WO2020148918A1 - 金属材料の設計支援方法及び設計支援装置 - Google Patents
金属材料の設計支援方法及び設計支援装置 Download PDFInfo
- Publication number
- WO2020148918A1 WO2020148918A1 PCT/JP2019/006147 JP2019006147W WO2020148918A1 WO 2020148918 A1 WO2020148918 A1 WO 2020148918A1 JP 2019006147 W JP2019006147 W JP 2019006147W WO 2020148918 A1 WO2020148918 A1 WO 2020148918A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- input
- design support
- data
- metal
- database
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/18—Manufacturability analysis or optimisation for manufacturability
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
Definitions
- the present invention relates to a design support method and a design support apparatus for a metal material having desired characteristics.
- Patent Document 1 proposes a method of performing material design using a mathematical model and optimization calculation in order to reduce the work load related to the design of non-metallic materials.
- Patent Document 2 a characteristic of a newly generated substance is simulated by combining a plurality of types of substances, and information of a substance to be combined with information of a simulation result of the characteristic of a newly generated substance is linked, There has been proposed a material development analysis device that extracts specific information according to a search condition input by a user.
- Patent Document 1 In designing metallic materials, as compared with designing non-metallic materials, many work processes and processing on equipment are performed, and therefore a huge amount of calculation is required for management and control of the work processes and processing on equipment. I need. If the technique of Patent Document 1 is applied to the design of a metallic material, it takes a huge amount of time for optimization calculation, and thus it cannot be said to be a realistic method. Further, in the design of the metal material, the metal structure of the metal material may significantly change according to the manufacturing conditions, and the characteristics of the metal material may significantly change in accordance with the change. However, in Patent Document 1 and Patent Document 2, this point was not taken into consideration.
- an object of the present invention made in view of the above problems is to provide a design support method and a design support apparatus capable of suppressing an increase in calculation load required for designing a metal material.
- a design support method for supporting the design of a metal material by a computer, It is created by using at least one mathematical model in which input information including elemental composition of elements in metal and manufacturing conditions and output information including characteristic values of the metal material are associated with each other, and corresponds to the input information. Input the desired characteristic value to the database in which the output data of the mathematical model for the input data for each mesh that divides the input range into multiple sections is stored in association with the input data.
- a design support device is A design support device for supporting the design of a metal material, It is created by using at least one mathematical model in which input information including elemental composition of elements in metal and manufacturing conditions and output information including characteristic values of the metal material are associated with each other, and corresponds to the input information. Input the desired characteristic value to the database in which the output data of the mathematical model for the input data for each mesh that divides the input range into multiple sections is stored in association with the input data.
- FIG. 1 is a schematic diagram of a design support device according to the first embodiment.
- 3 is a schematic diagram of a manufacturing process of a steel material according to Embodiment 1.
- FIG. FIG. 3 is a conceptual diagram of creating a mathematical model according to the first embodiment.
- 3 is a conceptual diagram of creating a database according to the first embodiment.
- FIG. 6 is a conceptual diagram of a search process based on a database according to the first embodiment.
- FIG. 3 is a flowchart showing the operation of the design support device according to the first embodiment.
- 9 is a conceptual diagram of creation of a mathematical model according to the second embodiment.
- FIG. FIG. 11 is a conceptual diagram of a search process based on a database according to the third embodiment.
- FIG. 1 is a schematic diagram of a design support device 1 according to the first embodiment of the present invention.
- a design support device 1 according to an embodiment of the present invention includes a data aggregating unit 11, a model creating unit 12, a database creating unit 13, a searching unit 14, and a presenting unit 15. It is a calculator.
- the data aggregating unit 11 aggregates the actual result data relating to the manufacturing of steel materials necessary for creating a mathematical model described later.
- the data aggregating unit 11 may include a communication interface for aggregating the record data.
- the data aggregating unit 11 may receive the record data from a plurality of external devices or the like using a predetermined communication protocol.
- the actual data collected by the data collecting unit 11 includes the component composition of elements in steel, manufacturing conditions, and characteristic values of steel materials.
- the data on the composition of elements of the steel in the steel collected by the data aggregating unit 11 includes the addition ratio of the elements to be melted in the steel in the converter or the secondary refining.
- Such elements include, for example, C, Si, Mn, P, S, Al, N, Cr, V, Sb, Mo, Cu, Ni, Ti, Nb, B, and Ca.
