Metal material quality prediction model generation method, metal material quality prediction method, metal material manufacturing method, quality prediction model generation method, data processing method, metal material quality prediction model generation device, metal material quality prediction device, and data processing device

Abstract

A metal material quality prediction model generation method including a first collection step, a second collection step, a third collection step, a saving step, and a model generation step, wherein in the saving step, prescribed ranges for each work stage which correspond to prescribed ranges for a final work stage are specified on the basis of at least one ore more among longitudinal direction and width direction dimensions of a metal material that change in each work stage, and longitudinal direction and width direction cutting positions of the metal material, collected in the third collection step, and the manufacturing conditions of the metal material in the specified prescribed ranges for each work stage and the quality of the metal material in the final work stage are associated and saved per each specified range in the final work stage.

Description

METALLIC MATERIAL QUALITY PREDICTION MODEL GENERATION METHOD, METALLIC MATERIAL QUALITY PREDICTION METHOD, METALLIC MATERIAL MANUFACTURING METHOD, QUALITY PREDICTION MODEL GENERATION METHOD, DATA PROCESSING METHOD, METALLIC MATERIAL QUALITY PREDICTION MODEL GENERATION DEVICE, METALLIC MATERIAL QUALITY PREDICTION DEVICE AND DATA PROCESSING DEVICE

[0001] The present invention relates to a method for generating a quality prediction model for metallic materials, a method for predicting the quality of metallic materials, a manufacturing method for metallic materials, a quality prediction model generating method, a data processing method, a device for generating a quality prediction model for metallic materials, a device for predicting the quality of metallic materials, and a data processing device.

[0002] Patent Document 1 discloses a method for predicting the quality of a material using a model that has been trained with data that links manufacturing condition data during the manufacturing of the material with quality data in the final process, taking into account the quality judgment position in the longitudinal direction of the material.

[0003] Patent No. 7207547

[0004] It is known that the quality of materials, particularly the mechanical properties of metallic materials such as tensile strength and yield stress, are significantly affected by the temperature of the metallic material and its changes over time. This is because the state of the metal structure varies depending on the temperature of the metallic material and is closely related to the mechanical properties of the metallic material. Therefore, in order to predict the quality of a metallic material with high accuracy, it is necessary to accurately correlate manufacturing condition data, such as the temperature of the metallic material at the position (site) of the metallic material whose quality is to be predicted, with quality data of the metallic material.

[0005] However, in conventional technology, while manufacturing condition data collected in detail in the longitudinal direction, which is the processing direction of the metal material, is utilized, manufacturing condition data collected in detail in the width direction, which is the direction perpendicular to the longitudinal direction, is not effectively utilized.

[0006] Specifically, the prior art did not take into account the cutting off of widthwise ends of the metal material or the division of the metal material at a predetermined, arbitrarily determined width position in each process of manufacturing the metal material. Furthermore, the prior art did not take into account changes in the widthwise length due to plastic deformation caused by rolling the metal material or changes in the widthwise length due to plastic deformation caused by longitudinal tension applied to the metal material in each process. Therefore, the prior art had limitations in improving the accuracy of metal material quality predictions.

[0007] The present invention has been made in consideration of the above, and aims to provide a quality prediction model generation method for metallic materials, a quality prediction method for metallic materials, a manufacturing method for metallic materials, a quality prediction model generation method, a data processing method, a quality prediction model generation device for metallic materials, a quality prediction device for metallic materials, and a data processing device, which are capable of predicting the quality of materials under any manufacturing conditions with high accuracy.

[0008] In order to solve the above-mentioned problems and achieve the object, the method for generating a quality prediction model for metallic materials according to the present invention includes a first collection step of collecting manufacturing conditions for metallic materials in each process within a predetermined range in the longitudinal and width directions of the metallic material, for each predetermined range of the process; a second collection step of evaluating and collecting the quality of the metallic material in the final process of each of the processes within the predetermined range of the final process; a third collection step of collecting at least one of the cutting positions in the longitudinal and width directions of the metallic material in each process and the dimensions of the metallic material in the longitudinal and width directions that change in each process; and The method includes a storing step of storing the quality of the metal material in each specified range of the final process in association with each specified range of the final process, and a model generation step of generating a quality prediction model that predicts the quality for each specified range of the final process from the manufacturing conditions for each specified range of the final process stored in the storing step, wherein the storing step identifies a specified range of each process corresponding to the specified range of the final process based on at least one of the longitudinal and widthwise cutting positions of the metal material and the longitudinal and widthwise dimensions of the metal material that change in each process, collected in the third collection step, and stores the manufacturing conditions of the metal material in the specified range of each process and the quality of the metal material in the final process in association with each specified range of the final process.

[0009] In addition, in the method for generating a quality prediction model for metal materials according to the present invention, in the above invention, the predetermined range of each process in the longitudinal direction of the metal material is determined based on the movement distance of the metal material according to the conveying direction in each process, and the predetermined range of each process in the width direction of the metal material is determined based on a range division distance predetermined in each process.

[0010] In addition, in the method for generating a quality prediction model for metal materials according to the present invention, in the above invention, the third collection step collects cutting positions, as cutting positions in the width direction of the metal material in each process, including truncation of the width direction ends of the metal material in each process and division of the metal material with a predetermined width position as a boundary.

[0011] In addition, in the method for generating a quality prediction model for metallic materials according to the present invention, in the above invention, the third collection step collects dimensions as the widthwise dimensions of the metallic material in each of the processes, including changes in the widthwise length expansion due to plastic deformation caused by rolling the metallic material in each of the processes, and changes in the widthwise length narrowing due to plastic deformation caused in response to the longitudinal tension applied to the metallic material in each of the processes.

[0012] In addition, in the method for generating a quality prediction model for a metallic material according to the present invention, in the above invention, the first collection step collects, for each predetermined range of each process, manufacturing conditions for the metallic material in each process, the manufacturing conditions including at least one of the temperature of the metallic material, the thickness of the metallic material before and after rolling, the tension applied to the metallic material, and the elongation rate of the metallic material.

[0013] Furthermore, in the method for generating a quality prediction model for metallic materials according to the present invention, in the above invention, the model generation step generates the quality prediction model using a statistical analysis method and a machine learning method, including linear regression, local regression, principal component regression, PLS regression, logistic regression, support vector machine, decision tree, regression tree, random forest, gradient boosting tree, and neural network.

[0014] In addition, the quality prediction model generation method for metallic materials according to the present invention, in the above invention, includes a re-storing step of associating the manufacturing conditions of the metallic material in each process collected in the first collection step and the quality of the metallic material in the final process collected in the second collection step for each predetermined range and re-storing them, and a model regeneration step of regenerating a quality prediction model that predicts the quality for each predetermined range from the manufacturing conditions for each predetermined range re-saved in the re-storing step, wherein, among the predetermined ranges, a predetermined range in the longitudinal direction of the metallic material is set to a predetermined range in the longitudinal direction of the metallic material in a process to which a manufacturing condition belongs that has the greatest importance index for quality output for each manufacturing condition that is an input to the quality prediction model generated in the model generation step, and among the predetermined ranges, a predetermined range in the width direction of the metallic material is determined based on a range division distance predetermined in the final process.

[0015] In order to solve the above-mentioned problems and achieve the objectives, the quality prediction method for metal materials of the present invention includes a quality prediction step of predicting the quality of metal materials manufactured under any manufacturing conditions for each specified range of the final process using a quality prediction model generated by the above-mentioned method for generating a quality prediction model for metal materials.

