Method for predicting the cold dimensions of structural steel, method for manufacturing structural steel, method for generating a cold dimension prediction model, and apparatus for predicting the cold dimensions of structural steel

Abstract

This invention provides a method and apparatus for predicting the cold working dimensions of structural steel, which can predict the cold working dimensions of structural steel with high accuracy using the hot working dimensions of the structural steel. A method for predicting the cold dimensions of a structural steel, which is manufactured by hot-rolling a steel billet heated in a heating furnace in multiple passes and then cooling it, wherein input data including the hot dimensions of the structural steel, the hot temperature of the structural steel, and the cold temperature of the structural steel is input into a cold dimension prediction model, the cold dimensions of the structural steel are output and predicted, the cold dimensions of the structural steel output from the cold dimension prediction model are one or more of the cross-sectional dimensions of the structural steel, and the hot dimensions of the structural steel included in the input data include one or more of the cross-sectional dimensions.

Description

Technical Field

[0001] The present invention relates to a method for predicting the cold dimensions of section steel, a method for manufacturing section steel, a method for generating a cold dimension prediction model of section steel, and a cold dimension prediction apparatus for section steel.

Background Art

[0002] An H-shaped steel, which is a type of section steel, is mainly manufactured by hot rolling. In the manufacturing process of H-shaped steel, a steel slab is heated in advance and then subjected to hot rolling including rough rolling, intermediate rolling, and finish rolling by a plurality of rolling mills to be formed into a predetermined cross-sectional dimension.

[0003] The finish-rolled H-shaped steel is cooled on a cooling bed. The H-shaped steel cooled on the cooling bed is conveyed to an inspection table after being corrected for bending and warping by a roll straightening machine called a leveler. For the H-shaped steel, the cross-sectional dimension after cooling (hereinafter, the cross-sectional dimension after cooling is referred to as "cold dimension") is measured by an operator on the inspection table. In some cases, the cold dimension of the H-shaped steel is automatically measured by a cold dimension meter.

[0004] When the measured cold dimension is out of the range of the target dimension, for example, the roll gap (roll clearance) when performing rough rolling, intermediate rolling, and finish rolling is changed so that the cold dimension is within the range of the target dimension. However, when the cold dimension of the H-shaped steel is measured, several hours have passed since the finish rolling was completed by cooling on the cooling bed and correction by the roll straightening machine. Therefore, there is a problem that the adjustment of the roll gap is delayed, and a large amount of defective products outside the range of the target dimension are produced.

[0005] One way to address this is to measure the cross-sectional dimensions of the H-beam immediately after finish rolling (hereinafter, the hot cross-sectional dimensions after finish rolling will be referred to as "hot dimensions") and adjust the roll spacing during rough rolling, intermediate rolling, and finish rolling based on these hot dimensions, thereby resolving dimensional defects early on. However, the temperature of the H-beam immediately after finish rolling is much higher than that of room temperature. Therefore, the hot dimensions of the H-beam will be larger than the cold dimensions, and even if the hot dimensions are within the target range, the cold dimensions may not necessarily be within the target range. For this reason, it is necessary to convert the hot dimensions to cold dimensions and then adjust the roll spacing during rough rolling, intermediate rolling, and finish rolling.

[0006] Patent Document 1 discloses a technique for converting hot dimensions to cold dimensions, which involves measuring the cross-sectional dimensions after finish rolling and simultaneously measuring the temperature of the object to be measured, and then correcting the hot dimensions to cold dimensions using a linear formula for that temperature. According to Patent Document 1, by using a linear formula for the temperature at which the dimensions were measured, it is possible to correct the hot dimensions to cold dimensions, taking into account the change in cross-sectional dimensions due to thermal expansion.

[0007] Patent Document 2 discloses a technique for converting hot dimensions to cold dimensions by measuring the temperature distribution in the width direction of the flange and the temperature distribution in the height direction of the web, correcting the amount of thermal expansion of the measured parts with a linear expansion coefficient corresponding to the temperature, and calculating the sum of the amounts of thermal expansion for each part. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Japanese Patent Application Publication No. 7-120235 [Patent Document 2] Japanese Patent Publication No. 2008-249352 [Overview of the Initiative] [Problems that the invention aims to solve]

[0009] However, since Patent Document 1 corrects the dimensions using a linear formula of the temperature at which they were measured, it does not take into account volume expansion due to phase transformation when the temperature decreases. Therefore, the technology disclosed in Patent Document 1 has the problem that the accuracy of the correction to cold dimensions decreases when volume expansion due to phase transformation occurs.

[0010] Patent Document 2 has the problem that temperature measurement is time-consuming because the thermometer is moved in the height or width direction to measure the temperature distribution. This technology also has the problem that the accuracy of conversion from hot dimensions to cold dimensions decreases when volume expansion due to phase transformation occurs, as it does not model in detail the volume expansion due to phase transformation.

[0011] This invention was made in view of the above prior art, and its objective is to provide a method for predicting the cold working dimensions of structural steel and a device for predicting the cold working dimensions of structural steel that can predict the cold working dimensions of structural steel with high accuracy using the hot working dimensions of structural steel. Furthermore, another objective of this invention is to provide a method for generating a cold working dimension prediction model used in the method for predicting the cold working dimensions of structural steel and a method for manufacturing structural steel. [Means for solving the problem]