- the manufacturing condition data collected by the data collecting unit 11 includes various conditions in each step of the manufacturing process of the steel material.
- FIG. 2 shows a schematic diagram of a manufacturing process of a steel material.
- raw iron ore is charged into a blast furnace together with limestone and coke to produce molten pig iron.
- Pig iron tapped in the blast furnace is adjusted for components such as carbon in the converter, and finally adjusted for secondary refining.
- a continuous casting machine casts refined steel to produce an intermediate material called a slab.
- a cold-rolled coil that is a product is generated through a plurality of processing steps such as a heating step in a heating furnace, a hot rolling step, a cooling step, a pickling step, a cold rolling step, an annealing step, and a plating step. ..
- the combination of these processing steps differs depending on the product to be manufactured.
- the combination of processing steps has, for example, the following patterns.
- ⁇ Pattern 1 Heating process ⁇ hot rolling process ⁇ cooling process ⁇ Pattern 2: Heating process ⁇ Hot rolling process ⁇ Cooling process ⁇ Pickling process ⁇ Cold rolling process ⁇ Pattern 3: Heating process ⁇ Hot rolling process ⁇ Cooling Process ⁇ Pickling process ⁇ Cold rolling process ⁇ Annealing process ⁇ Pattern 4: Heating process ⁇ Hot rolling process ⁇ Cooling process ⁇ Pickling process ⁇ Cold rolling process ⁇ Annealing process ⁇ Plating process ⁇ Heat treatment process
- the various conditions in each of the above steps include, for example, the following.
- ⁇ Heating process Heating temperature, heating time
- ⁇ Hot rolling process Plate thickness, plate width, cumulative reduction, rolling start temperature, rolling end temperature, Winding temperature, cooling speed-Cooling process: cooling start temperature, cooling speed-Pickling process: pickling chemical concentration, pickling chemical temperature, pickling speed-Cold rolling process: strip thickness, strip width, reduction rate-annealing Process: Heating speed, heating temperature, holding time, cooling speed, cooling method-Plating process: Melting plating temperature, plating adjustment gas spray amount-Heat treatment process: Heating speed, heating temperature
- the data of the characteristic values of the steel material collected by the data collecting unit 11 include, for example, tensile strength, yield stress, elongation, hardness, impact absorption energy, r value, n value, hole expansion rate, and BH amount.
- the characteristic value can be obtained, for example, by carrying out a sampling test for evaluating the characteristic of the steel material from a part of the manufactured steel material product.
- the data aggregating unit 11 manages the aggregated performance data in association with each other. In other words, the data aggregating unit 11 unitarily combines the actual data of the component composition of the elements in the steel, the actual data of the manufacturing conditions, and the actual data of the characteristic values of the steel material for each unit of the manufactured steel material product. , Collect them so that they can be handled.
- the model creating unit 12 creates a mathematical model in which input information including the elemental composition of elements in steel and manufacturing conditions and output information including characteristic values of steel materials are associated with each other, based on the result data collected by the data collecting unit 11. ..
- the input information is information on actual values used to create a mathematical model.
- the output information is information on actual values used to create a mathematical model.
- FIG. 3 shows a conceptual diagram of creation of the mathematical model according to the first embodiment. As shown in FIG. 3, the model creating unit 12 creates a mathematical model by associating the input information with the output information.
- the model creating unit 12 creates the mathematical model by an arbitrary algorithm based on the actual result data.
- input information is represented by three inputs 1 to 3 and output information is represented by one for the sake of simplicity.
- the inputs 1 to 3 respectively correspond to the component composition of the elements in the steel or the manufacturing conditions in the performance data collected by the data collecting unit 11.
- the output corresponds to the characteristic value of the steel material in the actual data.
- the database creation unit 13 creates a database using the mathematical model created by the model creation unit 12.
- FIG. 4 shows a conceptual diagram of creating a database according to the first embodiment.
- the database creation unit 13 defines the input data mesh by dividing the range of the input data into a plurality of sections for the input information, and defines the representative value corresponding to each input data mesh as the input data.
- the representative value may be, for example, a median value in the mesh, or an end value of the upper limit value or the lower limit value, which is representative of the mesh.
- the range of the input data does not have to be the same as the input information which is the actual data.
- the database creation unit 13 also inputs the input data for each mesh determined in this way into the mathematical model created by the model creation unit 12 to obtain output data for each mesh.