[0016] In order to solve the above-mentioned problems and achieve the objectives, the quality prediction method for metallic materials of the present invention includes a quality prediction step of predicting the quality of metallic materials manufactured under any manufacturing conditions for each specified range using a quality prediction model generated by the above-mentioned method for generating a quality prediction model for metallic materials.

[0017] In order to solve the above-mentioned problems and achieve the objectives, the method for manufacturing a metallic material according to the present invention includes an importance index calculation step for calculating, using a quality prediction model generated by the above-mentioned method for generating a quality prediction model for a metallic material, an importance index indicating the impact on the prediction of output quality for each manufacturing condition that serves as input to the quality prediction model; a manufacturing condition estimation step for estimating manufacturing conditions with high importance indices as manufacturing conditions that are factors that affect the quality of the metallic material; a quality prediction step for predicting output quality using the quality prediction model that includes as input manufacturing conditions that are factors that affect the estimated quality of the metallic material; a manufacturing condition determination step for determining manufacturing conditions that are factors that affect the quality of the metallic material so that the predicted quality falls within a predetermined range; and a metallic material manufacturing step for manufacturing a metallic material based on the determined manufacturing conditions.

[0018] In order to solve the above-mentioned problems and achieve the object, the quality prediction model generation method of the present invention includes a first collection step of collecting production conditions of a material in each process within a predetermined range in the longitudinal and width directions of the material, for each predetermined range of the process; a second collection step of evaluating and collecting the quality of the material in the final process of each of the processes within the predetermined range of the final process; a third collection step of collecting at least one of the cutting positions of the material in the longitudinal and width directions in each process and the dimensions of the material in the longitudinal and width directions that change in each process; and a third collection step of collecting the production conditions of the material in each process collected in the first collection step and the pre-cutting position of the material in the final process collected in the second collection step. and a model generation step of generating a quality prediction model that predicts the quality of the final process for each predetermined range from the manufacturing conditions for each predetermined range of the final process stored in the storing step, wherein the storing step identifies a predetermined range of each process corresponding to the predetermined range of the final process based on at least one of the cutting positions in the longitudinal and width directions of the material and the dimensions of the material in the longitudinal and width directions that change in each process, which are collected in the third collecting step, and stores the manufacturing conditions of the material in the identified predetermined range of each process and the quality of the material in the final process in association with each predetermined range of the final process.

[0019] In order to solve the above-mentioned problems and achieve the object, the data processing method of the present invention includes a first collection step of collecting manufacturing conditions for a metal material in each process within a predetermined range in the longitudinal and width directions of the metal material, for each predetermined range of the process; a second collection step of evaluating and collecting the quality of the metal material in the final process of each of the processes within a predetermined range of the final process; a third collection step of collecting at least one of cutting positions in the longitudinal and width directions of the metal material in each process and dimensions of the metal material in the longitudinal and width directions that change in each process; and The method includes a storage step of storing the manufacturing conditions of the metal material and the quality of the metal material in the final process collected in the second collection step in association with each predetermined range of the final process, wherein the storage step identifies a predetermined range of each process corresponding to the predetermined range of the final process based on at least one of the longitudinal and widthwise cutting positions of the metal material collected in the third collection step and the longitudinal and widthwise dimensions of the metal material that change in each process, and stores the manufacturing conditions of the metal material in the identified predetermined range of each process and the quality of the metal material in the final process in association with each predetermined range of the final process.

[0020] In order to solve the above-mentioned problems and achieve the object, the quality prediction model generation device for metallic materials according to the present invention comprises a first collection unit that collects manufacturing conditions for metallic materials in each process within a predetermined range in the longitudinal and width directions of the metallic material, for each predetermined range of the process; a second collection unit that evaluates and collects the quality of the metallic material in the final process of each of the processes within the predetermined range of the final process; a third collection unit that collects at least one of the cutting positions in the longitudinal and width directions of the metallic material in each process and the longitudinal and width directions dimensions of the metallic material that change in each process; and a third collection unit that collects the manufacturing conditions for the metallic material in each process collected by the first collection unit and the quality of the metallic material in the final process collected by the second collection unit. The system comprises a storage unit that stores the quality of the metal material in each specified range of the final process in association with each specified range of the final process, and a model generation unit that generates a quality prediction model that predicts the quality for each specified range of the final process from the manufacturing conditions for each specified range of the final process stored in the storage unit, wherein the storage unit identifies the specified range of each process corresponding to the specified range of the final process based on at least one of the longitudinal and widthwise cutting positions of the metal material and the longitudinal and widthwise dimensions of the metal material that change in each process, collected by the third collection unit, and stores the manufacturing conditions of the metal material in the identified specified range of each process and the quality of the metal material in the final process in association with each specified range of the final process.

[0021] In addition, in the quality prediction model generation device for metallic materials according to the present invention, in the above invention, the storage unit re-stores the manufacturing conditions of the metallic material in each process collected by the first collection unit and the quality of the metallic material in the final process collected by the second collection unit, in association with each other for each predetermined range, and the model generation unit re-generates a quality prediction model that predicts the quality for each predetermined range from the re-stored manufacturing conditions for each predetermined range, and among the predetermined ranges, the predetermined range in the longitudinal direction of the metallic material is set to be the predetermined range in the longitudinal direction of the metallic material in the process to which the manufacturing conditions that are input to the quality prediction model generated by the model generation unit belong, with the highest importance index for quality output for each manufacturing condition, which is the input to the quality prediction model generated by the model generation unit, and among the predetermined ranges, the predetermined range in the width direction of the metallic material is determined based on a range division distance predetermined in the final process.

[0022] In order to solve the above-mentioned problems and achieve the objectives, the quality prediction device for metal materials according to the present invention includes a quality prediction unit that uses a quality prediction model generated by the above-mentioned metal material quality prediction model generation device to predict the quality of metal materials manufactured under any manufacturing conditions for each specified range of the final process.

[0023] In order to solve the above-mentioned problems and achieve the objectives, the quality prediction device for metal materials according to the present invention includes a quality prediction unit that uses a quality prediction model generated by the above-mentioned metal material quality prediction model generation device to predict the quality of metal materials manufactured under any manufacturing conditions for each specified range.

[0024] In order to solve the above-mentioned problems and achieve the object, the data processing device of the present invention includes a first collection unit that collects manufacturing conditions for a metal material in each process within a predetermined range in the longitudinal and width directions of the metal material, for each predetermined range of the process; a second collection unit that evaluates and collects the quality of the metal material in the final process of each of the processes within a predetermined range of the final process; a third collection unit that collects at least one of cutting positions in the longitudinal and width directions of the metal material in each of the processes and dimensions of the metal material in the longitudinal and width directions that change in each of the processes; and The system is provided with a storage unit that stores the manufacturing conditions of the metal material and the quality of the metal material in the final process collected by the second collection unit in association with each predetermined range of the final process, and the storage unit identifies a predetermined range of each process corresponding to the predetermined range of the final process based on at least one of the longitudinal and widthwise cutting positions of the metal material collected by the third collection unit and the longitudinal and widthwise dimensions of the metal material that change in each process, and stores the manufacturing conditions of the metal material in the identified predetermined range of each process and the quality of the metal material in the final process in association with each predetermined range of the final process.