[0012] The means to solve the above problems are as follows: [1] A method for predicting the cold dimensions of a structural steel produced by hot rolling a steel billet in multiple passes and then cooling it, wherein input data including the hot dimensions of the structural steel, the hot temperature of the structural steel, and the cold temperature of the structural steel is input to a cold dimension prediction model, the cold dimensions of the structural steel are output to predict the cold dimensions of the structural steel, the cold dimensions of the structural steel output from the cold dimension prediction model are one or more of the cross-sectional dimensions of the structural steel, and the hot dimensions of the structural steel included in the input data include the cross-sectional dimensions corresponding to the cold dimensions of the structural steel output from the cold dimension prediction model. [2] The cold dimensional prediction method for a structural steel according to [1], wherein the input data includes at least one of the rolling parameters for the hot rolling and at least one of the attribute parameters for the structural steel. [3] The cold dimensional prediction method for a structural steel according to [1] or [2], wherein the structural steel is straightened with a roller straightening machine after cooling, and the input data includes one or more straightening parameters of the roller straightening machine. [4] A method for manufacturing a structural steel, which is produced by hot-rolling a steel billet in multiple passes and then cooling it, comprising: identifying the hot dimensions of the structural steel such that the cold dimensions of the structural steel predicted by the cold-dimension prediction method for structural steel described in any of [1] to [3] are within the range of target dimensions; using a hot-dimension prediction model that takes input data including one or more of the hot-rolling parameters as input and outputs the hot dimensions of the structural steel as output, identifying hot-rolling parameters that enable hot-rolling of the structural steel of the identified hot dimensions; and hot-rolling the steel billet in multiple passes under rolling conditions that include the identified hot-rolling parameters. [5] A method for generating a cold dimension prediction model used to predict the cold dimensions of structural steel produced by hot rolling a steel billet in multiple passes and then cooling it, wherein a machine learning model is trained using multiple datasets as training data, each dataset consisting of actual values ​​of input data including the hot dimensions, hot temperature, and cold temperature of structural steel produced in the past, and actual values ​​of the cold dimensions of the structural steel, and the cold dimension prediction model is generated which takes the input data as input and outputs the cold dimensions of the structural steel, the cold dimensions of the structural steel output from the cold dimension prediction model being one or more of the cross-sectional dimensions of the structural steel, and the hot dimensions of the structural steel included in the input data include cross-sectional dimensions corresponding to the cold dimensions of the structural steel output from the cold dimension prediction model. [6] A cold dimension prediction device for structural steel manufactured by hot rolling a steel billet in multiple passes and then cooling it, comprising a cold dimension prediction unit that inputs input data including the hot dimensions of the structural steel, the hot temperature of the structural steel, and the cold temperature of the structural steel into a cold dimension prediction model, outputs the cold dimensions of the structural steel, and predicts the cold dimensions of the structural steel, wherein the cold dimensions of the structural steel output from the cold dimension prediction model are one or more of the cross-sectional dimensions of the structural steel, and the hot dimensions of the structural steel included in the input data include the cross-sectional dimensions corresponding to the cold dimensions of the structural steel output from the cold dimension prediction model. [7] The cold dimensional prediction device for a structural steel according to [6], wherein the input data includes at least one of the rolling parameters for the hot rolling and at least one of the attribute parameters for the structural steel. [8] The cold dimensional prediction device for structural steel according to [6] or [7], wherein the structural steel is straightened with a roller straightening machine after cooling, and the input data includes one or more straightening parameters of the roller straightening machine. [9] A cold-work-size prediction device for a structural steel according to any one of [6] to [8], comprising a rolling parameter identification unit that identifies rolling parameters for hot rolling that enable hot rolling of a structural steel to be within a target cold-work-size range, wherein the rolling parameter identification unit identifies the hot dimensions of the structural steel to be within a target cold-work-size range, and identifies the rolling parameters for hot rolling that enable hot rolling of a structural steel to be hot-work-size the identified hot dimensions using a hot-work-size prediction model that takes input data including one or more of the hot-work-size parameters for hot rolling as input and outputs the hot dimensions of the structural steel as output. [Effects of the Invention]

[0013] The cold dimension prediction model used in the cold dimension prediction method for structural steel according to the present invention is generated using actual values ​​of the cold dimensions of structural steel that have undergone volume expansion due to phase transformation. Therefore, the prediction model is a cold dimension prediction model that takes into account volume expansion due to phase transformation. Thus, in the cold dimension prediction method according to this embodiment, cold dimensions are predicted using a prediction model that takes into account volume expansion due to phase transformation, making it possible to predict cold dimensions from hot dimensions with high accuracy.

[0014] The cold dimension prediction model used in the cold dimension prediction method for shaped steel according to the present invention uses the temperature measured during hot rolling. The temperature measured by a non-contact thermometer does not capture the temperature distribution in the thickness direction. Even if the temperature from the same temperature to room temperature contracts with different thicknesses, the amount of contraction is different. In contrast, in this cold dimension prediction model, since the temperature and thickness during hot measurement are measured simultaneously, it becomes a prediction model that takes into account the difference in thermal contraction due to different thicknesses. Thus, in the cold dimension prediction method for shaped steel according to the present invention, since the cold dimension is predicted using a cold dimension prediction model that takes into account the difference in thermal contraction due to different thicknesses, the cold dimension can be predicted with high accuracy from the hot dimension.

Brief Description of the Drawings

[0015] [Figure 1] Figure 1 is a cross-sectional schematic view showing the cross-sectional shape of an H-shaped steel. [Figure 2] Figure 2 is a schematic view showing an example of a rolling facility including a cold dimension prediction device for shaped steel in which the cold dimension prediction method for shaped steel according to an embodiment of the present invention can be implemented. [Figure 3] Figure 3 is a schematic view showing a configuration example of a cold dimension prediction device for shaped steel. [Figure 4] Figure 4 is a flowchart showing the flow of rolling parameter identification processing.

Embodiments for Carrying Out the Invention

[0016] First, using FIG. 1, the cross-sectional dimensions of the H-shaped steel 10, which is an example of a structural steel, will be described. FIG. 1 is a cross-sectional schematic view showing the cross-sectional shape of the H-shaped steel 10. As shown in FIG. 1, in the H-shaped steel 10, the portion with a relatively thick plate thickness is the flange 12, and the portion with a relatively thin plate thickness that is connected to the pair of flanges 12 is the web 14. The cross-sectional dimensions of the H-shaped steel 10 include the thickness of the web 14 (twb), the thicknesses of the upper, lower, left, and right of the flange 12 (tf1b, tf2b, tf3b, tf4b), the height of the web 14 (hwb), and the width of the flange 12 (wfb). In the following description, the thickness of the web 14 will be described as "twb". The thicknesses of the upper, lower, left, and right of the flange 12 will be described as "tf1b, tf2b, tf3b, tf4b". The height of the web 14 will be described as "hwb", and the width of the flange 12 will be described as "wfb".

[0017] There are numerous combinations of the cross-sectional dimensions of the H-shaped steel 10, about several tens for the H-shaped steel with a constant inner dimension (JISH) and about several hundreds for the H-shaped steel with a constant outer dimension. Here, the outer dimension is the web height hwb of the H-shaped steel 10, and the inner dimension is the dimension obtained by subtracting the flange thickness from the outer dimension.

[0018] When roughly classifying the strength of the H-shaped steel 10, it can be divided into three categories with a tensile strength greater than 400 N / mm 2 class, 490 N / mm 2 class, and greater than 490 N / mm 2 . The chemical composition of the H-shaped steel 10 is adjusted for each steel type classification (three types: 40k steel, 50k steel, and greater than 50k) according to this strength classification. Such H-shaped steel 10 is mainly manufactured by hot rolling. The H-shaped steel 10 manufactured by hot rolling is called rolled H-shaped steel.

[0019] In the cold dimension prediction method for structural steel and the cold dimension prediction device 46 for structural steel according to the present embodiment, the cross-sectional dimensions in the hot state (hot dimensions) of the hot-rolled structural steel are used to predict the cross-sectional dimensions in the cold state (cold dimensions) of the structural steel after cooling. Hereinafter, an example in which an embodiment of the present invention is applied to an H-shaped steel, which is an example of a structural steel, will be described.

[0020] Figure 2 is a schematic diagram showing an example of a rolling mill 100 including a cold dimension prediction device 46 for structural steel that can implement the cold dimension prediction method for structural steel according to one embodiment of the present invention. The rolling mill 100 is installed in a rolling mill. The rolling mill 100 has a heating furnace 20, a roughing mill 22, an intermediate rolling mill 24, a finishing mill 30, a hot shape gauge 32, and a hot thermometer 34.

