Quality prediction model generation method, quality prediction method for metallic materials, manufacturing method for metallic materials, manufacturing condition presentation method for metallic materials, quality prediction model generation device, quality prediction device for metallic materials, manufacturing condition presentation device for metallic materials, and manufacturing system for metallic materials

Abstract

The quality prediction model generation method includes an acquisition step (S1) of acquiring explanatory variables selected from the manufacturing conditions of each process and a dependent variable which is the state of quality defects in the manufactured metal material, a storage step (S3) of associating the explanatory variables and the dependent variable and storing them as learning data, calculation steps (S4 to S6) of dividing the learning data into groups and testing whether there are significant differences in the state of the quality defects, a search step (S7) of searching for the most significant grouping, and a generation step (S8) of generating a quality prediction model by machine learning with a group according to the most significant grouping found.

Description

[Technical field]

[0001] The present disclosure relates to a quality prediction model generating method, a quality prediction method for metallic materials, a manufacturing method for metallic materials, a manufacturing condition presentation method for metallic materials, a quality prediction model generating device, a quality prediction device for metallic materials, a manufacturing condition presentation device for metallic materials, and a manufacturing system for metallic materials. [Background technology]

[0002] As a method for predicting quality for any required condition, for example, a method is known in which the distance between a plurality of past observation conditions stored in a performance database and a desired required condition is calculated, and the quality is predicted using the calculated distance. For example, according to Patent Document 1, a method is known in which the weight of the observed data (performance data) is calculated from the calculated distance, a function is created from the calculated weight to fit the vicinity of the required condition, and the quality for the required condition is predicted using the created function. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7207547 Summary of the Invention [Problem to be solved by the invention]

[0004] The quality prediction method of Patent Document 1 associates the manufacturing conditions of each process with the quality of the metal material manufactured under these manufacturing conditions for each predetermined range, accumulates performance data, and generates a quality prediction model that predicts the quality of the metal material for each predetermined range. The predetermined range is specified by taking into consideration, for example, the cutting position of the metal material. Here, the quality prediction method of Patent Document 1 can predict the quality for any manufacturing condition with high accuracy, but further improvement in prediction accuracy is required. For example, further improvement in prediction accuracy is expected by objectively classifying data so as to correspond to changes in the quality of the metal material and generating a quality prediction model for each classification.

[0005] In consideration of the above circumstances, the object of the present disclosure is to provide a quality prediction model generation method, a metallic material quality prediction method, a metallic material manufacturing method, a quality prediction model generation device, and a metallic material quality prediction device that enable highly accurate prediction of the quality of metallic materials. [Means for solving the problem]

[0006] (1) A quality prediction model generating method according to an embodiment of the present disclosure includes: A quality prediction model generation method for generating a quality prediction model of a metal material manufactured through one or more processes, comprising the steps of: An acquisition step of acquiring explanatory variables selected from manufacturing conditions of each process and a response variable which is a state of quality defects of the manufactured metal material; a storing step of storing the explanatory variables and the objective variables as learning data in association with each other; a calculation step of dividing the learning data into groups and testing whether there is a significant difference in the state of the quality defect; a search step for searching for the most significant grouping; A generating step of generating the quality prediction model by machine learning using groups according to the most significant groupings found.

[0007] (2) As an embodiment of the present disclosure, in (1), The explanatory variables are taken as numerical or categorical variables.

[0008] (3) As an embodiment of the present disclosure, in (2), The calculation step, when the explanatory variables are acquired as categorical variables, performs a numerical conversion to classify the categorical variables by assigning different binary numerical values, and calculates an index to evaluate the significant difference based on the assigned numerical values ​​and the state of the quality defect.

[0009] (4) As an embodiment of the present disclosure, in (3), The calculating step calculates the index for each of a plurality of groupings.

[0010] (5) As an embodiment of the present disclosure, in any one of (1) to (4), The calculation step, after the search step is performed, further groups one of the groups according to the most significant grouping found and the test is performed.

[0011] (6) As an embodiment of the present disclosure, in any one of (1) to (5), The machine learning techniques include at least linear regression, local regression, principal component regression, PLS regression, neural networks, regression trees, random forests, LightGBM, and XGBoost.

[0012] (7) As an embodiment of the present disclosure, in any one of (1) to (6), When the prediction accuracy of the quality prediction model falls outside a predetermined range, or at predetermined intervals, the quality prediction model is regenerated.

[0013] (8) A method for predicting quality of a metallic material according to an embodiment of the present disclosure, The state of quality defects in the metal material is predicted using a quality prediction model generated by any one of the quality prediction model generation methods (1) to (7).

[0014] (9) A method for producing a metal material according to an embodiment of the present disclosure includes: (8) Predicting the state of quality defects of the metallic material by the metallic material quality prediction method; When the metallic material is predicted to have a quality defect, at least a part of the manufacturing conditions is changed.

[0015] (10) A method for presenting manufacturing conditions for a metallic material according to an embodiment of the present disclosure, (8) The state of a quality defect of the metallic material is predicted by the metallic material quality prediction method, and if a quality defect of the metallic material is predicted, a change in at least a part of the manufacturing conditions that will eliminate the quality defect is searched for and presented.