- the output data is the output value of the model corresponding to the input data.
- each input data for each mesh can be used as the input data with the median value of the data section defined as the mesh as a representative value, for example.
- the database creating unit 13 creates a database by accumulating and storing, for each mesh, the correspondence between the output data and the input data obtained by inputting the input data for each mesh into the mathematical model. That is, the database creation unit 13 creates a database that stores output data for each mesh that divides the range of input data into a plurality of sections.
- the range of input data that defines the input data mesh is the entire range of input, including the composition and manufacturing conditions of elements in steel that can be assumed as a steel material. That is, the range of input data is limited to a predetermined range based on a predetermined condition such as a metallurgical knowledge or an evaluation function. Table 1 is an example of restrictions on the range of the input data.
- the search unit 14 searches the database for input data corresponding to a certain index, on condition that an arbitrary index matches the output data.
- FIG. 5 shows a conceptual diagram of the search process based on the database according to the first embodiment.
- FIG. 5 shows an example in which one output data item is associated with m or more input data items.
- the search unit 14 searches the database by using the arbitrarily specified desired characteristic value as an index, and as a search result, obtains the component composition of the element in the steel and the manufacturing conditions in a plurality of steps corresponding to the desired characteristic value.
- “y2” is searched as an index as a desired characteristic value
- the "component composition of the element in the steel corresponding to the desired characteristic value and the manufacturing conditions in multiple steps” means the composition of the element in the steel that produces a characteristic value that matches the desired characteristic value designated as an index.
- the composition of elements in the steel that produces a characteristic value similar to the desired characteristic value designated as an index and the manufacturing conditions in a plurality of steps are included.
- “the characteristic values are similar” means that the absolute value of the difference between the characteristic values is small, and the similar range can be defined for each characteristic value. That is, if there is data included in a range similar to the value given as the desired characteristic value, the search result is output as a match.
- FIG. 5 shows an example in which one search result is obtained by using “y2” as an index, but the search result is not limited to this.
- a plurality of search results may be extracted. By doing so, it is possible to obtain a plurality of patterns of the component composition of the elements in the steel and the manufacturing conditions in the plurality of steps that exhibit the desired characteristics, and it is possible to efficiently design the steel material.
- the presenting unit 15 presents to the user the search results searched by the searching unit 14, that is, the component composition and manufacturing conditions of the elements in the steel corresponding to the desired characteristic value.
- the user can efficiently design the steel material by using the component composition of the elements in the steel presented by the presentation unit 15 and the production conditions in the plurality of steps as target values or reference values during the production of the steel material.
- the data aggregating unit 11 aggregates the performance data necessary for creating the mathematical model (step S10).
- the actual data collected by the data collecting unit 11 includes the component composition of the elements in the steel, the manufacturing conditions, and the characteristic value of the steel material related to the manufactured steel material.
- the model creating unit 12 associates the input information including the component composition and manufacturing conditions of the elements in the steel with the output information including the characteristic values of the steel material, based on the actual data collected by the data collecting unit 11, and the mathematical model. Is created (step S20).
- the database creating unit 13 creates a database for supporting the design of the steel material using the mathematical model created by the model creating unit 12 (step S30). Specifically, the database creation unit 13 creates a database in which the output data corresponding to the input data for each mesh obtained by dividing the range of the input data into a plurality of sections is associated with the input data and accumulated.
- the search unit 14 searches the component composition and manufacturing conditions of the elements in the steel corresponding to the desired characteristic value based on the database (step S40).
- the presentation unit 15 presents the component composition and manufacturing conditions of the element in the steel corresponding to the desired characteristic value retrieved by the retrieval unit 14 (step S50).
- design support apparatus 1 instead of performing optimization calculation, a database that stores output data for each mesh that divides a range of input data into a plurality of sections is used. Then, the composition and production conditions of the element in the metal corresponding to the desired characteristic value are searched based on the database, and the composition and production condition of the element in the metal corresponding to the desired characteristic value are presented. That is, according to the design support apparatus 1 according to the first embodiment, design support can be performed without performing optimization calculation, and thus an increase in calculation load related to the design of steel materials can be suppressed.
- the performance data collected by the data collecting unit 11 of the design support device 1 according to the second embodiment includes, in addition to the component composition of the elements in the steel, the manufacturing conditions, and the characteristic value of the steel material, an index indicating the state of the metal structure. ..