[0021] The heating furnace 20 is a device for heating the steel billet to a predetermined temperature. In the heating furnace 20, the steel billet is heated to above the austenite temperature range (for example, 1100-1300°C or higher). The steel billet extracted from the heating furnace 20 is rolled in the roughing mill 22. In the roughing mill 22, the rectangular cross-section steel billet is roughly shaped to approach the cross-sectional dimensions of an H-beam by reverse rolling with 5 to 30 passes using a die roll. The example shown in Figure 2 illustrates the case where the steel billet is heated in the heating furnace 20, but if hot rolling is possible, the steel billet does not necessarily have to be heated in the heating furnace 20. The steel billet may be heated in a tunnel furnace or induction heating may be used. Alternatively, the cast steel billet may be hot-rolled directly without heating.

[0022] The intermediate rolling mill 24 uses a combination of one or more intermediate universal rolling mills 26 and edger rolling mills 28. The intermediate universal rolling mill 26, which is one of the intermediate rolling mills 24, is a rolling mill having a total of four rolls: upper and lower horizontal rolls that are driven and rotate around a horizontal axis, and left and right vertical rolls that rotate freely around a vertical axis.

[0023] The diameter of the upper and lower horizontal rolls of the intermediate universal rolling mill 26 is approximately 1000 to 1500 mm. The width of the upper and lower horizontal rolls is adjusted according to the hwb of the H-beam 10 to be rolled, and rolls of appropriate width are incorporated into the rolling mill. The diameter of the left and right vertical rolls is approximately 600 to 1000 mm. These upper and lower horizontal rolls and left and right vertical rolls are structured so that their positions can be adjusted by the reduction device, and the spacing between the upper and lower horizontal rolls and the spacing between the sides of the horizontal rolls and the vertical rolls can be set arbitrarily. The intermediate universal rolling mill 26 can reduce the web 14 and flange 12 simultaneously, and the flange 12 is reduced with a maximum inclination of approximately 10° outward. The intermediate universal rolling mill 26 performs reverse rolling for approximately 5 to 25 passes.

[0024] The edger rolling mill 28, one of the intermediate rolling mills 24, has two horizontal rolls, one above and one below, with holes (hereinafter, the horizontal rolls will be referred to as "E1 rolls"). The diameter of the E1 rolls is approximately 800 to 1200 mm, and both the upper and lower E1 rolls are driven. The wfb is adjusted by pressing down on the tip of the flange 12 from above and below with these E1 rolls. In the edger rolling mill 28, reverse rolling is performed by the E1 rolls for approximately 5 to 25 passes to form the cross-sectional dimensions of the H-shaped steel.

[0025] A finishing universal rolling mill is used for the finishing rolling mill 30. The finishing rolling mill 30 is a rolling mill that finishes the intermediate-rolled material to the product cross-sectional shape in a single pass of rolling. The dimensions and configuration of the rolls of the finishing universal rolling mill are the same as those of the intermediate universal rolling mill 26, but in the finishing universal rolling mill, the material is rolled so that the flange 12 is perpendicular to the web 14. The finishing-rolled H-beam 10 is cut to a predetermined product length by a hot saw (not shown).

[0026] The hot shape gauge 32 is installed downstream of the finishing rolling mill 30 and measures the hot dimensions, which are the cross-sectional dimensions of the H-shaped steel beam 10 rolled by the roughing rolling mill 22, the intermediate rolling mill 24, and the finishing rolling mill 30. The hot shape gauge 32 measures the twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb of the H-shaped steel beam 10 while it is hot.

[0027] In the hot shape gauge 32, it is preferable to measure the cross-sectional dimensions along the entire length and use the average value of the measured cross-sectional dimensions. Furthermore, it is even more preferable to use the average value of the cross-sectional dimensions excluding a few meters from the ends of the H-shaped steel beam 10, which are the non-steady sections. The hot dimensions of the H-shaped steel beam 10 measured by the hot shape gauge 32 are output to the process computer 44.

[0028] The hot thermometer 34 is installed near the hot shape gauge 32 and measures the temperature of the flange 12 and web 14 of the H-beam 10 whose hot dimensions have been measured by the hot shape gauge 32 (hereinafter, this temperature will be referred to as "hot temperature"). For example, the hot thermometer 34 measures the temperature of the flange 12 at a position on the outer surface of the flange that is 1 / 4 of the height of the flange 12, and for example, the temperature of the web 14 at a position on the top surface of the web that is in the center of the web 14. The hot thermometer 34 outputs the hot temperature to the process computer 44. If the hot shape gauge 32 has a temperature measuring function, the hot temperature of the H-beam 10 may be measured with the hot shape gauge 32.

[0029] The rolling mill 100 further includes a cooling bed 36, a roller straightening machine 38, a cold shape gauge 40, a cold thermometer 42, a process computer 44, and a cold dimension prediction device 46 for the structural steel. The H-shaped steel 10, whose hot dimensions and hot temperature have been measured, is transported to the cooling bed 36. The cooling bed 36 is a device for cooling the hot-rolled H-shaped steel 10. The H-shaped steel 10 is left on the cooling bed 36 for several hours and cooled to room temperature by air cooling. In some cases, mist is sprayed onto the H-shaped steel 10 on the cooling bed 36 to speed up the cooling rate. Room temperature refers to the room temperature of the rolling mill where the rolling mill 100 is installed and the H-shaped steel 10 is manufactured.

[0030] The H-shaped steel beams 10, cooled on the cooling bed 36, are straightened by a roller straightening machine 38 to correct any bends or warps. The roller straightening machine 38 is a device that straightens the bends and warps of the flanges 12 and webs 14 of the H-shaped steel beams 10. The roller straightening machine 38 has a structure that allows adjustment of the roll pitch and reduction amount of the straightening rolls, and these are adjusted so that the bends and warps of the H-shaped steel beams 10 are properly straightened.

[0031] The H-beams 10, whose bends and warps have been corrected by the roller straightening machine 38, are transported to the cold shape gauge 40. The cold shape gauge 40 measures the cold dimensions, which are the cross-sectional dimensions of the cooled H-beams 10. The cold shape gauge 40 measures the twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb of the H-beams while they are cold. The cold dimensions may also be measured by an operator on an inspection table (not shown).

[0032] In cold working dimensions, it is preferable to measure the cross-sectional dimensions along the entire length and use the average value of the measured cross-sectional dimensions. Furthermore, it is even more preferable to use the average value of the cross-sectional dimensions excluding a few meters from the ends of the H-shaped steel beam 10, which constitutes the transient section. The measured cold working dimensions are output to the process computer 44.