[0025] According to the present invention, by dividing a metal material into predetermined ranges in the width direction as well as the longitudinal direction, it is possible to create training data that takes into account variations in manufacturing conditions across the width direction of the metal material. This allows differences in manufacturing conditions across the width direction of the metal material to be appropriately reflected in the quality prediction model, even at the same position in the longitudinal direction, making it possible to generate a highly accurate quality prediction model.

[0026] FIG. 1 is a block diagram showing an example of the configuration of an information processing device functioning as a quality prediction model generation device, a quality prediction device, and a data processing device for metallic materials according to an embodiment of the present invention. FIG. 2 is a flowchart showing the flow of a quality prediction model generation method, a quality prediction method, and a data processing method for metallic materials according to a first embodiment of the present invention. FIG. 3 is a diagram showing an example of manufacturing condition data for each predetermined range in the longitudinal and width directions of a metallic material collected by a manufacturing condition data collection unit in the quality prediction model generation method, a quality prediction method, and a data processing method for metallic materials according to an embodiment of the present invention. FIG. 4 is a diagram showing an example of quality data for each predetermined range in the longitudinal and width directions of a metallic material collected by a quality data collection unit in the quality prediction model generation method, a quality prediction method, and a data processing method for metallic materials according to an embodiment of the present invention. FIG. 5 is a diagram showing an example of a case where a metallic material is manufactured through multiple processes in the quality prediction model generation method, a quality prediction method, and a data processing method for metallic materials according to an embodiment of the present invention. FIG. 6 is a diagram showing an example of the transition of a metallic material through each process in the quality prediction model generation method, a quality prediction method, and a data processing method for metallic materials according to an embodiment of the present invention. 7 is a diagram showing an example of learning data edited by an integrated process data editing unit in the quality prediction model generation method, quality prediction method, and data processing method for metallic materials according to an embodiment of the present invention. FIG. 8 is a flowchart showing the flow of the quality prediction model generation method and quality prediction method for metallic materials according to a second embodiment of the present invention.

[0027] A method for generating a quality prediction model for metal materials, a method for predicting the quality of metal materials, a quality prediction model generation method, a data processing method, a device for generating a quality prediction model for metal materials, a device for predicting the quality of metal materials, and a data processing device according to embodiments of the present invention will be described with reference to the drawings.

[0028] (Device Configuration) The configurations of a quality prediction model generation device, a quality prediction device, and a data processing device for metallic materials according to this embodiment will be described with reference to FIG. 1 . The quality prediction model generation device is a device for generating a quality prediction model for predicting the quality of metallic materials manufactured through one or more processes. The quality prediction device is a device for predicting the quality of metallic materials manufactured through one or more processes using the quality prediction model. The data processing device is a device for creating learning data for generating the quality prediction model. Note that metallic materials in this embodiment include, for example, steel products, such as semi-finished products such as slabs, and products such as steel plates manufactured by rolling these slabs.

[0029] In the following, we will explain the quality prediction model generation device, quality prediction device, and data processing device for metal materials. However, the devices can also be applied to materials other than metal materials, such as materials that are manufactured through multiple processes on a continuous production line, and other manufactured products.

[0030] The quality prediction model generation device, the quality prediction device, and the data processing device can be realized by, for example, an information processing device 1 as shown in Fig. 1. Specifically, the information processing device 1 is configured by a personal computer, a workstation, or the like. The information processing device 1 has as its main components a processor such as a CPU (Central Processing Unit) and a memory (main storage unit) such as a RAM (Random Access Memory) and a ROM (Read Only Memory).

[0031] The information processing device 1 includes a manufacturing condition data collection unit 11, a quality data collection unit 12, a cutting performance data collection unit 13, a dimensional performance data collection unit 14, an integrated process data editing unit 15, an integrated process database 16, a model generation unit 17, a quality prediction unit 18, and an influencing factor estimation unit 19. The quality prediction model generation device for metallic materials according to the embodiment is composed of the elements of the information processing device 1 excluding the quality prediction unit 18 and the influencing factor estimation unit 19. The quality prediction device for metallic materials according to the embodiment is composed of the elements of the information processing device 1 excluding the influencing factor estimation unit 19. The data processing device according to the embodiment is composed of the elements of the information processing device 1 excluding the model generation unit 17, the quality prediction unit 18, and the influencing factor estimation unit 19.

[0032] A sensor (not shown) is connected to the manufacturing condition data collection unit 11. The manufacturing condition data collection unit 11 uses this sensor to collect data on the manufacturing conditions of each process (hereinafter referred to as "manufacturing condition data") and outputs the data to the integrated process data editing unit 15. Examples of the "manufacturing conditions of metal material" include the components of the metal material in each process, the temperature of the metal material, the pressure applied to the metal material, the tension applied to the metal material, the thickness of the metal material before and after rolling, and the threading speed of the metal material.

[0033] The manufacturing condition data for each process collected by the manufacturing condition data collection unit 11 includes not only actual values ​​of the manufacturing conditions measured by sensors but also preset values ​​of the manufacturing conditions. That is, since there are cases where sensors are not installed in some processes, in such cases, the set values ​​are collected as manufacturing condition data instead of actual values.

[0034] The manufacturing condition data collection unit 11 collects manufacturing condition data for the metal material in each process, which is a predetermined range in the longitudinal and width directions of the metal material, for each predetermined range of each process. For example, if the metal material is a slab or a steel plate, the "predetermined range of each process" refers to a certain range (position) in the longitudinal and width directions of the metal material in each process. Furthermore, the predetermined range of each process may be different for each process or may be the same for each process.

[0035] Of the predetermined ranges for each process, the predetermined range in the longitudinal direction of the metal material is determined, for example, based on the movement distance of the metal material in the conveying direction in each process. Furthermore, of the predetermined ranges for each process, the predetermined range in the width direction of the metal material is determined, for example, based on a range division distance predetermined for each process. The predetermined range in the width direction of the metal material is determined, for example, as the range division distance, i.e., the predetermined range in the width direction, using the highest resolution (finest data granularity) of the resolutions of each sensor that comprehensively collects manufacturing condition data for the metal material in the width direction. For example, if the minimum resolution of each sensor is 10 mm, this 10 mm is set as the predetermined range in the width direction.

[0036] 1, it is assumed that only one manufacturing condition data collection unit 11 is provided, and that manufacturing condition data for each process is collected by this single manufacturing condition data collection unit 11. However, unlike the configuration shown in FIG. 1, for example, a plurality of manufacturing condition data collection units 11 may be provided in accordance with the number of processes, and the manufacturing condition data for each process may be collected by a separate manufacturing condition data collection unit 11.

[0037] A sensor (not shown) or a data input terminal (not shown) is connected to the quality data collection unit 12. The quality data collection unit 12 collects data on the quality of the metal material manufactured through each process (hereinafter referred to as "quality data") using this sensor or data input terminal, and outputs the data to the integrated process data editing unit 15. Examples of the "quality of the metal material" include the tensile strength of the metal material in the final process, the defect contamination rate (the number of defects appearing per unit area) of the metal material in the final process, etc.

[0038] The quality data of metal materials collected by the quality data collection unit 12 includes not only actual quality values ​​measured by sensors, but also quality judgment values ​​entered from a data input terminal, where pass / fail is evaluated based on certain standards based on the actual measurement values ​​or by visual inspection by an inspector, etc.

[0039] The quality data collection unit 12 evaluates and collects quality data of the metal material in the final process of each process for each predetermined range of the final process. For example, if the metal material is a slab or a steel plate, the "predetermined range of the final process" refers to a certain range (position) in the longitudinal and width directions of the metal material in the final process.