[0016] (11) A method for producing a metal material according to an embodiment of the present disclosure includes: (10) A metallic material is manufactured based on the change in manufacturing conditions presented by the method for presenting manufacturing conditions for a metallic material.

[0017] (12) A quality prediction model generation device according to an embodiment of the present disclosure, A quality prediction model generation device that generates a quality prediction model of a metal material manufactured through one or more processes, an acquisition unit that acquires explanatory variables selected from manufacturing conditions of each process and a target variable that is a state of quality defects of the manufactured metal material; a storage unit that stores the explanatory variables and the objective variables in association with each other as learning data; a calculation unit that divides the learning data into groups and tests whether there is a significant difference in the state of the quality defect; a search part for searching for the most significant grouping; A generation unit generates the quality prediction model by machine learning using a group according to the most significant grouping found.

[0018] (13) A quality prediction device for a metallic material according to an embodiment of the present disclosure, (12) The state of quality defects of the metal material is predicted using the quality prediction model generated by the quality prediction model generation device.

[0019] (14) An apparatus for presenting manufacturing conditions for a metal material according to an embodiment of the present disclosure, (13) Based on the presence or absence of quality defects in the metal material predicted by the metal material quality prediction device, at least a part of the changes in the manufacturing conditions of the metal material are suggested.

[0020] (15) A system for producing a metal material according to an embodiment of the present disclosure, (14) Controlling the manufacturing equipment for each process of manufacturing the metal material according to the change in the manufacturing conditions output from the output unit of the metal material manufacturing condition presentation device. Effect of the Invention

[0021] According to the present disclosure, it is possible to provide a quality prediction model generation method, a quality prediction method for a metallic material, a manufacturing method for a metallic material, a quality prediction model generation device, and a quality prediction device for a metallic material that enable highly accurate prediction of the quality of a metallic material. [Brief description of the drawings]

[0022] [Figure 1] FIG. 1 is a block diagram illustrating an example configuration of a quality prediction model generating device according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is a flowchart showing the process of a quality prediction model generating method according to an embodiment of the present disclosure. [Diagram 3] FIG. 3 is a diagram for explaining grouping. [Figure 4] FIG. 4 is a diagram showing combinations of two groups and the presence or absence of a quality defect. [Diagram 5] FIG. 5 is a diagram illustrating the slab casting process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] Hereinafter, a quality prediction model generating method, a quality prediction method for a metallic material, a manufacturing method for a metallic material, a manufacturing condition presenting method for a metallic material, a quality prediction model generating device 10 (see FIG. 1), a quality prediction device for a metallic material, a manufacturing condition presenting device for a metallic material, and a manufacturing system for a metallic material according to an embodiment of the present disclosure will be described with reference to the drawings. The quality prediction model generating method and the quality prediction model generating device 10 according to the present embodiment, in summary, acquire manufacturing conditions and quality data for each process of manufacturing a metallic material, perform testing on the acquired data, and divide the data into groups with significant differences in quality. Then, the quality prediction model generating method and the quality prediction model generating device 10 generate a quality prediction model that predicts the quality for each group. The quality prediction method for a metallic material according to the present embodiment predicts the state of quality defects of a metallic material using the quality prediction model generated by the quality prediction model generating method or the quality prediction model generating device 10 according to the present embodiment. The state of quality defects is divided into a first state and a second state. For example, the first state may correspond to "present", "predetermined number or more", or "distributed in a predetermined region or more", and the second state may correspond to "absent", "less than a predetermined number", or "distributed in a region less than a predetermined region". In the following description, the first state is "present" and the second state is "absent", but the "presence" of a quality defect is an example of the "state" of the quality defect, and "presence" can be replaced with "state". The manufacturing method of the metallic material according to the present embodiment is a method for manufacturing a metallic material, in which the presence or absence of a quality defect of the metallic material is predicted by the quality prediction method of the metallic material according to the present embodiment, and at least a part of the manufacturing conditions is changed when the quality defect of the metallic material is predicted. The quality prediction method of the metallic material according to the present embodiment predicts the presence or absence of a quality defect of the metallic material using a quality prediction model generated by the quality prediction model generation method or the quality prediction model generation device 10 according to the present embodiment. In addition, the manufacturing condition presentation method of the metallic material according to the present embodiment presents at least a part of the change of the manufacturing conditions of the metallic material based on the result of the presence or absence of a quality defect of the metallic material predicted by the quality prediction method of the metallic material or the quality prediction device according to the present embodiment.