[0033] The cold thermometer 42 is installed near the cold shape gauge 40 and measures the temperature of the flange 12 and web 14 of the H-beam 10 whose cold dimensions have been measured (hereinafter referred to as the "cold temperature"). For example, the cold thermometer 42 measures the temperature of the flange 12 at a position that is 1 / 4 of the height of the flange 12, and for example, the temperature of the web 14 at the center of the web 14. Since the temperature of the H-beam 10 cooled to room temperature is the same as the room temperature of the rolling mill, the cold thermometer 42 may measure the room temperature of the rolling mill instead of measuring the temperature of the flange 12 and web 14 of the H-beam 10, and use that temperature as the cold temperature. In this embodiment, an example in which the room temperature of the rolling mill is used as the cold temperature will be described. The cold thermometer 42 outputs the cold temperature to the process computer 44.

[0034] The process computer 44 can be a general-purpose computer such as a workstation or personal computer. The process computer 44 is connected by wire or wireless to the heating furnace 20, roughing mill 22, intermediate rolling mill 24, finishing mill 30, hot shape gauge 32, hot thermometer 34, cold shape gauge 40 and cold thermometer 42, and oversees the manufacturing process of the H-shaped steel 10 by the rolling equipment 100. The process computer 44 also obtains attribute parameters of the H-shaped steel 10 to be manufactured from a higher-level computer. The attribute parameters of the H-shaped steel 10 include information on the target dimensions of the H-shaped steel 10, steel grade classification (3 types: 40k steel, 50k steel, and above), chemical composition (content of C, Si, Mn, Cr, Mo, V, etc.), and target values ​​of mechanical properties (yield stress, tensile strength, elongation, toughness, hardness, etc.). The target dimensions for H-beam 10 are the target dimensions (typical values ​​and tolerances) for twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb.

[0035] The process computer 44 sets hot rolling parameters such as the number of passes for each rolling mill and the roll spacing of each rolling roll, according to the attribute parameters of the H-shaped steel beam 10. The normal hot rolling parameter settings are set as table values ​​associated with the attribute parameters of the H-shaped steel beam 10, based on past rolling performance.

[0036] The process computer 44 outputs the set hot rolling parameters to the roughing mill 22, intermediate universal rolling mill 26, edger rolling mill 28, and finishing rolling mill 30. Each rolling mill controls the roll spacing, roll position, and rolling load of its respective rolling mill in accordance with the acquired hot rolling parameters. The process computer 44 acquires attribute parameters, hot rolling parameters, measured values ​​of hot / cold dimensions, measured values ​​of hot / cold temperatures, and identification numbers that identify the roll sets of each rolling mill. The process computer 44 stores this data in its database, associating it with the manufacturing lot number of the H-shaped steel beams 10.

[0037] The cold dimension prediction device 46 for structural steel predicts the cold dimension of the H-shaped steel 10 after the hot dimension of the H-shaped steel 10 is measured by the hot shape gauge 32 and the hot temperature is measured by the hot thermometer 34. The cold dimension predicted by the cold dimension prediction device 46 for structural steel only needs to be one or more of the twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb of the H-shaped steel 10. However, since all of the above cold dimensions are under management for the H-shaped steel 10, it is preferable for the cold dimension prediction device 46 to predict all of the cold dimensions of twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb. In the following embodiment, an example in which the cold dimension prediction device 46 for structural steel predicts all of the cold dimensions of the H-shaped steel 10 will be used for explanation.

[0038] The cold dimension prediction device 46 for structural steel acquires input data from the process computer 44, including the hot dimensions, hot temperature, and cold temperature (room temperature of the rolling mill) of the H-shaped steel 10, when the hot dimensions are measured by the hot shape gauge 32 and the hot temperature is measured by the hot thermometer 34. The cold dimension prediction device 46 for structural steel inputs the acquired input data into a cold dimension prediction model and outputs the cold dimensions, thereby predicting the cold dimensions of the H-shaped steel 10 after it has been cooled to room temperature and any bending or warping has been corrected.

[0039] The cold dimension prediction device 46 for structural steel may identify the hot dimensions at which the predicted cold dimensions of the H-shaped steel 10 fall within the range of the target dimensions, and identify the rolling parameters for hot rolling that enable the manufacture of the H-shaped steel 10 with the identified hot dimensions. The cold dimension prediction device 46 for structural steel may also identify the rolling parameters for hot rolling that enable the manufacture of the H-shaped steel 10 with hot dimensions within the range of the target dimensions by performing an inverse analysis of a hot dimension prediction model that takes input data including rolling parameters for hot rolling as input and outputs the hot dimensions of the H-shaped steel 10. Here, inverse analysis of a hot dimension prediction model means an analysis method that uses a hot dimension prediction model to identify combinations of input data that produce the desired output data through optimization calculations. In other words, the cold dimension prediction device 46 for structural steel uses a hot dimension prediction model to identify the rolling parameters for hot rolling that produce the desired hot dimensions of the H-shaped steel 10 (hot dimensions of the H-shaped steel 10 that fall within the range of the target dimensions) through optimization calculations. The cold-working dimension prediction device 46 for structural steel may output the identified hot-rolling rolling parameters to the process computer 44, thereby reflecting those rolling parameters in the hot-rolling rolling conditions.

[0040] Next, a cold working dimension prediction device 46 for structural steel that predicts the cold working dimensions of H-shaped steel 10 will be described. Figure 3 is a schematic diagram showing an example configuration of the cold working dimension prediction device 46 for structural steel. The cold working dimension prediction device 46 for structural steel can be a general-purpose computer such as a workstation or personal computer. The cold working dimension prediction device 46 for structural steel has a control unit 48, an input unit 50, an output unit 52, and a storage unit 54. The control unit 48 is, for example, a CPU, and by executing a program stored in the storage unit 54, it functions as a data acquisition unit 56, a cold working dimension prediction unit 58, a rolling parameter identification unit 60, and a prediction model generation unit 62.

[0041] The input unit 50 is, for example, a keyboard, a touch panel integrated with a display, etc. The output unit 52 is, for example, an LCD or CRT display, etc. The storage unit 54 is, for example, an information recording medium such as a flash memory that can be updated and recorded, a hard disk that is built-in or connected via a data communication terminal, a memory card, etc., and a device for reading and writing the same. The storage unit 54 stores programs and data for realizing each function of the cold-worked dimension prediction device 46 for structural steel.

[0042] The storage unit 54 further stores a database 64, a cold dimension prediction model 66, and a hot dimension prediction model 68. The database 64 stores more than 1000 datasets, each containing a set of actual values ​​for the hot dimensions, hot temperature, cold temperature, and cold dimensions of H-beams 10 previously manufactured by the rolling mill 100. Preferably, more than 3000 datasets are stored, each containing a set of actual values ​​for the hot dimensions, hot temperature, cold temperature, and cold dimensions of H-beams 10. These datasets are used as training data for generating the cold dimension prediction model 66.

[0043] The database 64 also stores more than 1000 datasets, each containing a set of rolling parameters and actual hot dimensions of H-shaped steel beams 10 previously manufactured by the rolling mill 100. Preferably, the database also stores more than 3000 datasets containing a set of rolling parameters and actual hot dimensions of H-shaped steel beams 10. These datasets are used as training data to generate the hot dimension prediction model 68.