[0040] Of the predetermined ranges for the final process, the predetermined range in the longitudinal direction of the metal material is determined, for example, based on the movement distance of the metal material in the conveying direction in the final process. Furthermore, of the predetermined ranges for the final process, the predetermined range in the width direction of the metal material is determined, for example, based on a range division distance predetermined for the final process. The predetermined range in the width direction of the metal material is determined, for example, as the range division distance, i.e., the predetermined range in the width direction, using the highest resolution (finest data granularity) of the resolutions of each sensor that comprehensively collects manufacturing condition data for the metal material in the width direction. For example, if the minimum resolution of each sensor is 10 mm, this 10 mm is set as the predetermined range in the width direction.

[0041] The cutting history data collection unit 13 collects cutting positions in the longitudinal and width directions of the metal material in each process. Cutting machines (not shown) for cutting the leading and trailing ends of the metal material in the longitudinal direction and at predetermined arbitrary positions in the longitudinal direction in each process are connected to the cutting history data collection unit 13. Also connected to the cutting history data collection unit 13 are cutting machines (not shown) for cutting the ends of the metal material in the width direction and at predetermined arbitrary cutting positions in the width direction.

[0042] The cutting performance data collection unit 13 collects, for each metal material, the cutting positions in the width direction of the metal material and the number of cuts (hereinafter referred to as "cutting positions, etc.") in each process using these cutting machines. The "cutting positions of the metal material" refer to the distance from the tip of the metal material when cutting in the longitudinal direction and the distance from the end of the metal material when cutting in the width direction. In addition, the "cutting positions of the metal material" include, for example, the truncation of the end of the metal material in the width direction in each process and the division of the metal material at a predetermined width position as a boundary.

[0043] The cutting performance data collection unit 13 outputs performance data relating to the cutting position of the metal material, etc. to the integrated process data editing unit. As with the manufacturing condition data collection unit 11, only one cutting performance data collection unit 13 may be provided, or multiple cutting performance data collection units 13 may be provided in accordance with the number of processes.

[0044] The actual dimension data collection unit 14 collects the longitudinal and width dimensions of the metal material that change in each process. A sensor (not shown) for collecting the longitudinal dimension of the shape that changes due to processing of the metal material in each process and a sensor (not shown) for collecting the width dimension of the shape that changes due to processing of the metal material are connected to the actual dimension data collection unit 14.

[0045] The actual dimension data collection unit 14 collects the widthwise dimension of the metal material in each process for each metal material through these sensors. The "widthwise dimension of the metal material" includes, for example, changes in the widthwise length due to plastic deformation caused by rolling the metal material in each process, and changes in the widthwise length due to plastic deformation caused by the longitudinal tension applied to the metal material in each process.

[0046] The dimension actual data collection unit 14 outputs actual data relating to the dimensions of the metal material to the integrated process data editing unit 15. As with the manufacturing condition data collection unit 11, only one dimension actual data collection unit 14 may be provided, or multiple units may be provided in accordance with the number of processes.

[0047] The integrated process data editing unit 15 edits the performance data input from the manufacturing condition data collecting unit 11, the quality data collecting unit 12, the cutting performance data collecting unit 13, and the dimension performance data collecting unit 14. Specifically, the integrated process data editing unit 15 associates the manufacturing condition data of each process collected by the manufacturing condition data collecting unit 11 with the quality data of the metal material collected by the quality data collecting unit 12 for each predetermined range of the final process, and stores the data in the integrated process database 16.

[0048] First, the integrated process data editing unit 15 identifies the predetermined range of each process corresponding to the predetermined range of the final process based on at least one of the cutting positions in the longitudinal and width directions of the metal material and the longitudinal and width dimensions of the metal material that change in each process. Note that the "cutting positions in the longitudinal and width directions of the metal material" are collected by the cutting performance data collecting unit 13. Also, the "longitudinal and width dimensions of the metal material" are collected by the dimension performance data collecting unit 14. Also, the "predetermined range of each process corresponding to the predetermined range of the final process" refers to the predetermined range of each process that corresponds to the same position as the predetermined range of the final process.

[0049] In the production of metal materials, the longitudinal and width dimensions of the metal material change as it passes through each process, as shown in FIG. 6 (described later). Therefore, the integrated process data editing unit 15 traces and identifies the predetermined ranges of each process that correspond to the predetermined range of the final process, tracing back to processes prior to the final process. The integrated process data editing unit 15 then associates the manufacturing condition data of the metal material within the identified predetermined ranges of each process with the quality data of the metal material in the final process in the integrated process database 16, with each predetermined range of the final process. The integrated process data editing unit 15 then stores the manufacturing condition data and quality data of the metal material in the integrated process database 16 in a manner that allows the cutting position and dimensional data of the metal material in each process to be distinguished.

[0050] The model generation unit 17 generates a quality prediction model that predicts the quality of each predetermined range of the final process of the metal material from the manufacturing condition data and quality data for each predetermined range of the final process stored in the integrated process database 16. The model generation unit 17 uses, for example, a gradient boosting tree as a statistical analysis method and a machine learning method when generating the quality prediction model. Note that, in addition to the above, various other statistical analysis methods and machine learning methods can also be used, such as linear regression, local regression, principal component regression, PLS regression, logistic regression, support vector machine, decision tree, regression tree, random forest, and neural network.

[0051] The quality prediction unit 18 predicts the quality of a metal material manufactured under any manufacturing conditions for each predetermined range in the final process using the quality prediction model generated by the model generation unit 17. For example, if the metal material to be predicted is a slab, the conventional method predicts the quality for each predetermined range in the longitudinal direction of the slab, but in this embodiment, it is possible to predict the quality for each predetermined range in the longitudinal and width directions of the slab.

[0052] The influencing factor estimation unit 19 uses the quality prediction model generated by the model generation unit 17 to calculate the variable importance (importance index) of the quality prediction model for each manufacturing condition, and estimates manufacturing conditions with high variable importance as factors that affect the quality of the metal material. Furthermore, the influencing factor estimation unit 19 may predict output quality using the quality prediction model generated by the model generation unit 17, which includes, as input, manufacturing conditions estimated as factors that affect the quality of the metal material. The influencing factor estimation unit 19 may then determine a control range for the manufacturing conditions estimated as factors that affect the quality of the metal material so that the predicted quality falls within a predetermined range. The control range for the manufacturing conditions can be determined, for example, by a mathematical programming method such as a branch-and-bound method.

[0053] The information processing device 1 may further include an output unit that outputs, as an output signal, the control range of the manufacturing conditions estimated as factors affecting the quality of the metal material, as determined by the influencing factor estimation unit 19. If the output signal is output to an equipment control device, the output signal may be a control signal for the equipment that sets the manufacturing conditions. Furthermore, if the output signal is output to an operator's operation terminal, the output signal may be operator guidance information that is displayed as guidance on a screen. Then, the manufacturing conditions are adjusted so that the manufacturing conditions estimated as factors affecting the quality of the metal material fall within the determined control range, and the metal material is manufactured.

[0054] <First Embodiment> (Quality Prediction Method) A quality prediction method, a quality prediction model generation method, and a data processing method according to a first embodiment will be described with reference to Figures 2 to 7. The quality prediction method according to this embodiment performs the processing of steps S1 to S6 shown in Figure 2. The quality prediction model generation method according to this embodiment performs the processing of steps S1 to S5 excluding step S6 shown in Figure 2. The data processing method according to this embodiment performs the processing of steps S1 to S4 excluding steps S5 and S6 shown in Figure 2.