[0024] FIG. 1 is a block diagram showing an example of the configuration of a quality prediction model generation device 10 according to the present embodiment. The quality prediction model generation device 10 is used in a manufacturing process for manufacturing a metal material. The metal material may be, for example, a steel product, and may include a semi-finished product such as a slab, and a finished product such as a steel plate manufactured by rolling a slab. The quality prediction model generation device 10 generates a quality prediction model of a metal material manufactured through one or more processes. The quality prediction model generation device 10 executes a quality prediction model generation method (see FIG. 2). In the example of FIG. 1, the quality prediction model generation device 10 also functions as a quality prediction device for a metal material that predicts the presence or absence of a quality defect in the metal material using the generated quality prediction model. The quality prediction device for a metal material executes a quality prediction method for a metal material (see FIG. 2). Here, the quality prediction model generation device 10 may only have a function of generating a quality prediction model. In this case, the quality prediction device for a metal material may be configured as a device separate from the quality prediction model generation device 10 that can acquire a quality prediction model from the quality prediction model generation device 10. 1, the quality prediction model generating device 10 also functions as a manufacturing condition presenting device for metallic material that presents at least a part of the changes to the manufacturing conditions for metallic material based on the presence or absence of quality defects of the metallic material predicted by the prediction method for metallic material. The manufacturing condition presenting device for metallic material executes the manufacturing condition presenting method for metallic material. The manufacturing condition presenting device may be configured as a device separate from the quality prediction model generating device 10.

[0025] As shown in FIG. 1, the quality prediction model generating device 10 includes a communication unit 11, a storage unit 12, and a control unit 13. The control unit 13 includes an acquisition unit 131, a storage unit 132, a calculation unit 133, a search unit 134, a generation unit 135, a prediction unit 136, a manufacturing condition presentation unit 137, and an output unit 138. The quality prediction model generating device 10 may be realized as a hardware configuration by a general-purpose computer (information processing device) such as a personal computer or a workstation. The computer includes a processor such as a CPU (Central Processing Unit). The computer also includes a memory such as a RAM (Random Access Memory) and a ROM (Read Only Memory) as a main component. Details of the components of the quality prediction model generating device 10 will be described later.

[0026] The quality prediction model generating device 10 acquires quality data and operation data from the operation data server 20. The operation data includes manufacturing conditions for each process of manufacturing a metal material. In the example of FIG. 1, the quality prediction model generating device 10 acquires quality data and operation data stored in a performance database provided in the operation data server 20. The operation data server 20 is capable of communicating with the quality prediction model generating device 10 via a network, and may be realized by, for example, a computer that manages the manufacturing process. The network is, for example, the Internet. In this embodiment, the display unit 30 displays the prediction result output from the quality prediction device for metal materials (integrated with the quality prediction model generating device 10 in the example of FIG. 1).

[0027] Here, the manufacturing conditions stored in the performance database may include at least one of the components, temperature, pressure, plate thickness, and plate threading speed of the metal material in each process. The quality data stored in the performance database may include at least one of the following: material properties such as tensile strength, the presence or absence of defects on the surface of the material, the presence or absence of defects on the internal defects of the material, the number of defects, and the defect positions. The manufacturing conditions and quality data are obtained as operation performance data, operation target values, analysis values, or quality judgment results from sensors, manufacturing equipment, control devices, or quality control devices installed in a factory, for example. For example, the presence or absence of internal defects in a steel plate (an example of a metal material) may be obtained as quality data by an ultrasonic flaw detector (an example of a sensor and a quality control device). The performance database is a database in which the manufacturing conditions, operation performance data, and various analysis values ​​of the corresponding processes are associated with data on the quality of the manufactured metal material.

[0028] Here, the quality prediction model generating device 10 is a computer separate from the operation data server 20 in the example of FIG. 1, but may be configured as the same computer as the operation data server 20. The display unit 30 may be a device for an operator working in the manufacturing process to obtain information. The display unit 30 may be realized by a display of a terminal device such as a smartphone or a tablet. When it is predicted that there is a quality defect in the metal material, a manufacturing method for the metal material may be executed in which at least a part of the manufacturing conditions is changed based on an instruction from an operator. The display unit 30 may be a display device such as a liquid crystal display (LCD) or an organic electro-luminescence panel (OLED).

[0029] The components of the quality prediction model generating device 10 will be described in detail below. The communication unit 11 includes one or more communication modules that connect to a network. The communication unit 11 may include a communication module that supports mobile communication standards such as 4G (4th Generation) and 5G (5th Generation). The communication unit 11 may include a communication module that supports a wired or wireless LAN standard.

[0030] The storage unit 12 is one or more memories. The memory may be, for example, a semiconductor memory, a magnetic memory, an optical memory, or the like, but is not limited to these, and may be any memory. The storage unit 12 is, for example, built into the quality prediction model generation device 10, but may also be configured to be accessed from the outside by the quality prediction model generation device 10 via any interface.

[0031] The storage unit 12 stores various data used in various calculations performed by the control unit 13. The storage unit 12 may also store results and intermediate data of various calculations performed by the control unit 13. In this embodiment, the storage unit 12 stores learning data and a generated model.

[0032] The control unit 13 is one or more processors. The processor may be, for example, a general-purpose processor or a dedicated processor specialized for a specific process, but is not limited to these and may be any processor. The control unit 13 controls the overall operation of the quality prediction model generation device 10.

[0033] Here, the quality prediction model generating device 10 may have the following software configuration. One or more programs used to control the operation of the quality prediction model generating device 10 are stored in the storage unit 12. When the programs stored in the storage unit 12 are read by the processor of the control unit 13, they cause the control unit 13 to function as an acquisition unit 131, a storage unit 132, a calculation unit 133, a search unit 134, a generation unit 135, and a prediction unit 136.