[0044] Preferably, the database 64 stores the above-mentioned datasets for the H-shaped steel 10 manufactured by the rolling mill 100, which are sequentially acquired from the process computer 44. In this case, an upper limit may be set on the number of datasets stored in the database 64, and older datasets may be deleted in order to prevent the number from exceeding that limit.

[0045] The cold dimension prediction model 66 is a pre-trained machine learning model that has been trained using multiple datasets stored in the database 64 as training data. The cold dimension prediction model 66 is a pre-trained machine learning model that takes the hot dimensions (twb, tf1b, tf2b, tf3b, tf4b, hwb, wfb), hot temperature, and cold temperature of the H-beam 10 as input and outputs the cold dimensions (twb, tf1b, tf2b, tf3b, tf4b, hwb, wfb) of the H-beam 10. The cold dimension prediction model 66 is used to predict the cold dimensions of the H-beam 10.

[0046] The hot dimension prediction model 68 is also a trained machine learning model that has been trained using multiple datasets stored in the database 64 as training data. The hot dimension prediction model 68 is a trained machine learning model that takes hot rolling parameters as input and outputs the hot dimensions of the H-beam 10. The input hot rolling parameters include at least the roll spacing in each pass of the roughing mill 22, intermediate rolling mill 24, and finishing mill 30. The hot rolling parameters may also include the rolling load in each pass of the roughing mill 22, intermediate rolling mill 24, and finishing mill 30. The hot dimension prediction model 68 is used to identify the hot rolling parameters that can hot roll an H-beam 10 to a specific hot dimension by inverse analysis of the prediction model.

[0047] Next, the processes performed by the data acquisition unit 56 and the cold dimension prediction unit 58 when predicting the cold dimensions of the H-shaped steel beam 10 will be described. The data acquisition unit 56 acquires the hot dimensions, hot temperature, and cold temperature of the H-shaped steel beam 10 as input data from the process computer 44. The data acquisition unit 56 outputs the acquired input data to the cold dimension prediction unit 58. The cold temperature is the room temperature of the rolling mill at the time the cold temperature is predicted.

[0048] When the cold dimension prediction unit 58 acquires input data, it reads the cold dimension prediction model 66 from the storage unit 54. The cold dimension prediction unit 58 inputs the input data into the cold dimension prediction model 66 and outputs the cold dimensions of the H-beam 10. In this way, the cold dimension prediction unit 58 predicts the cold dimensions of the H-beam 10 using the hot dimensions of the finished-rolled H-beam 10. The cold dimension prediction unit 58 may also display the predicted cold dimensions of the H-beam 10 on the output unit 52. This allows the operator to confirm the predicted cold dimensions of the H-beam 10 by visually checking the output unit 52.

[0049] The cold dimension prediction model 66 used to predict the cold dimensions of the H-shaped steel beam 10 is generated using actual values ​​of the cold dimensions of the H-shaped steel beam 10 after volume expansion due to phase transformation. Therefore, this model is a cold dimension prediction model that takes volume expansion due to phase transformation into account. Consequently, it can be seen that predicting the cold dimensions of the H-shaped steel beam 10 using this prediction model can predict the cold dimensions with higher accuracy than prediction models that do not take phase transformation into account.

[0050] The input data for the cold dimension prediction model 66 includes the hot dimensions, hot temperature, and cold temperature of the H-beam 10. The hot dimensions, hot temperature, and cold temperature of the H-beam 10 all affect the output cold dimensions of the H-beam 10. Therefore, including the hot dimensions, hot temperature, and cold temperature of the H-beam 10 in the input data for the cold dimension prediction model improves the accuracy of the cold dimension prediction by the cold dimension prediction model 66.

[0051] Preferably, the input data for the cold dimensional prediction model 66 further includes one or more of the rolling parameters for hot rolling. The rolling parameters for hot rolling are, for example, the parameters listed below.

[0052] <Rolling parameters for hot rolling> • Roll spacing of the rolling rolls in each pass of the roughing mill 22, intermediate rolling mill 24, and finishing mill 30 • Time from extraction from the heating furnace 20 to finish rolling in the finish rolling mill 30 Rolling loads in each pass of the roughing mill 22, intermediate rolling mill 24, and finishing mill 30 • Steel billet thickness, steel billet width, steel billet length, steel billet weight

[0053] These hot-rolling parameters affect the hot dimensions of the H-beam 10. As mentioned above, since hot dimensions affect cold dimensions, it can be said that the hot-rolling parameters also affect cold dimensions. Therefore, by including these hot-rolling parameters in the input data of the cold-dimension prediction model, the prediction accuracy of the cold-dimension prediction model 66 is improved.

[0054] Preferably, the input data for the cold dimension prediction model 66 further includes one or more of the straightening parameters of the roller straightening machine 38. The straightening parameters of the roller straightening machine 38 are, for example, the parameters listed below.

[0055] <Orthodontic parameters> • Roll pitch of the straightening roll • Reduction amount of the straightening roll · Roll axis position

[0056] Since the roller straightening machine 38 is a device that corrects the bending and warping of cooled H-beams 10, the straightening parameters of the roller straightening machine 38 have a significant impact on the cold dimensions of the H-beams 10. For this reason, including the straightening parameters of the roller straightening machine 38 in the input data of the cold dimension prediction model improves the prediction accuracy of the cold dimension prediction model 66.

[0057] Preferably, the input data for the cold dimension prediction model 66 used to predict the cold dimension of the H-shaped steel beam 10 further includes one or more of the attribute parameters of the H-shaped steel beam 10. The attribute parameters of the H-shaped steel beam 10 are, for example, the parameters listed below.

[0058] <Attribute Parameters> • Typical values ​​of target dimensions • Steel grade classification (3 types: 40k steel, 50k steel, and steel greater than 50k) • Chemical composition (content of C, Si, Mn, Cr, Mo, V, etc.) • Target values ​​for mechanical properties (yield stress, tensile strength, elongation, toughness, hardness, etc.)

[0059] Of these attribute parameters, the representative value of the target dimension affects the hot and cold dimensions of the H-shaped steel beam 10. The target values ​​of steel grade, chemical composition, and mechanical properties affect the composition and structure of the oxide scale in the heating furnace 20, and thus affect the deformation resistance during hot rolling. Therefore, these attribute parameters also affect the hot and cold dimensions of the H-shaped steel beam 10. Accordingly, by including the above attribute parameters in the input data of the cold dimension prediction model 66, the prediction accuracy of the cold dimension prediction model 66 is improved.

[0060] Next, the rolling parameter identification process by the rolling parameter identification unit 60 will be described. Figure 4 is a flowchart showing the flow of the rolling parameter identification process. The flow shown in Figure 4 is started, for example, when the process computer 44 has acquired the hot dimensions and hot temperature of the finished-rolled H-shaped steel beam 10.