- the index indicating the state of the metal structure includes, for example, the grain size and structure fraction of ferrite, the structure fraction of cementite, the structure fraction of pearlite, the structure fraction of bainite, and the structure fraction of martensite. Any method can be adopted as a method of collecting the indexes indicating the state of the metal structure.
- the data aggregating unit 11 may be obtained by performing a sampling test for evaluating an index indicating the state of the metal structure from a part of the manufactured steel material product.
- the data aggregating unit 11 associates the index data thus obtained with the manufacturing data of the steel material product and the characteristics of the steel material.
- the data aggregating unit 11 may be obtained by a measuring device capable of evaluating an index indicating the state of the metal structure during manufacturing.
- the data aggregating unit 11 associates the index data thus obtained with the manufacturing data of the product and the characteristics of the steel material.
- the data aggregating unit 11 may obtain the index indicating the state of the metal structure during manufacturing by a simulation capable of evaluating.
- the data aggregating unit 11 may associate the index data thus obtained with the manufacturing data of the product and the characteristics of the steel material.
- the model creation unit 12 of the design support device 1 provides input information including the component composition of elements in steel, manufacturing conditions, and an index indicating the state of the metal structure, and output information including characteristic values of steel materials. Create related mathematical models.
- FIG. 7 shows a conceptual diagram of creation of a mathematical model according to the second embodiment.
- the input information is represented by three inputs 1 to 3 for the sake of simplicity of description. These inputs 1 to 3 respectively correspond to the component composition of elements in steel, manufacturing conditions, or an index indicating the state of the metal structure.
- the model creating unit 12 creates a mathematical model by associating the input information with the output information.
- the creation of the database by the database creating unit 13 is the same as that of the first embodiment, and thus the description thereof is omitted.
- the search processing by the search unit 14 searches the database by using the arbitrarily specified desired characteristic value as an index, and as a search result, in addition to the component composition and manufacturing conditions of the elements in the steel corresponding to the desired characteristic value, the metal Obtain an index that represents the state of the organization.
- the information presenting process by the presenting unit 15 is the same as that in the first embodiment, and thus the description thereof is omitted.
- the design support device 1 since the data of the state of the metallographic structure, which is a direct factor for expressing the characteristics of the steel material, is used, the accuracy of the mathematical model created can be improved. .. Furthermore, based on the information on the state of the metallographic structure, it is possible to obtain an index indicating the state of the metallographic structure in addition to the component composition and manufacturing conditions of the elements in the steel that produce the desired characteristic value as the search result. , The design accuracy of steel materials can be improved. Therefore, according to the design support apparatus 1 according to the second embodiment, it is possible to accurately obtain the component composition and manufacturing conditions of the elements in the metal that can obtain a desired characteristic value of the characteristics of the steel material, and perform a highly accurate design. You can
- the third embodiment of the present invention will be described below.
- the same components as those in the first embodiment are designated by the same reference numerals and the description thereof will be omitted.
- the design support apparatus 1 according to the third embodiment is different from the configuration according to the first embodiment in that the model creating unit 12 creates a mathematical model for each characteristic value.
- the characteristics of the metal material include, for example, tensile strength, yield stress, elongation, hardness, impact absorption energy, r value, n value, hole expansion ratio, and BH amount, as described in the first embodiment.
- the model creation unit 12 of the design support device 1 according to the third embodiment creates separate mathematical models for each of these multiple types of characteristics. In other words, the model creation unit 12 of the design support device 1 according to the third embodiment creates a plurality of mathematical models.
- the database creation unit 13 creates a database using the plurality of mathematical models created by the model creation unit 12. Specifically, the database creation unit 13 defines the input data mesh by dividing the input range into a plurality of sections, with the component composition and manufacturing conditions of the elements in the steel that can be assumed as the steel material being the entire input range. Here, the range of input data input to the database does not have to match the range of input information.
- the input data is a representative value of each data mesh (similar to the first embodiment).
- the database creation unit 13 inputs the determined input data for each mesh into each of the plurality of mathematical models created by the model creation unit 12 to obtain output data for each mesh.
- the database creation unit 13 accumulates and stores, for each mesh, the correspondence between the output data and the input data obtained by inputting the input data for each mesh into a plurality of mathematical models, and creates a database.
- the database creation unit 13 creates a database that stores output data for each mesh that divides the range of input data into a plurality of sections.