[0034] The acquiring unit 131 acquires the manufacturing conditions and quality data from the performance database. Here, the performance database stores a large amount of quality data including the manufacturing conditions of each process and measurements of the presence or absence of quality defects in the metal material manufactured under those manufacturing conditions, for example, as time-series data. The acquiring unit 131 may selectively acquire explanatory variables and objective variables that can be used as learning data for machine learning, which will be described later, from the performance database. In this embodiment, the acquiring unit 131 acquires explanatory variables selected from the manufacturing conditions of each process and an objective variable that is the presence or absence of quality defects in the manufactured metal material.

[0035] The storage unit 132 associates explanatory variables, which are data acquired by the acquisition unit 131, with objective variables and stores them as learning data. That is, the storage unit 132 associates (links) the manufacturing conditions of each process acquired as explanatory variables with quality data including measured values ​​of the presence or absence of quality defects of the metal material manufactured under the manufacturing conditions, and sets them as learning data. The storage destination of the learning data may be, for example, the memory unit 12. Here, if the manufacturing conditions of each process are a large amount of data acquired over a long period of time or at a high frequency, the storage unit 132 may thin out or convert the data into a representative value such as an average value or a median value to an appropriate amount of data before storing the learning data. In addition, as shown in FIG. 3, the storage unit 132 may associate "the manufacturing conditions of each process and the presence or absence of quality defects (data group)" for each predetermined range of the metal material. In the example of FIG. 3, the learning data is associated for each predetermined range from the head of the conveying direction of the metal material, and is divided into data group 1, data group 2, data group 3, data group 4, ..., data group N in order. Here, N is an integer equal to or greater than 2. Furthermore, the predetermined range may be determined based on the cutting position when, for example, a metal material is cut during manufacturing, but is not limited to this determination method.

[0036] The calculation unit 133 divides the learning data into groups and performs a test (significant difference test) to see if there is a significant difference in the presence or absence of quality defects. The calculation unit 133 divides the above data groups 1 to N into groups X and Y, for example, as in patterns 1 to M in FIG. 3. Here, M is an integer equal to or greater than 2. The calculation unit 133 performs a plurality of groupings and performs a significant difference test for each of the plurality of groupings. The calculation unit 133 may perform grouping comprehensively for all combinations, or may perform grouping only for some combinations. For example, a constraint such as an upper limit of calculation time may be given, and the calculation unit 133 may perform grouping for combinations that can be calculated within the constraints. In addition, a priority order for grouping may be determined in advance based on manufacturing knowledge or the like.

[0037] In this embodiment, the calculation unit 133 uses Fisher's exact test. However, the method of the significant difference test is not limited. In this embodiment, the calculation unit 133 uses a P value as an index for evaluating a significant difference. The P value is a value indicating the probability of a combination that produces a difference equal to or greater than the observed difference among all possible combinations. Therefore, the smaller the P value, the higher the significance (the higher the probability of a significant difference). As shown in FIG. 4, the numbers corresponding to the combinations of two groups and the presence or absence of quality defects are a, b, c, and d. In this case, the P value is calculated by the following formula. n is the total number of learning data.

[0038]

number

[0039] Here, the grouping is performed based on explanatory variables. The explanatory variables are acquired as numerical variables or categorical variables. For example, when the explanatory variables used for grouping are acquired as numerical variables, the calculation unit 133 may classify the explanatory variables as group X if they are equal to or greater than a threshold, and may classify the explanatory variables as group Y if they are less than the threshold.

[0040] For example, when an explanatory variable used for grouping is acquired as a categorical variable, the calculation unit 133 may perform a numerical conversion to classify the categorical variable by assigning different binary numerical values. The different binary numerical values ​​may be, for example, 0 or 1, and in the present embodiment, a description will be given assuming that 0 or 1 is used. In addition, for example, one-hot encoding may be used as the numerical conversion, but is not limited thereto. Then, the calculation unit 133 may calculate an index (P value) for evaluating a significant difference based on the assigned 0 or 1 and the presence or absence of a quality defect. That is, the calculation unit 133 may classify the explanatory variable after the numerical conversion as group X if it is 1 (i.e., if it is a certain category), and as group Y if it is 0 (i.e., if it is not a certain category).

[0041] Here, the explanatory variables are selected from the manufacturing conditions of each process. The manufacturing conditions include many variables such as the composition of the metal materials and the temperature in various processes. Therefore, the explanatory variables used for grouping can be selected in various ways. For example, if n explanatory variables are used for grouping, 2 n The P value may be calculated for all of the patterns, or may be calculated for some of the patterns. In addition, candidates for explanatory variables used for grouping may be determined in advance based on manufacturing knowledge.

[0042] The search unit 134 searches for the most significant grouping. In this embodiment, the search unit 134 identifies the grouping with the lowest P value. In the example of Fig. 3, the calculation unit 133 calculates a P value for each of patterns 1 to M, and the search unit 134 identifies the pattern with the lowest P value (pattern 2 as an example) as the most significant grouping. In this example, the search unit 134 determines that there is the most significant difference between group X including data groups 1 and 2 and group Y including data groups 3 to N.