[0061] First, the data acquisition unit 56 acquires the hot dimensions, hot temperature, and cold temperature of the H-shaped steel beam 10 as input data from the process computer 44 (step S101). The data acquisition unit 56 outputs the acquired input data to the cold dimension prediction unit 58.

[0062] When the cold dimension prediction unit 58 acquires input data, it reads the cold dimension prediction model 66 from the storage unit 54, inputs the input data into the cold dimension prediction model 66, and outputs the cold dimensions of the H-beam 10. As a result, the cold dimension prediction unit 58 predicts the cold dimensions of the H-beam 10 cooled to room temperature (step S102). The cold dimension prediction unit 58 outputs the predicted cold dimensions to the rolling parameter identification unit 60.

[0063] The rolling parameter identification unit 60 determines whether all predicted cold dimensions are within the range of the target dimensions of the H-beam (step S103). The range of the target dimensions of the H-beam may be stored in advance in the storage unit 54, or it may be entered by the operator from the input unit 50.

[0064] The rolling parameter identification unit 60 compares all predicted cold dimensions with the range of the target dimensions for each dimension. If even one of the predicted cold dimensions does not fall within the range of the target dimensions, the rolling parameter identification unit 60 determines that the predicted cold dimensions are not within the range of the target dimensions (Step S103: No).

[0065] If the rolling parameter identification unit 60 determines that the dimensions are not within the range of the target dimensions, it changes the hot dimensions used as input data (step S104). The rolling parameter identification unit 60 changes the hot dimensions at the same position as the cold dimensions that were determined not to satisfy the target dimensions, and executes the process from step S102. In step S103, the rolling parameter identification unit 60 repeatedly executes the processes of steps S104, S102, and S103 until all predicted cold dimensions are within the range of the target dimensions. One unit of hot dimensions that the rolling parameter identification unit 60 changes may be stored in advance in the storage unit 54 for each hot dimension position, or it may be entered by the operator from the input unit 50.

[0066] On the other hand, the rolling parameter identification unit 60 determines that the predicted cold dimensions are within the range of the target dimensions if all predicted cold dimensions are within the range of the target dimensions (Step S103: Yes). The rolling parameter identification unit 60 identifies the hot dimensions used in the input data as hot dimensions that result in the cold dimensions being within the range of the target dimensions (Step S105). If the processing in the first step S103 determines that the cold dimensions are within the range of the target dimensions, there is no need to change the rolling parameters for hot rolling, so the processing in steps S105 to S107 can be skipped.

[0067] The rolling parameter identification unit 60 identifies the rolling parameters for hot rolling that will allow the H-shaped steel beam 10 with the specified hot dimensions to be hot rolled (step S106). Here, the rolling parameters for hot rolling are, for example, the roll spacing in each pass of the roughing mill 22, the intermediate rolling mill 24, and the finishing mill 30. The rolling parameter identification unit 60 reads the hot dimension prediction model 68 from the storage unit 54 and uses the hot dimension prediction model 68 to identify the roll spacing in each pass that will allow the H-shaped steel beam 10 with the specified hot dimensions to be hot rolled.

[0068] The rolling parameter identification unit 60 adjusts the roll spacing of the rolling rolls in each pass of each rolling mill by, for example, 0.1 mm increments to 10 levels each, for both widening and narrowing the roll spacing uniformly across each pass of each rolling mill, centered on the standard rolling conditions. By inputting this combined data into the hot dimension prediction model 68, it outputs predicted values ​​for the hot dimensions. Among the outputted predicted values ​​for the hot dimensions, the roll spacing included in the input data used to output the predicted value for the hot dimension that is the same as, or closest to, the specified hot dimension is identified as the roll spacing that allows rolling of the H-shaped steel beam 10 with the specified hot dimension. In this way, the roll spacing for each pass that allows rolling of the H-shaped steel beam 10 with the specified hot dimension can be identified.

[0069] The rolling parameter identification unit 60 outputs the identified hot rolling parameters to the process computer and reflects the identified hot rolling parameters in the rolling conditions of each rolling mill (step S107), and this process ends. Subsequently, the H-shaped steel beams 10 are manufactured in the rolling equipment 100 by hot rolling under rolling conditions that include the identified hot rolling parameters. This makes it possible to manufacture H-shaped steel beams 10 that are within the target cold dimensions range.

[0070] Next, a method for generating a cold dimension prediction model 66 used to predict the cold dimensions of H-shaped steel beams 10 will be described. The data acquisition unit 56 acquires actual values ​​of hot dimensions, cold dimensions, hot temperatures, and cold temperatures of H-shaped steel beams 10 previously manufactured by the rolling mill 100 from the process computer. The data acquisition unit 56 stores each acquired actual value as a dataset in the database 64 of the storage unit 54. The number of datasets stored in the database 64 is preferably 1000 or more, and more preferably 3000 or more.

[0071] The prediction model generation unit 62 reads a machine learning model pre-stored in the storage unit 54 and uses multiple datasets stored in the database 64 as training data to train the machine learning model and generate a trained machine learning model. This trained machine learning model becomes the cold dimension prediction model 66. The cold dimension prediction model 66 uses a dataset containing actual values ​​of the cold dimensions of structural steel that have undergone volume expansion due to phase transformation as training data. Therefore, the generated cold dimension prediction model 66 is a prediction model that takes into account volume expansion due to phase transformation. The prediction model generation unit 62 stores the generated cold dimension prediction model 66 in the storage unit 54. The machine learning model used in this embodiment may be any of the commonly used neural networks, decision tree learning, random forests, support vector regression, Gaussian processes, and nearest neighbor methods.

[0072] The prediction model generation unit 62 may update the cold dimension prediction model 66 to a new cold dimension prediction model by, for example, training the machine learning model again every month or every year. The data acquisition unit 56 acquires the data set each time an H-shaped steel beam 10 is manufactured in the rolling mill 100 and stores it in the database 64. As a result, the database 64 stores measurement data of newly manufactured H-shaped steel beams 10, so the number of data sets stored in the database 64 increases.

[0073] As the number of datasets and training data increases, the prediction accuracy of the cold dimension prediction model 66 improves, allowing for more accurate prediction of cold dimensions. Furthermore, by using new datasets, the recent state of the rolling equipment 100 is reflected in the cold dimension prediction model 66, so updating the cold dimension prediction model 66 allows for even more accurate prediction of cold dimensions from hot dimensions. The prediction model generation unit 62 can also generate and update a hot dimension prediction model 68 using datasets stored in the database 64, similar to the cold dimension prediction model 66.

[0074] As described above, the cold dimension prediction method used in the cold dimension prediction device 46 of this embodiment predicts cold dimensions from hot dimensions using a cold dimension prediction model 66 that reflects the actual cold dimensions of the steel structure that has undergone volume expansion due to phase transformation. This makes it possible to predict cold dimensions from hot dimensions with higher accuracy than methods that predict cold dimensions without considering volume expansion due to phase transformation.