[0055] 2, step S1 corresponds to the first collection step, step S2 corresponds to the second collection step, step S3 corresponds to the third and fourth collection steps, step S4 corresponds to the storage step, step S5 corresponds to the model generation step, and step S6 corresponds to the prediction step.

[0056] First, the manufacturing condition data collection unit 11 collects data on the manufacturing conditions of each process (step S1). In step S1, the manufacturing condition data collection unit 11 collects manufacturing condition data for each process for each metal material. The manufacturing condition data collected by the manufacturing condition data collection unit 11 is data in which actual values ​​or set values ​​of multiple manufacturing conditions are listed for each position from the leading end to the trailing end in the longitudinal direction of the metal material in each process and for each position from one end to the other in the width direction, as shown in Fig. 3, for example.

[0057] The manufacturing condition data shown in FIG. １ , l ２ , ... and the widthwise position w in each process １ , w ２ , ... and a plurality of manufacturing conditions x measured by the sensor at the position １ １１ , x １ １２ , ..., x ２ ２１ , x ２ ２２ , . . . .

[0058] Next, the quality data collection unit 12 collects quality data of the metal material manufactured through each process (step S2). The quality data collected by the quality data collection unit 12 is data in which actual measured values ​​or judged values ​​of quality are arranged for each position from the leading end to the trailing end in the longitudinal direction and for each position from one end to the other in the width direction of the metal material manufactured through each process, as shown in Fig. 4, for example.

[0059] The performance data shown in FIG. 4 is the longitudinal position l １ , l ２ , ... and the widthwise position w １ , w ２ , ... and multiple qualities y １１ , y １２ , . . . .

[0060] Next, the cutting performance data collection unit 13 and the dimension performance data collection unit 14 collect performance data on the cutting positions and dimensions of the metal material in each process (step S3). Next, the integrated process data editing unit 15 aligns and combines the performance data of all processes in units of position in the longitudinal direction and position in the width direction of the metal material (step S4).

[0061] The integrated process data editing unit 15 aligns and combines the multiple manufacturing conditions and quality data for the metal material for all processes based on the performance data related to the longitudinal position, width position, cutting position, etc. of the metal material and dimensions, for each predetermined range of the final process. Note that "for each predetermined range of the final process" refers to the positional units of the metal material in the longitudinal and width directions at the exit side of the final process.

[0062] In this way, the integrated process data editing unit 15 associates the manufacturing condition data for each process with the quality data of the metal material manufactured under these manufacturing conditions for each predetermined range of the final process in the longitudinal and width directions of the metal material, and stores the data in the integrated process database 16. An example of processing by the integrated process data editing unit 15 will be described below with reference to Figures 5 to 7.

[0063] For example, consider the case of manufacturing a metal material (material) through steps 1, 2, and 3, as shown in FIG. 5 . Steps 1, 2, and 3 are, for example, rolling steps, and with each rolling step, the material undergoes plastic deformation, increasing its length in the longitudinal direction and its length in the width direction. Furthermore, when tension is applied to the material in the longitudinal direction, the plastic deformation caused by the tension narrows its width direction. Also, as shown in FIG. 5 , when moving from step 1 to step 2, material A is divided longitudinally into material A1 and material A2, and when moving from step 2 to step 3, material A1 is divided widthwise into material A11 and material A12. Note that step 3 is, for example, the final step.

[0064] 6 shows an image of the material in each process, focusing on part B in FIG. 5. For example, material A in process 1 is divided by the manufacturing condition data collection unit 11 into 50 mm intervals from the leading end to the trailing end in the longitudinal direction and 20 mm intervals from one end to the other in the width direction, for example, X １ １ ~X １ Ｍ１ The performance data for M1 items is collected.

[0065] Furthermore, for material A, the leading end portion from 0 mm (leading end) to 250 mm in the longitudinal direction is truncated, and performance data is collected for material A1 from 250 mm to 3300 mm in the longitudinal direction by the cutting performance data collection unit 13. Furthermore, for material A, the leading end portion from 3300 mm to 4950 mm in the longitudinal direction is truncated, and performance data is collected for material A2 from 4950 mm to 5300 mm (tail end) in the longitudinal direction.

[0066] Furthermore, for material A, the actual dimension data collection unit 14 collects actual data on the longitudinal dimension of 5300 mm and the width dimension of 920 mm of material A before cutting, and the longitudinal dimension of 3050 mm and the width dimension of 920 mm of material A1 after cutting material A. Furthermore, for material A, the actual dimension data collection unit 14 collects actual data on the longitudinal dimension of 1650 mm and the width dimension of 920 mm of material A2 after cutting material A.

[0067] Next, the material A1 in the process 2 is divided by the manufacturing condition data collection unit 11 into 100 mm intervals from the leading end to the trailing end in the longitudinal direction and 10 mm intervals from one end to the other in the width direction, for example, X ２ １ ~X ２ Ｍ２ Performance data for M2 items is collected.

[0068] Furthermore, for material A1, one end portion from 0 mm (one end) to 20 mm in the width direction is truncated by the cutting performance data collection unit 13, and performance data is collected for material A11 from 20 mm to 500 mm in the width direction. Furthermore, for material A1, material A12 is taken from 500 mm to 940 mm in the width direction by the cutting performance data collection unit 13, and performance data is collected for material A12 from 940 mm to 960 mm (the other end) in the width direction is truncated.

[0069] Furthermore, for material A1, the actual dimension data collection unit 14 collects actual data on the longitudinal dimension of 68,000 mm and the width dimension of 960 mm before cutting, which have changed due to rolling of material A1. Furthermore, for material A1, the actual dimension data collection unit 14 collects actual data on the longitudinal dimension of 68,000 mm and the width dimension of 480 mm of material A11 after cutting of material A1. Furthermore, for material A1, the actual dimension data collection unit 14 collects actual data on the longitudinal dimension of 68,000 mm and the width dimension of 440 mm of material A12 after cutting of material A1.

[0070] Next, the material A11 in process 3 is divided into 500 mm pieces, for example, X pieces, by the manufacturing condition data collection unit 11 and the quality data collection unit 12, from the leading end to the trailing end in the longitudinal direction and from one end to the other end in the width direction. ３ １ ~X ３ Ｍ３ Performance data for M3 items is collected.

[0071] Furthermore, the dimension performance data collection unit 14 collects performance data on the longitudinal dimension of 136,000 mm and the width dimension of 500 mm before cutting, which are changes caused by rolling the material A11.

[0072] The integrated process data editing unit 15 combines the performance data of multiple manufacturing conditions for all processes, which are collected in detail in the longitudinal and width directions by sensors (not shown), for each predetermined range of the final process, while taking into account the performance data of the cutting position and dimensions of the metal material in each process. That is, the integrated process data editing unit 15 combines the performance data of multiple manufacturing conditions for each process in units of the longitudinal position and width position of the metal material in the final process.

[0073] As a result, as shown by the dashed lines in Figure 6, the longitudinal and width directions of the material in processes 1 and 2 are scaled to match the longitudinal and width directions of the material in process 3, the final process. The integrated process data editing unit 15 then identifies the cutting position of each metal material, taking into account the cutting position and dimensions of the metal material in each process. As a result, the integrated process data editing unit 15 identifies the predetermined range of each process in the metal material that corresponds to the predetermined range of the final process. Next, the integrated process data editing unit 15 associates multiple manufacturing conditions and quality performance data for all processes with each predetermined range of the final process of the metal material, and stores the associated data in the integrated process database 16.