[0043] Here, the calculation unit 133 and the search unit 134 may perform multi-stage grouping starting from the group with the lowest P value. In other words, after the search unit 134 searches for the most significant grouping, the calculation unit 133 may perform further grouping and significance test (calculation of P value) for one of the groups according to the grouping. In the example of FIG. 3, after the search unit 134 identifies pattern 2 as the most significant grouping, the calculation unit 133 may perform further grouping and calculation of P value for group Y including data groups 3 to N. Then, the search unit 134 may search for the most significant grouping for group Y, and may determine that, for example, group Y1 including data groups 3 to i and group Y2 including data groups i to N have the most significant difference. Here, i is an integer of 3 or more. In addition, the number of times (repetition number) of multi-stage grouping is not limited, but it is necessary that the number of learning data included in one group is not too small so that the accuracy of the generated quality prediction model does not decrease.

[0044] Here, for multiple explanatory variables, those with a P-value equal to or less than a predetermined value (e.g., 1 / n where n is the total number of learning data) may be determined as important variables, and grouping may be performed using the important variables. Furthermore, among the important variables, only one that is easy to explain the relationship with product quality may be selected, or two or more important variables may be selected. Furthermore, when the accuracy of prediction using a quality prediction model generated in the past is obtained, the important variables may be selected taking that accuracy into consideration.

[0045] The generating unit 135 generates a quality prediction model by machine learning using a group according to the most significant grouping searched for by the searching unit 134. The machine learning method is not particularly limited, and may be, for example, linear regression, local regression, principal component regression, PLS regression, neural network, regression tree, random forest, LightGBM, or XGBoost. In other words, the machine learning method that the generating unit 135 can use includes at least linear regression, local regression, principal component regression, PLS regression, neural network, regression tree, random forest, LightGBM, and XGBoost. For example, in the above example in which the calculation unit 133 and the searching unit 134 perform multi-stage grouping, the generating unit 135 generates three quality prediction models using the learning data of each of the groups X, Y1, and Y2. The generating unit 135 stores the generated quality prediction model in, for example, the storage unit 12.

[0046] The prediction unit 136 uses the generated quality prediction model to predict the presence or absence of quality defects in a metal material manufactured under any manufacturing conditions. The prediction result is displayed, for example, on the display unit 30. When the presence of a quality defect in the metal material is predicted, at least a part of the manufacturing conditions is changed based on an instruction from an operator, thereby improving the quality of the manufactured metal material.

[0047] In order to maintain the prediction accuracy of the presence or absence of quality defects in the metallic material, the prediction unit 136 may output a command to the calculation unit 133, the search unit 134, and the generation unit 135 to regenerate the quality prediction model of the metallic material. That is, the prediction unit 136 may cause the calculation unit 133 to regroup the learning data and retest whether there is a significant difference in the state of the quality defects. The prediction unit 136 may output a command to cause the search unit 134 to re-search for the most advantageous grouping and to cause the generation unit 135 to regenerate the quality prediction model by machine learning for the group according to the most advantageous grouping searched for. The prediction unit 136 may output a command to regenerate the quality prediction model of the metallic material when, for example, the prediction accuracy of the presence or absence of quality defects in the metallic material falls outside a predetermined range (is not within a predetermined range) or at predetermined intervals (for example, every three months). Then, the quality prediction model may be regenerated based on the command.

[0048] The manufacturing condition presentation unit 137 searches for at least a part of the manufacturing conditions of each process that will eliminate the quality defect in the prediction result based on the presence or absence of the quality defect of the metal material predicted by the prediction unit 136, and presents the change of the found manufacturing conditions via the display unit 30. The manufacturing condition presentation unit 137 changes at least a part of the operable manufacturing conditions of each process from the current value to another value, and inputs the other manufacturing conditions as the current values ​​as explanatory variables of the quality prediction model generated by the generation unit 135. Then, the manufacturing condition presentation unit 137 predicts the presence or absence of the quality defect of the metal material. If the prediction result is "absent" of the quality defect, the changed manufacturing conditions are presented via the display unit 30. If the prediction result is "present" of the quality defect, the values ​​of the operable manufacturing conditions of each process are further changed to another value until the prediction result by the quality prediction model becomes "absent" of the quality defect. For example, an optimization method such as an iterative method may be used. The manufacturing condition presenting unit 137 may use the quality prediction model generated by the generating unit 135 through machine learning to calculate variable importance (feature importance) for the objective variable, which is the presence or absence of quality defects in the manufactured metal material. Then, the manufacturing condition presenting unit 137 may select, based on the variable importance, a manufacturing condition that has a high contribution to the objective variable among the manufacturing conditions of each process, which are explanatory variables, as a manufacturing condition to be changed. The manufacturing condition presenting unit 137 may calculate the variable importance (feature importance) by applying a mathematical method such as SHAP (Shapley Additive exPlanations), Gini impurity, frequency, gain (amount of reduction in the objective function depending on the presence or absence of the feature).