[0075] The embodiments of the present invention are not limited to the embodiments described above and can be modified in various ways. In this embodiment, the cold dimension prediction method has been described as an example of predicting all cross-sectional dimensions of the H-shaped steel 10 in the cold state, but it is not limited to this. In the cold dimension prediction method according to this embodiment, it is sufficient to predict one or more of the cold twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb. In this case, the input data of the cold dimension prediction model 66 used for predicting cold dimensions should include the cross-sectional dimensions corresponding to the cold dimensions of the steel shape to be predicted. For example, when predicting the cold dimension twb using the cold dimension prediction method, it is sufficient that the input data of the cold dimension prediction model 66 includes at least the hot dimension twb, and other cross-sectional dimensions may also be included.

[0076] Preferably, the hot and cold temperatures included in the input data include the hot and cold temperatures at the locations corresponding to the predicted cross-sectional dimensions. For example, when predicting the cold twb using a cold dimension prediction method, it is preferable that the input data includes the hot and cold temperatures of the web 14.

[0077] In this embodiment, the rolling mill 100 shown in Figure 2 is shown as having a process computer 44 and a cold-working dimension prediction device 46 for structural steel, but it is not limited to this. For example, the process computer 44 may have the function of the cold-working dimension prediction device 46, and these may be configured in a single device.

[0078] In this embodiment, the cold dimension prediction device 46 for structural steel shown in Figure 3 is shown as having a control unit 48 which includes a data acquisition unit 56, a cold dimension prediction unit 58, a rolling parameter identification unit 60, and a prediction model generation unit 62, but it is not limited to this. If the cold dimension prediction method is to be implemented in the cold dimension prediction device 46 for structural steel, the control unit 48 does not need to have a rolling parameter identification unit 60. If the cold dimension prediction model 66 and the hot dimension prediction model 68 are generated using a separate calculation unit and the generated prediction models are stored in the storage unit 54, the control unit 48 does not need to have a prediction model generation unit 62.

[0079] In this embodiment, an example of predicting the cold dimensions of an H-shaped steel beam 10 using the cold dimension prediction device 46 for structural steel beams has been described, but the structural steel beams whose cold dimensions are to be predicted are not limited to H-shaped steel beams 10. The cold dimension prediction device 46 for structural steel beams can similarly predict the cold dimensions of other structural steel beams manufactured by hot rolling in the roughing mill 22, intermediate rolling mill 24, and finishing mill 30, such as channel steel beams, I-beams, angle steel beams, steel sheet piles, or rails. [Examples]

[0080] Next, we will describe an example in which steel billets were hot-rolled into H-beams using the rolling equipment 100 shown in Figure 2, the cold dimensions were predicted from the hot dimensions of the H-beams using a cold dimension prediction model, and 100 H-beams were manufactured while adjusting the rolling conditions using the predicted cold dimensions. In this example, of the 100 H-beams manufactured, the cold dimensions were predicted from the hot dimensions for 99 H-beams, excluding the first H-beam immediately after rearranging the rolling rolls.

[0081] In Invention Example 1, a cold working dimension prediction model was used that takes the following input data as input and outputs cold working dimensions. The cold working dimensions to be output are twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb. In Invention Example 1, 900 H-beams with hwb of 400-900 mm and wfb of 200-400 mm were manufactured in advance, and the dataset obtained from this manufacturing example was used as training data to train a machine learning model and generate a cold working dimension prediction model. The machine learning model used a neural network with 3 hidden layers and 5 nodes each. The sigmoid function was used as the activation function.

[0082] <Input data for Invention Example 1> ·Hot dimensions (twb, tf1b, tf2b, tf3b, tf4b, hwb, wfb) • Hot temperature (web temperature, flange temperature) • Cold temperature (room temperature of the rolling mill at the time of prediction)

[0083] In Invention Example 2, a cold working dimension prediction model was used that takes the following input data as input and outputs cold working dimensions. The cold working dimensions to be output are twb, tf1b, tf2b, tf3b, tf4b, hwb, and wfb. In Invention Example 2, 900 H-beams with hwb of 400-900 mm and wfb of 200-400 mm were manufactured in advance, and the dataset obtained from this manufacturing example was used as training data to train a machine learning model and generate a cold working dimension prediction model. The machine learning model used a neural network with 3 hidden layers and 5 nodes each. The sigmoid function was used as the activation function.

[0084] <Input data for Invention Example 2> ·Hot dimensions (twb, tf1b, tf2b, tf3b, tf4b, hwb, wfb) • Hot temperature (web temperature, flange temperature) • Cold temperature (room temperature of the rolling mill at the time of prediction) • Steel grade classification (3 types: 40k steel, 50k steel, and steel greater than 50k) ·Billette thickness, billet width, billet length, billet weight • Time from extraction from the heating furnace to finish rolling • Rolling loads for each pass in the roughing mill, intermediate rolling mill, and finishing mill

[0085] In Invention Examples 1 and 2, the rolling conditions for hot rolling were adjusted based on the cold dimensions predicted using the hot dimensions. The adjustment of the rolling conditions for hot rolling was performed using a hot dimension prediction model that takes the roll spacing in each pass of the roughing mill, intermediate mill, and finishing mill as input and outputs the hot dimensions. Specifically, the roll spacing in each pass of the hot rolling conditions was adjusted using the following procedure 1 to 3.

[0086] Step 1: Using a cold dimension prediction model, the hot dimensions of the H-beams that fall within the target cold dimension range were identified. Step 2: The roll spacing for each pass was varied by 0.1 mm increments to 10 levels for both the widening and narrowing sides of each pass on each rolling mill, centered around the standard rolling conditions. These combinations were then input as input data into a hot dimension prediction model to output the hot dimensions. Step 3: The roll interval of each pass was adjusted so that it matched the roll interval of each pass included in the input data used to output the predicted value of the hot dimension that was the same as the identified hot dimension.

[0087] On the other hand, in Comparative Example 1, H-beams were manufactured by adjusting the rolling conditions based on measured cold dimensions, without using a predictive model to predict cold dimensions from hot dimensions. In Comparative Example 2, H-beams were manufactured by adjusting the rolling conditions based on measured hot dimensions. The manufacturing results for Invention Examples 1 and 2 and Comparative Examples 1 and 2 are shown in Table 1 below.

[0088] [Table 1]

[0089] In Invention Example 1, out of the 99 structural steel sections evaluated, 85 sections had all cold working dimensions (twb, tf1b, tf2b, tf3b, tf4b, hwb, wfb) within the target range. In Invention Example 2, out of the 99 structural steel sections evaluated, 96 sections had all cold working dimensions within the target range.

[0090] Such results cannot be obtained unless the cold working dimensions of the H-beams can be predicted with high accuracy. Therefore, from the above results, it was confirmed that in Invention Example 1 and Invention Example 2, the cold working dimensions of the H-beams can be predicted with high accuracy, and by adjusting the rolling conditions using these cold working dimensions, it is possible to manufacture H-beams with the target cold working dimensions.