[0074] For example, in Figure 6, the shaded area where material A11 was taken in process 3, the final process, is identified by tracing back to material A1 in process 2 and material A in process 1, and this process is repeated for all metal materials. As a result, the integrated process data editing unit 15 creates performance data in which multiple production condition and quality performance data for metal materials in all processes are aligned and combined in units of position in the longitudinal direction and position in the width direction of the metal material, as shown in Figure 7. Below, we will continue the explanation by returning to Figure 2.

[0075] The model generation unit 17 generates a quality prediction model for predicting the quality of the metallic material for each predetermined range in the final process from the manufacturing condition data and quality data for each predetermined range in the final process (step S5). Subsequently, the quality prediction unit 18 predicts the quality of the metallic material manufactured under any manufacturing conditions for each predetermined range in the final process using the quality prediction model generated by the model generation unit 17 (step S6).

[0076] In this embodiment, quality prediction is performed in step S6 after generating a quality prediction model in step S5, but it is not necessary to perform all steps up to step S5 every time step S6 is performed. For example, steps S1 to S5 may be performed in advance to generate a quality prediction model, and then only steps S1, S3, S4, and S6 may be performed every time step S6 is performed using the generated quality prediction model.

[0077] According to the above-described embodiments of the method for generating a quality prediction model for a metal material, the method for predicting quality of a metal material, the data processing method, the device for generating a quality prediction model for a metal material, and the data processing device, by dividing a predetermined range in the width direction as well as the longitudinal direction of the metal material, it is possible to generate training data that takes into account variations in manufacturing conditions across the width direction of the metal material. This makes it possible to appropriately reflect differences in manufacturing conditions across the width of the metal material in the quality prediction model, even for the same position in the longitudinal direction, and therefore to generate a highly accurate quality prediction model.

[0078] Furthermore, according to the method for generating a quality prediction model for metal materials, the method for predicting quality of metal materials, the quality prediction model generation method, the data processing method, the device for generating a quality prediction model for metal materials, the device for predicting quality of metal materials, and the data processing device according to the embodiments, by generating a quality prediction model that associates the manufacturing conditions of each process with the quality of the metal material manufactured under these manufacturing conditions for each specified range of the final process, it is possible to predict the quality of the metal material for any manufacturing conditions with higher accuracy than conventional methods.

[0079] That is, in the metallic material quality prediction model generation method, metallic material quality prediction method, quality prediction model generation method, data processing method, metallic material quality prediction model generation device, metallic material quality prediction device, and data processing device according to the embodiments, actual data on multiple manufacturing conditions and quality for all processes are aligned and combined in units of position in the longitudinal direction and position in the width direction of the metallic material at the exit side of the final process, taking into account the cutting position and dimensions of the metallic material in each process. Therefore, quality prediction is made by effectively utilizing actual data on multiple manufacturing conditions for all processes that is collected in detail by sensors in the longitudinal and width directions of the metallic material, thereby enabling quality predictions to be made with higher accuracy than before.

[0080] Second Embodiment A method for generating a quality prediction model for a metallic material, a method for predicting the quality of a metallic material, and a method for manufacturing a metallic material according to a second embodiment of the present invention will be described with reference to the drawings.

[0081] The first embodiment is based on the premise that the manufacturing condition data for each process in the production of a metal material and the quality data for the final process are aligned in the unit of length of the metal material for the quality data and manufacturing condition data for the final process for evaluating quality. Note that the "unit of length of the metal material for the quality data and manufacturing condition data for the final process" refers to a predetermined range in the longitudinal direction of the metal material within a predetermined range of the final process. In contrast, the second embodiment determines the coarseness of the data (unit of length of the metal material) for aligning and associating the manufacturing condition data and quality data for each process, including the final process, based on an importance index (variable importance) that indicates the influence on the prediction of the quality of the final process.

[0082] When each process is a rolling process, the length of the metal material increases, mainly in the longitudinal direction, as it passes through each process. If the length units of the metal material in the final process are aligned, the length of the metal material in the corresponding intermediate process will become shorter as the process progresses. If the measurement period is shorter than the measurement period for the actual manufacturing condition data of the intermediate process, even if data interpolation (e.g., linear interpolation) is performed, there is a possibility that fluctuations in the actual manufacturing conditions will not be accurately reflected.

[0083] In particular, when manufacturing conditions change rapidly or fluctuate nonlinearly, linear interpolation may not be able to capture the fluctuations, and there is a risk that the prediction accuracy of the quality prediction model will decrease. Furthermore, if the number of data points for each process increases due to interpolation, the quality prediction model will overfit, causing a decrease in generalization performance.

[0084] Therefore, in the second embodiment, the manufacturing condition data for intermediate processes, which can be a major cause of quality defects, is aligned with the longitudinal roughness of the metal material, and the manufacturing conditions for each of the other processes and the quality of the final process are correlated. This makes it possible to reflect the maximum information regarding the target quality in the generated quality prediction model. As a result, the quality of a material for any manufacturing condition can be predicted with higher accuracy.

[0085] (Quality Prediction Method) A quality prediction method and a quality prediction model generation method according to this embodiment will be described with reference to Fig. 8. The quality prediction model generation method according to this embodiment performs the processes of steps S11 to S18 shown in Fig. 8. Furthermore, the quality prediction method according to this embodiment performs the processes of steps S11 to S19 shown in Fig. 8. Furthermore, in Fig. 8, steps S11 to S15 perform the same processes as steps S1 to S5 in Fig. 2, so their description will be omitted, and steps S16 to S19 will be described below.

[0086] First, the influencing factor estimation unit 19 calculates, for each manufacturing condition, an importance index (variable importance) that indicates the influence that the manufacturing condition that is the input of each quality prediction model has on the prediction of quality that is the output, using the quality prediction model generated by the model generation unit 17. Then, the influencing factor estimation unit 19 estimates, from among the calculated manufacturing conditions, the manufacturing condition that has the largest importance index (step S16).

[0087] The influencing factor estimation unit 19 calculates, as an importance index, an index corresponding to the influence of each factor on the entire quality prediction model. The importance index may be calculated using an algorithm that calculates an importance index specific to the algorithm that created the quality prediction model, such as importance based on Gini impurity in decision tree algorithms such as random forest. Alternatively, the importance index may be calculated using a known method that calculates an importance index universally regardless of the model creation algorithm, such as "Permutation Importance."

[0088] Next, the integrated process data editing unit 15 associates the manufacturing condition data for each process collected by the manufacturing condition data collection unit 11 with the quality data of the metal material collected by the quality data collection unit 12 for each specified range based on the importance index of the manufacturing conditions calculated by the influence factor estimation unit 19, and re-stores the data in the integrated process database 16 (step S17).

[0089] Specifically, the integrated process data editing unit 15 determines the predetermined range in the longitudinal direction of the metal material within the predetermined range in which the performance data of all processes are collected and combined as follows: That is, the predetermined range in the longitudinal direction of the metal material is determined to be the predetermined range in the longitudinal direction of the metal material in the process to which the manufacturing conditions that are input to the quality prediction model generated in step S15 belong, which have the greatest importance index for quality (variable importance).