[0049] The quality prediction model generating device 10 may also include an output unit 138 that outputs (transmits) the changes in the manufacturing conditions presented by the manufacturing condition presenting unit 137 to a control computer that controls the manufacturing equipment of each process for manufacturing the metal material via a network. The control computer controls the manufacturing equipment of each process for manufacturing the metal material according to the transmitted changes in the manufacturing conditions, and the quality prediction model generating device 10 and the control computer may constitute a manufacturing system for the metal material. The control computer may also accept modifications to the manufacturing conditions by an operator and control the manufacturing equipment of each process.

[0050] The output unit 138 may display the changes in the manufacturing conditions presented by the manufacturing condition presenting unit 137 through a display device (for example, a liquid crystal display) connected via a network.

[0051] FIG. 2 is a flowchart showing the processing of the quality prediction model generation method executed by the quality prediction model generation device 10 according to the present embodiment. In the present embodiment, the quality prediction model generation device 10 also functions as a quality prediction device for a metallic material, and predicts the presence or absence of a quality defect in the metallic material. FIG. 2 is also a flowchart showing the processing of the quality prediction method for a metallic material. In the present embodiment, the quality prediction model generation device 10 also functions as a manufacturing condition presentation device for a metallic material, and presents at least a partial change in the manufacturing conditions for the metallic material. FIG. 2 is also a flowchart showing the processing of the manufacturing condition presentation method for a metallic material. In the present embodiment, the quality prediction model generation device 10 constitutes a manufacturing system for a metallic material together with a control computer that controls a manufacturing device that manufactures the metallic material. FIG. 2 is also a flowchart showing the processing of the manufacturing method for a metallic material. Here, the prediction of the presence or absence of a quality defect in the metallic material (step S9), the regeneration of the quality prediction model (step S10), the presentation of manufacturing conditions for the metallic material (step S11), and the manufacturing of the metallic material (step S12) may be performed consecutively with the generation of the quality prediction model (step S8). In addition, steps S9 to S12 may be performed after a certain time has elapsed after the generation of the quality prediction model.

[0052] The acquiring unit 131 acquires manufacturing conditions and quality data from the performance database (step S1). Step S1 corresponds to an acquiring step.

[0053] The storage unit 132 eliminates multicollinearity from the data acquired by the acquisition unit 131 (step S2), and associates the explanatory variables with the objective variables and stores them as learning data (step S3). A known method may be used to eliminate multicollinearity. Step S2 may be omitted. Step S3 corresponds to a storage step.

[0054] When the explanatory variables used for grouping are acquired as categorical variables (Yes in step S4), the calculation unit 133 executes a numerical conversion such as one-hot encoding (step S5). When the explanatory variables used for grouping are acquired as numerical variables (No in step S4), the calculation unit 133 proceeds to the process of step S6. The calculation unit 133 divides the learning data into groups and executes a significant difference test (step S6). Steps S4 to S6 correspond to calculation steps.

[0055] The search unit 134 searches for the most significant grouping (step S7). Step S7 corresponds to a search step.

[0056] The generating unit 135 generates a quality prediction model using groups according to the grouping found (step S8). Step S8 corresponds to a generating step.

[0057] The prediction unit 136 predicts the presence or absence of quality defects in the metallic material using the generated quality prediction model (step S9).

[0058] The prediction unit 136 may output a command to regenerate the quality prediction model of the metallic material in order to maintain the prediction accuracy of the presence or absence of a quality defect in the metallic material (step S10). The prediction unit 136 may output a command to regenerate the quality prediction model of the metallic material, for example, when the prediction accuracy of the presence or absence of a quality defect in the metallic material falls outside a predetermined range (is not within the predetermined range), or at predetermined intervals. If the quality prediction model of the metallic material is to be regenerated (Yes in step S10), the process returns to step S6. If the quality prediction model of the metallic material is not to be regenerated (No in step S10), the process proceeds to step S11.

[0059] Based on the predicted presence or absence of quality defects in the metallic material, the manufacturing condition presentation unit 137 searches for at least a part of the manufacturing conditions for each process that will result in a prediction result of "no" quality defects, and presents the searched changes to the manufacturing conditions (step 11). The searched changes to the manufacturing conditions are transmitted as a manufacturing condition change command via the output unit 138 to a control computer that controls the manufacturing equipment for each process of manufacturing the metallic material, and are displayed via the display unit 30.

[0060] Based on the searched changes in the manufacturing conditions, the metal material is manufactured (step S12). The changes in the manufacturing conditions are transmitted to the control computer, and the metal material is manufactured according to the transmitted changes in the manufacturing conditions. Based on the display on the display unit 30, the operator may set the changes in the manufacturing conditions in the control computer.

[0061] The effects of the present disclosure will be specifically described below based on examples, but the present disclosure is not limited to these examples.

[0062] In this example, the quality prediction method for metallic materials according to the above embodiment was applied to predicting the presence or absence of defects in a structural steel plate. The structural steel plate in this example is manufactured from a slab, which is an intermediate product manufactured in a casting process (continuous casting process) shown in Fig. 5.

