such a way as to reduce the predicted ovality. 9. A method for regulating the ovality of a steel pipe, characterized in that making a prediction regarding the ovality of the steel pipe after the pipe expansion step, the prediction being performed before starting a target step for making changes selected from the end bending step, the pressure bending step, the weld gap reduction step, and the pipe expansion step, which are included in the steel pipe manufacturing process , and the prediction is performed using the method for predicting the ovality of a steel pipe according to claim 7, and changing one or more operating parameters selected from at least the operating parameters of the target step for making the change, or one or more operating parameters selected from the operating parameters of the molding processing steps performed after the specified target step for making the change, wherein said change is made based on the information obtained on predicting the ovality of the steel pipe. 10. A method for manufacturing a steel pipe, characterized in that a steel pipe is manufactured using the method for adjusting the ovality of a steel pipe according to claim 8 or 9. 11. A steel pipe ovality prediction device for predicting the ovality of a steel pipe after a pipe expansion step in which the inner diameter of a steel pipe whose ends are connected to each other is expanded in a steel pipe manufacturing process, which includes the following steps: a pressure bending step to process the steel sheet forming a molded blank having a U-shaped cross-section through a plurality of pressings with a punch; a step of reducing the weld gap by reducing a weld gap portion of the molded blank having a U-shaped cross-section to obtain an open pipe; a welding step for joining the ends of the open pipe to each other; and the pipe expansion stage; the specified device contains: an input data collection unit designed to perform a numerical operation having as input data a set of operating conditions data, including: one or more parameters selected from attribute information about the steel sheet, one or more operating parameters selected from the operating parameters of the pressure bending step, and one or more operating parameters selected from the operating parameters of the weld gap reduction step; and having as output data the ovality information of the steel pipe after the pipe expansion step, wherein the calculation is performed a plurality of times when the operating condition data set changes, and by means of said calculation, a plurality of data sets are generated regarding the ovality information of the steel pipe after the pipe expansion step corresponding datasets of operating conditions as training data; an ovality prediction model generating unit configured to generate an ovality prediction model containing a set of operating conditions data as input data, and containing information about the ovality of a steel pipe after the pipe expansion step as output data, wherein the model generation is performed by machine learning using a set of training data sets generated by the source data collection block; an operating parameter acquisition unit configured to obtain in real time a set of data of operating conditions to be set as operating conditions of the stages of the steel pipe manufacturing process; And an ovality prediction unit configured to predict ovality information of a steel pipe after the pipe expansion step corresponding to a set of operating conditions data obtained by the operating parameter acquisition unit, wherein the prediction is performed in real time using an ovality prediction model generated by the ovality prediction model generation unit. 12. A device for predicting the ovality of a steel pipe according to claim 11, containing a terminal device comprising an input unit configured to collect input information based on user operations, and an indication unit configured to display ovality information, wherein the operating parameter acquisition unit is configured to update part or all of the operating condition data set in the steel pipe manufacturing process based on input information received by the input unit, and the display unit is configured to display information about the ovality of the steel pipe predicted by the ovality prediction unit using the updated operating condition data set.