[0090] Furthermore, the integrated process data editing unit 15 determines a predetermined range in the width direction of the metal material within the predetermined range for aligning and combining the performance data of all processes based on the range division distance determined in advance for the final process.The integrated process data editing unit 15 then aligns and recombines the performance data of all processes in position units in the longitudinal direction and width direction of the metal material.

[0091] Next, the model generation unit 17 regenerates a quality prediction model that predicts the quality of a metallic material for each predetermined range based on the importance index of the metallic material from the re-saved manufacturing condition data and quality data for each predetermined range based on the importance index of the manufacturing condition with respect to the quality of the metallic material (step S18). Next, the quality prediction unit 18 predicts the quality of a metallic material manufactured under any manufacturing conditions for each predetermined range based on the importance index of the manufacturing condition with respect to the quality of the metallic material using the quality prediction model generated by the model generation unit 17 in step S18 (step S19).

[0092] According to the method for generating a quality prediction model for metallic materials and the method for predicting the quality of metallic materials according to the second embodiment described above, by aligning the manufacturing condition data for intermediate processes, which can be a major cause of quality defects, with the longitudinal roughness of the metallic material, and by relating the manufacturing conditions for each of the other processes and the quality of the final process, it is possible to reflect the maximum information related to the target quality in the generated quality prediction model. Therefore, the quality of a material for any manufacturing condition can be predicted with higher accuracy.

[0093] (Method for Manufacturing Metallic Material) The method for manufacturing a metallic material according to this embodiment includes an importance index calculation step, a manufacturing condition estimation step, a quality prediction step, a manufacturing condition determination step, and a metallic material manufacturing step.

[0094] In the importance index calculation step, the influence factor estimation unit 19 uses the quality prediction model generated by the model generation unit 17 to calculate an importance index (variable importance) that indicates the influence that each manufacturing condition that is the input of the quality prediction model has on the prediction of the output quality.

[0095] In the manufacturing condition estimation step, the influencing factor estimation unit 19 estimates manufacturing conditions with high importance indices as manufacturing conditions that are factors that affect the quality of the metal material.

[0096] In the quality prediction step, the influencing factor estimation unit 19 predicts the output quality using the quality prediction model generated by the model generation unit 17, which includes as input manufacturing conditions that are factors that affect the quality of the estimated metal material.

[0097] In the manufacturing condition determination step, the influencing factor estimation unit 19 determines manufacturing conditions that are factors that affect the quality of the metallic material so that the predicted quality falls within a predetermined range. That is, the influencing factor estimation unit 19 determines the control range of the manufacturing conditions that are estimated as factors that affect the quality of the metallic material so that the predicted quality falls within the predetermined range. The control range of the manufacturing conditions can be obtained by mathematical programming such as the branch and bound method.

[0098] The influencing factor estimation unit 19 may output the control range of the manufacturing conditions estimated as factors that affect the quality of the metal material as an output signal via the output unit of the information processing device 1. If the output signal is output to a control device of the equipment, the output signal may be a control signal of the equipment that sets the manufacturing conditions. Furthermore, if the output signal is output to an operation terminal of an operator, the output signal may be operator guidance information to be displayed as guidance on the screen.

[0099] In the metal material manufacturing step, the metal material is manufactured based on manufacturing conditions that are adjusted so that the manufacturing conditions estimated to be factors that affect the quality of the metal material are within the determined control range.

[0100] The method for generating a quality prediction model for metallic materials, the method for predicting quality of metallic materials, the method for manufacturing metallic materials, the quality prediction model generating method, the data processing method, the device for generating a quality prediction model for metallic materials, the device for predicting quality of metallic materials, and the data processing device according to the present invention have been specifically described above using a description of the preferred embodiment and examples, but the scope of the present invention is not limited to these descriptions and should be broadly interpreted based on the claims. Furthermore, it goes without saying that various changes and modifications based on these descriptions are also included in the scope of the present invention.

[0101] REFERENCE SIGNS LIST 1 Information processing device 11 Manufacturing condition data collection unit 12 Quality data collection unit 13 Cutting performance data collection unit 14 Dimension performance data collection unit 15 Integrated process data editing unit 16 Integrated process database 17 Model generation unit 18 Quality prediction unit 19 Influence factor estimation unit

Claims

1. A method for manufacturing a metal material comprising: a first collection step of collecting manufacturing conditions for each process within a predetermined range in the longitudinal and width directions of the metal material, for each predetermined range of the process; a second collection step of evaluating and collecting the quality of the metal material in the final process of each of the processes for each predetermined range of the final process; a third collection step of collecting at least one of the cutting positions in the longitudinal and width directions of the metal material in each process and the dimensions of the metal material in the longitudinal and width directions that change in each process; a storage step of correlating and storing the manufacturing conditions for the metal material in each process collected in the first collection step and the quality of the metal material in the final process collected in the second collection step for each predetermined range of the final process; and a model generation step of generating a quality prediction model that predicts the quality for each predetermined range of the final process from the manufacturing conditions for each predetermined range of the final process stored in the storage step, wherein the storage step comprises: a predetermined range for each process corresponding to the predetermined range of the final process based on at least one of the cutting positions in the longitudinal and width directions of the metal material and the dimensions of the metal material in the longitudinal and width directions that change in each process, collected in the third collection step; and a manufacturing condition for the metal material in the predetermined range for each process that has been identified and a quality of the metal material in the final process that are associated with each predetermined range of the final process and are saved.

2. A method for generating a quality prediction model for metallic materials as described in claim 1, wherein, of the predetermined ranges for each process, the predetermined range in the longitudinal direction of the metallic material is determined based on the moving distance of the metallic material according to the conveying direction in each process, and, of the predetermined ranges for each process, the predetermined range in the width direction of the metallic material is determined based on a range division distance predetermined in each process.

3. A method for generating a quality prediction model for metal materials as described in claim 1 or claim 2, wherein the third collection step collects cutting positions including the truncation of the widthwise ends of the metal material in each process and the division of the metal material at a predetermined width position as the boundary, as cutting positions in the widthwise direction of the metal material in each process.

4. A method for generating a quality prediction model for metallic materials as described in any one of claims 1 to 3, wherein the third collection step collects dimensions including, as the widthwise dimensions of the metallic material in each of the processes, changes in widthwise length due to plastic deformation caused by rolling the metallic material in each of the processes and changes in widthwise length due to plastic deformation caused in response to the longitudinal tension applied to the metallic material in each of the processes.

5. A method for generating a quality prediction model for metallic materials as described in any one of claims 1 to 4, wherein the first collection step collects, for each predetermined range of each process, manufacturing conditions for the metallic material in each process, including at least one of the temperature of the metallic material, the thickness of the metallic material before and after rolling, the tension applied to the metallic material, and the elongation rate of the metallic material.

6. A method for generating a quality prediction model for a metallic material according to any one of claims 1 to 5, wherein the model generation step generates the quality prediction model using a statistical analysis method and a machine learning method, including linear regression, local regression, principal component regression, PLS regression, logistic regression, support vector machine, decision tree, regression tree, random forest, gradient boosting tree, and neural network.