[0063] In the casting process shown in Figure 5, molten steel in a ladle is poured into a mold via a tundish. The molten steel in the mold is solidified from the surface by primary cooling in the mold, becoming a slab. The slab pulled out of the mold is secondary cooled by cooling water below the mold, and finally solidifies completely. The slab is then cut into predetermined lengths by a cutter to become slabs. The slabs are rolled in the next process, and become products after going through subsequent processing. Here, numbers 1 to N are assigned to each molten steel pot corresponding to the manufactured slabs in the order of cutting, from the top to the bottom, and the slab cutting order is determined.

[0064] The objective variable (quality) for quality prediction in this embodiment is the presence or absence of defects determined by ultrasonic inspection of the product (structural steel plate, steel for pressure vessels, steel for marine structures). The explanatory variables (manufacturing conditions) include the chemical composition of the metallic material in the smelting process. The explanatory variables also include the temperature of the metallic material in the casting process, the casting speed, the cooling conditions, and the slab cutting order. The explanatory variables also include the temperature of the metallic material in the heating process, the temperature of the metallic material in the hot rolling process, and the temperature of the metallic material in the cooling process. The explanatory variables also include the temperature of the metallic material in the cold rolling process and the cutting dimensions.

[0065] As in the quality prediction model generation method according to the above embodiment, data was acquired from a performance database to generate a quality prediction model, which is a machine learning model. LightGBM (Light Gradient Boosting Machine) was used as the machine learning method. Fisher's exact test was used as the significant difference test method. An arbitrary division point of a continuous variable was determined by dividing the difference between the maximum and minimum values ​​of the variable ("maximum minus minimum value") by 25 to find the minimum P value, and adopting the division point corresponding to the minimum P value as the division point of the variable.

[0066] Moreover, the slab cutting order is a kind of categorical variable. Data corresponding to cutting order 1 was categorized as data group 1, data corresponding to cutting orders 2 to N-1 were categorized as data groups 2 to N-1, respectively, and data corresponding to the last cutting order (N) was categorized as data group N. Each data group was classified by binarization using one-hot encoding. In this embodiment, it was determined that the most significant difference was observed when dividing the data into a first group including data groups 1 and 2, a second group including data groups N and N-1, and a third group including the other data groups.

[0067] A quality prediction model was generated for each of these groups using LightGBM. Furthermore, the presence or absence of quality defects was predicted using these quality prediction models. Here, the number of samples in the performance database was 307, and the number of explanatory variables was 260. As a result of the prediction, the accuracy rate of "quality defects are present" using this method was 89%. The accuracy rate of "quality defects are present" using the conventional method was 68%, confirming that the accuracy rate had increased significantly. Here, the conventional method generates one quality prediction model using all of the learning data, without performing grouping using a significance test.

[0068] As described above, the quality prediction model generating method, metallic material quality prediction method, metallic material manufacturing method, quality prediction model generating device 10, metallic material quality prediction device, and metallic material manufacturing system according to the present embodiment perform appropriate grouping based on significant differences in generating a quality prediction model. Therefore, by using the generated quality prediction model, it becomes possible to predict the quality of the metallic material with high accuracy.

[0069] In the model generation of this embodiment, the operating conditions at the time of manufacturing each process of a product manufactured in the past and the quality of the product obtained by manufacturing under those operating conditions are acquired as performance data, and the operating condition data and the quality data corresponding to each other are linked and stored. Then, grouping is performed for a plurality of explanatory variables, and the grouping with the highest significance for the quality data is identified. Conventionally, such grouping has not been performed. The method of the present disclosure enables objectively meaningful grouping even for workers who do not have deep knowledge or sufficient experience of the manufacturing process. The grouping by the method of the present disclosure classifies the manufacturing conditions and the quality data by a significant difference test so as to correspond to the change in the quality of the metal material (so as to reflect the factors that contribute to the quality). Therefore, the quality prediction model generated according to such grouping enables highly accurate prediction.

[0070] In addition, the manufacturing conditions that improve internal defects (defects detected by ultrasonic inspection) in structural steel plates are searched for and presented by the method and device for presenting manufacturing conditions for metallic materials according to the present embodiment. In this embodiment, a quality prediction model generated using the above-mentioned LightGBM was used. Among the manipulable manufacturing conditions in the refining process, casting process, heating process, hot rolling process, cooling process, and cold rolling process, the heating temperature of the heating process, which has a large gain in LightGBM and a high contribution to the presence or absence of internal defects, which is the target variable, was selected as the condition to be changed. Using the quality prediction model, the presence or absence of quality defects in the metallic material was predicted, and a heating temperature at which the predicted result was "no quality defects" was searched for using an iterative method (golden section method) within the range from the maximum value to the minimum value of the heating temperature acquired from the operation data server 20. Then, the change in the heating temperature of the heating process that was searched for was set in the control computer, and the metallic material was manufactured. As a result, the heating temperature increased by +30°C on average, and the occurrence rate of quality defects was 9%. The rate of occurrence of quality defects before application of the manufacturing condition presentation method for metal material according to this embodiment (conventional) was 10%, and it was confirmed that the rate of occurrence of quality defects was improved.