[0091] In Invention Example 2, the input data includes billet thickness, billet width, billet length, billet weight, time from extraction from the heating furnace to finish rolling, and rolling load, resulting in a cold dimension prediction model that takes these variations into account. In reality, billet weight, time from extraction from the heating furnace to finish rolling, and rolling load for each pass varied during the production of 100 H-beams. Therefore, Invention Example 2 was able to predict the cold dimensions while taking the effects of these variations into account, improving the accuracy of cold dimension prediction, and as a result, the number of acceptable pieces increased compared to Invention Example 1.

[0092] On the other hand, in Comparative Example 1, after the first H-beam was rolled, cooled on a cooling bed, straightened with a roller straightening machine, and then its cold dimensions were measured with a cold dimension gauge. At this point, four hours had passed since the second H-beam was rolled, and due to operational concerns that a large number of H-beams that did not meet the target cold dimensions would be produced, the production of H-beams was stopped midway.

[0093] In Comparative Example 2, the rolling conditions were adjusted so that the hot dimensions were within the target range. The adjustment of the rolling conditions was the same as in Procedure 3 of Invention Examples 1 and 2. However, since hot dimensions and cold dimensions are different, adjusting the rolling conditions based on the hot dimensions did not result in the production of H-beams with the target cold dimensions. Of the 99 structural steels evaluated, only 60 had cold dimensions that were entirely within the target range. For structural steels whose cold dimensions were not within the target range, additional dimensional correction processes or remanufacturing were necessary, resulting in a significant increase in the manufacturing cost of the structural steels. [Explanation of Symbols]

[0094] 10 H-beam 12 flanges 14 Web 20 Furnace 22 Roughing mill 24 Intermediate Rolling Mill 26 Intermediate Universal Rolling Mill 28 Edger Rolling Mill 30 Finishing Rolling Mill 32 Hot shape meter 34. Hot thermometer 36 Cooling bed 38 Roller Orthodontic Machine 40 Cold Shape Meter 42 Cold thermometer 44 Process Computers 46 Cold-work dimension prediction device for structural steel 48 Control Unit 50 Input section 52 Output section 54 Storage Unit 56 Data Acquisition Unit 58 Cold Dimension Prediction Section 60 Rolling parameter identification section 62 Prediction Model Generation Unit 64 databases 66 Cold Dimension Prediction Model 68. Hot Dimension Prediction Model 100 Rolling Mills

Claims

1. A method for predicting the cold dimensions of structural steel produced by hot-rolling a steel billet in multiple passes and then cooling it, Input data including the hot dimensions of the steel section, the hot temperature of the steel section, and the cold temperature of the steel section are input into a cold dimension prediction model, and the cold dimensions of the steel section are output to predict the cold dimensions of the steel section. The cold dimensions of the structural steel output from the cold dimension prediction model are one or more of the cross-sectional dimensions of the structural steel. A method for predicting the cold dimensions of a structural steel, wherein the hot dimensions of the structural steel included in the input data include cross-sectional dimensions corresponding to the cold dimensions of the structural steel output from the cold dimension prediction model.

2. The cold dimensional prediction method for a structural steel according to claim 1, wherein the input data includes at least one of the rolling parameters for the hot rolling and at least one of the attribute parameters for the structural steel.

3. The aforementioned steel section is straightened with a roller straightening machine after cooling. The cold-worked dimension prediction method for structural steel according to claim 1, wherein the input data includes one or more of the straightening parameters of the roller straightening machine.

4. The aforementioned steel section is straightened with a roller straightening machine after cooling. The cold-worked dimension prediction method for structural steel according to claim 2, wherein the input data includes one or more of the straightening parameters of the roller straightening machine.

5. A method for manufacturing structural steel by hot-rolling a steel billet in multiple passes and then cooling it, Identify the hot dimensions of the structural steel such that the cold dimensions of the structural steel predicted by the cold dimension prediction method for structural steel according to claim 1 or claim 3 fall within the range of the target dimensions. Using a hot-rolling dimension prediction model that takes input data including one or more of the hot-rolling parameters as input and outputs the hot dimensions of the steel section, the hot-rolling parameters for hot-rolling a steel section of the specified hot dimensions are identified. A method for manufacturing structural steel, comprising hot-rolling the steel billet in multiple passes under rolling conditions that include specified hot-rolling parameters.

6. A method for generating a cold dimension prediction model used to predict the cold dimensions of structural steel produced by hot rolling a steel billet in multiple passes and then cooling it, A machine learning model is trained using multiple datasets, each consisting of a set of actual input data including the hot dimensions, hot temperature, and cold temperature of previously manufactured structural steel sections, and actual cold dimensions of the structural steel sections, as training data. This generates a cold dimension prediction model that takes the input data as input and outputs the cold dimensions of the structural steel sections. The cold dimensions of the structural steel output from the cold dimension prediction model are one or more of the cross-sectional dimensions of the structural steel. A method for generating a cold dimension prediction model, wherein the hot dimensions of the structural steel included in the input data include cross-sectional dimensions corresponding to the cold dimensions of the structural steel output from the cold dimension prediction model.

7. A cold-work dimension prediction device for structural steel produced by hot-rolling a steel billet in multiple passes and then cooling it, The system includes a cold dimension prediction unit that inputs input data including the hot dimensions of the steel section, the hot temperature of the steel section, and the cold temperature of the steel section into a cold dimension prediction model, and outputs the cold dimensions of the steel section to predict the cold dimensions of the steel section. The cold dimensions of the structural steel output from the cold dimension prediction model are one or more of the cross-sectional dimensions of the structural steel. A cold dimension prediction device for structural steel, wherein the hot dimensions of the structural steel included in the input data include cross-sectional dimensions corresponding to the cold dimensions of the structural steel output from the cold dimension prediction model.

8. The cold dimensional prediction device for a structural steel according to claim 7, wherein the input data includes at least one of the rolling parameters for the hot rolling and at least one of the attribute parameters for the structural steel.

9. The aforementioned steel section is straightened with a roller straightening machine after cooling. The cold-worked dimension prediction device for structural steel according to claim 7, wherein the input data includes one or more of the straightening parameters of the roller straightening machine.

10. The aforementioned steel section is straightened with a roller straightening machine after cooling. The cold-worked dimension prediction device for structural steel according to claim 8, wherein the input data includes one or more of the straightening parameters of the roller straightening machine.

11. It has a rolling parameter identification unit that identifies rolling parameters for hot rolling that enable hot rolling of a steel section to be within the target cold dimensions range, The cold dimension prediction device for a structural steel according to claim 7 or 9, wherein the rolling parameter identification unit identifies the hot dimensions of the structural steel such that the cold dimensions predicted by the cold dimension prediction unit are within the range of target dimensions, and uses a hot dimension prediction model that takes input data including one or more of the rolling parameters for hot rolling as input and outputs the hot dimensions of the structural steel as output to identify the rolling parameters for hot rolling that enable hot rolling of the structural steel of the identified hot dimensions.