7. A method for generating a quality prediction model for a metallic material according to any one of claims 1 to 6, comprising: a re-storing step of associating the manufacturing conditions of the metallic material in each process collected in the first collecting step with the quality of the metallic material in the final process collected in the second collecting step, and re-storing them for each predetermined range; and a model re-generation step of re-generating a quality prediction model that predicts the quality for each predetermined range from the manufacturing conditions for each predetermined range re-saved in the re-storing step, wherein, of the predetermined ranges, the predetermined range in the longitudinal direction of the metallic material is set to be the predetermined range in the longitudinal direction of the metallic material in the process to which the manufacturing conditions belong that have the greatest importance index for quality output for each manufacturing condition that is input to the quality prediction model generated in the model generation step; and a method for generating a quality prediction model for a metallic material according to any one of claims 1 to 6, comprising:

8. A method for predicting the quality of a metallic material, comprising a quality prediction step of predicting the quality of a metallic material manufactured under any manufacturing conditions for each specified range of the final process, using a quality prediction model generated by the method for generating a quality prediction model for a metallic material as defined in any one of claims 1 to 6.

9. A method for predicting the quality of a metallic material, comprising a quality prediction step of predicting the quality of a metallic material manufactured under any manufacturing conditions for each specified range using a quality prediction model generated by the method for generating a quality prediction model for a metallic material as described in claim 7.

10. A method for manufacturing a metallic material, comprising: an importance index calculation step for calculating, using a quality prediction model generated by the method for generating a quality prediction model for metallic materials according to any one of claims 1 to 7, an importance index indicating the impact on the prediction of output quality for each manufacturing condition that is an input to the quality prediction model; a manufacturing condition estimation step for estimating manufacturing conditions with high importance indices as manufacturing conditions that are factors that affect the quality of the metallic material; a quality prediction step for predicting output quality using the quality prediction model that includes as input the estimated manufacturing conditions that are factors that affect the quality of the metallic material; a manufacturing condition determination step for determining manufacturing conditions that are factors that affect the quality of the metallic material so that the predicted quality falls within a predetermined range; and a metallic material manufacturing step for manufacturing a metallic material based on the determined manufacturing conditions.

11. A method for manufacturing a material comprising: a first collecting step of collecting manufacturing conditions for a material in each process within a predetermined range in the longitudinal and width directions of the material, for each predetermined range of the process; a second collecting step of evaluating and collecting the quality of the material in the final process of each of the processes for each predetermined range of the final process; a third collecting step of collecting at least one of the cutting positions in the longitudinal and width directions of the material in each process and the dimensions of the material in the longitudinal and width directions that change in each process; a storing step of correlating and storing the manufacturing conditions for the material in each process collected in the first collecting step and the quality of the material in the final process collected in the second collecting step for each predetermined range of the final process; and a model generating step of generating a quality prediction model that predicts the quality for each predetermined range of the final process from the manufacturing conditions for each predetermined range of the final process stored in the storing step, wherein the storing step comprises: a predetermined range for each process corresponding to the predetermined range of the final process based on at least one of the cutting positions in the longitudinal direction and the width direction of the material and the dimensions of the material in the longitudinal direction and the width direction that change in each process, collected in the third collection step; and a manufacturing condition for the material in the specified predetermined range for each process and a quality of the material in the final process are associated and saved for each predetermined range of the final process.

12. A method for manufacturing a metal material comprising: a first collection step of collecting manufacturing conditions for each process within a predetermined range in the longitudinal and width directions of the metal material, for each predetermined range of the process; a second collection step of evaluating and collecting the quality of the metal material in the final process of each of the processes for each predetermined range of the final process; a third collection step of collecting at least one of the cutting positions in the longitudinal and width directions of the metal material in each process and the longitudinal and width dimensions of the metal material that change in each process; and a storage step of correlating and storing the manufacturing conditions for the metal material in each process collected in the first collection step and the quality of the metal material in the final process collected in the second collection step for each predetermined range of the final process, wherein the storage step identifies a predetermined range for each process that corresponds to the predetermined range of the final process based on at least one of the cutting positions in the longitudinal and width directions of the metal material and the longitudinal and width dimensions of the metal material that change in each process collected in the third collection step, a data processing method for storing the specified manufacturing conditions for the metal material within a predetermined range of each process and the quality of the metal material in the final process in association with each other for each predetermined range of the final process.

13. A system comprising: a first collection unit that collects manufacturing conditions for a metal material in each process within a predetermined range in the longitudinal and width directions of the metal material, for each predetermined range of the process; a second collection unit that evaluates and collects the quality of the metal material in the final process of each of the processes for each predetermined range of the final process; a third collection unit that collects at least one of the cutting positions in the longitudinal and width directions of the metal material in each process and the longitudinal and width directions dimensions of the metal material that change in each process; a storage unit that associates and stores the manufacturing conditions for the metal material in each process collected by the first collection unit and the quality of the metal material in the final process collected by the second collection unit for each predetermined range of the final process; and a model generation unit that generates a quality prediction model that predicts the quality for each predetermined range of the final process from the manufacturing conditions for each predetermined range of the final process stored in the storage unit, wherein the storage unit: a predetermined range for each process corresponding to the predetermined range of the final process based on at least one of the cutting positions in the longitudinal direction and width direction of the metal material and the dimensions of the metal material in the longitudinal direction and width direction that change in each process, collected by the third collection unit; and a manufacturing condition for the metal material in the specified predetermined range for each process and a quality of the metal material in the final process are associated with each predetermined range of the final process and saved.

14. The quality prediction model generating device for metallic materials according to claim 13, wherein the storage unit associates the manufacturing conditions of the metallic material in each process collected by the first collection unit with the quality of the metallic material in the final process collected by the second collection unit, and re-stores them in association with each other for a predetermined range; the model generation unit re-generates a quality prediction model that predicts the quality for each of the predetermined ranges from the re-stored manufacturing conditions for each of the predetermined ranges; the predetermined range in the longitudinal direction of the metallic material is set to be the predetermined range in the longitudinal direction of the metallic material in the process to which the manufacturing condition that outputs the greatest importance index for quality for each manufacturing condition that is input to the quality prediction model generated by the model generation unit belongs; and the predetermined range in the width direction of the metallic material is determined based on a range division distance predetermined for the final process.

15. A quality prediction device for metallic materials, comprising a quality prediction unit that uses a quality prediction model generated by the metallic material quality prediction model generation device described in claim 13 to predict the quality of metallic materials manufactured under any manufacturing conditions for each specified range of the final process.

16. A quality prediction device for metallic materials, comprising a quality prediction unit that predicts the quality of metallic materials manufactured under any manufacturing conditions for each specified range using a quality prediction model generated by the metallic material quality prediction model generation device described in claim 14.

17. A system comprising: a first collection unit that collects manufacturing conditions for a metal material in each process within a predetermined range in the longitudinal and width directions of the metal material, for each predetermined range of the process; a second collection unit that evaluates and collects the quality of the metal material in the final process of each of the processes for each predetermined range of the final process; a third collection unit that collects at least one of the cutting positions in the longitudinal and width directions of the metal material in each of the processes and the longitudinal and width dimensions of the metal material that change in each of the processes; and a storage unit that associates and stores the manufacturing conditions for the metal material in each of the processes collected by the first collection unit and the quality of the metal material in the final process collected by the second collection unit for each predetermined range of the final process, wherein the storage unit identifies the predetermined range of each process that corresponds to the predetermined range of the final process based on at least one of the cutting positions in the longitudinal and width directions of the metal material and the longitudinal and width dimensions of the metal material that change in each of the processes collected by the third collection unit, and storing the identified manufacturing conditions for the metal material within a predetermined range of each process and the quality of the metal material in the final process in association with each predetermined range of the final process.

Images