[0071] Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art can easily make various modifications or corrections based on the present disclosure. Therefore, it should be noted that these modifications or corrections are included in the scope of the present disclosure. For example, the functions included in each component or each step can be rearranged so as not to be logically inconsistent, and multiple components or steps can be combined into one or divided. The embodiments of the present disclosure can also be realized as a program executed by a processor included in the device or a storage medium on which a program is recorded. It should be understood that these are also included in the scope of the present disclosure.

[0072] Here, as described above, the quality prediction model generating device 10 does not need to have the function of a quality prediction device for metallic materials. Specifically, the quality prediction model generating device 10 may have only the function of generating a quality prediction model, and the prediction unit 136 may be included in a quality prediction device for metallic materials that is a computer (information processing device) different from the quality prediction model generating device 10. In this case, the quality prediction device for metallic materials obtains the model generated by the quality prediction model generating device 10 and stored in the storage unit 12 via a network, and predicts the presence or absence of a quality defect in the metallic material.

[0073] As another example, the display unit 30 may be included in the quality prediction model generation device 10. [Explanation of symbols]

[0074] 10. Quality prediction model generation device (quality prediction device for metal materials) 11 Communications Department 12 Storage section 13 Control section 20 Operational Data Server 30 Display section 131 Acquisition Department 132 Preservation Department 133 Calculation Department 134 Search Department 135 Generation part 136 Prediction Department 137 Manufacturing condition presentation department 138 Output section

Claims

1. A quality prediction model generation method for generating a quality prediction model of a metal material manufactured through one or more processes, comprising: an acquisition step of acquiring explanatory variables selected from the manufacturing conditions of each process and a response variable that is the state of quality defects in the manufactured metal material; a storing step of storing the explanatory variables and the objective variables as learning data in association with each other; a calculation step of dividing the learning data into groups and testing whether there is a significant difference in the state of the quality defect; a search step for searching for the most significant grouping; A quality prediction model generation method comprising: a generation step of generating the quality prediction model by machine learning using a group according to the most significant grouping found.

2. The quality prediction model generating method according to claim 1 , wherein the explanatory variables are acquired as numerical variables or categorical variables.

3. 3. The quality prediction model generation method according to claim 2, wherein the calculation step, when the explanatory variables are acquired as categorical variables, performs a numerical conversion to classify the categorical variables by assigning different binary numerical values ​​to the categorical variables, and calculates an index to evaluate the significant difference based on the assigned numerical values ​​and the state of the quality defect.

4. The quality prediction model generating method according to claim 3 , wherein the calculation step calculates the index for each of a plurality of groupings.

5. 5. The quality prediction model generation method according to claim 1, wherein the calculation step, after the search step is performed, further grouping and the test are performed on one of the groups found according to the most significant grouping found.

6. 5. The quality prediction model generation method according to claim 1, wherein the machine learning techniques include at least linear regression, local regression, principal component regression, PLS regression, neural network, regression tree, random forest, LightGBM, and XGBoost.

7. 5. The quality prediction model generation method according to claim 1, further comprising the step of: regenerating the quality prediction model when the prediction accuracy of the quality prediction model falls outside a predetermined range or at predetermined intervals.

8. A quality prediction method for a metallic material, which predicts a state of quality defects in the metallic material using a quality prediction model generated by the quality prediction model generation method according to any one of claims 1 to 4.

9. A quality defect state of the metallic material is predicted by the metallic material quality prediction method according to claim 8, A method for manufacturing a metal material, wherein at least a part of the manufacturing conditions is changed when it is predicted that the metal material has a quality defect.

10. A method for presenting manufacturing conditions for metal materials, which predicts the state of quality defects in the metal material using the quality prediction method for metal materials described in claim 8, and if a quality defect in the metal material is predicted, searches for and presents changes to at least some of the manufacturing conditions that will eliminate the quality defect.

11. A method for producing a metallic material, which produces a metallic material based on the changes in the manufacturing conditions presented by the method for presenting manufacturing conditions for a metallic material according to claim 10.

12. A quality prediction model generation device that generates a quality prediction model of a metal material manufactured through one or more processes, an acquisition unit that acquires explanatory variables selected from the manufacturing conditions of each process and a target variable that is the state of quality defects in the manufactured metal material; a storage unit that stores the explanatory variables and the objective variables in association with each other as learning data; a calculation unit that divides the learning data into groups and tests whether there is a significant difference in the state of the quality defect; a search part for searching for the most significant grouping; A quality prediction model generation device comprising: a generation unit that generates the quality prediction model by machine learning using a group according to the most significant grouping found.

13. A quality prediction device for a metallic material, which predicts a state of quality defects in the metallic material using a quality prediction model generated by the quality prediction model generation device according to claim 12.

14. A manufacturing condition suggestion device for metallic materials, which suggests changes to at least some of the manufacturing conditions for metallic materials based on the presence or absence of quality defects in the metallic material predicted by the quality prediction device for metallic materials according to claim 13.

15. A metallic material manufacturing system that controls manufacturing equipment for each process of manufacturing a metallic material in accordance with the change in manufacturing conditions output from the output unit of the metallic material manufacturing condition presentation device according to claim 14.

Images