- the search process by the search unit 14 is the same as that of the first embodiment, but the index at the time of search can be specified by a plurality of types of characteristics.
- FIG. 8 shows a conceptual diagram of the search process based on the database according to the third embodiment.
- “y12, y22,...” As a desired characteristic value is searched as an index, and “x12, x22,..., Xm2,...” Is obtained as a search result.
- the presentation unit 15 presents the search result to the user.
- the presentation unit 15 presents the component composition of the elements in the steel and the production conditions corresponding to the plurality of desired characteristic values retrieved by the retrieval unit 14.
- the user can efficiently design the steel material by using the component composition of the elements in the steel and the manufacturing conditions in a plurality of steps presented by the presentation unit 15 as the target value or the reference value at the time of manufacturing the steel material.
- FIG. 9 shows a conceptual diagram of another search process based on the database according to the third embodiment.
- the search unit 14 outputs at least one search result having output data in a range similar to the index. For example, when a part of the plurality of characteristic values among the desired characteristic values, which is an index, partially matches, the search unit 14 presents, as candidates, a characteristic that partially matches and does not match. To do. At this time, there is a method of determining the similarity based on the distance of a vector composed of characteristic values that do not match.
- the search unit 14 outputs, as a search result, data having output data that is most similar to any one of desired characteristic values that are indexes.
- search result candidates “x12, x22,..., Xm2,...” And “x1n, x2n,. ..
- the presentation unit 15 may present the two search results to the user.
- the similar range can be determined by normalizing the values between the respective elements of the vector and then determining that the normalized distance between the vectors is a predetermined distance. It is also possible to make a determination by defining a similar range for each element.
- the design support apparatus 1 it is possible to easily associate the input/output relations such as the composition of elements in the complex steel and the manufacturing conditions in a plurality of processes that exhibit a plurality of characteristics of the steel material. As a result, it becomes possible to efficiently design steel materials.
- Embodiment 4 of the present invention will be described.
- the same components as those in the first embodiment are designated by the same reference numerals and the description thereof will be omitted.
- the design support device 1 according to the fourth embodiment is different from the configuration according to the first embodiment in the content of the result data aggregated by the data aggregation unit 11 and the configuration of the mathematical model created by the model creation unit 12.
- the actual data collected by the data collecting unit 11 of the design support apparatus 1 according to the fourth embodiment includes the composition of elements in steel, the manufacturing conditions, and the characteristic value of the steel material, as well as an index indicating the state of the metal structure. .. Further, the model creating unit 12 associates the input information including the component composition and manufacturing conditions of the elements in the steel with the intermediate output information including the index indicating the state of the metal structure, the intermediate output information and the metal. A second mathematical model associated with output information including material properties is created.
- FIG. 10 shows a conceptual diagram of creation of a mathematical model according to the fourth embodiment. As shown in FIG.
- the index indicating the state of the metal structure is used as the output information (intermediate output information) of the first mathematical model, and the input information and the output information are output via the intermediate output information.
- the first mathematical model and the second mathematical model correspond by the first mathematical model and the second mathematical model.
- the database creation unit 13 creates a database using the plurality of mathematical models created by the model creation unit 12, that is, the first mathematical model and the second mathematical model. Specifically, the database creation unit 13 defines the input data mesh by dividing the input range into a plurality of sections, with the component composition and manufacturing conditions of the elements in the steel that can be assumed as the steel material being the entire input range. Here, the range of input data input to the database does not have to match the range of input information.
- the input data is a representative value of each data mesh (similar to the first embodiment).
- the database creation unit 13 inputs the determined input data for each mesh into the first mathematical model to obtain intermediate output data for each mesh.
- the database creation unit 13 also inputs the intermediate output data to the second mathematical model to obtain output data for each mesh.
- the database creation unit 13 accumulates and stores, for each mesh, the correspondence between the output data and the input data obtained by inputting the input data for each mesh into a plurality of mathematical models, and creates a database.
- the database creation unit 13 creates a database that stores output data for each mesh that divides the range of input data into a plurality of sections.
- the search processing by the search unit 14 for example, using a desired characteristic value as an index, the range of the index indicating the state of the metallographic structure that is the intermediate output is limited to a predetermined range, and the search is performed within the limited range. Using the index indicating the metal structure as an index, the composition and manufacturing conditions of elements in steel are searched.
- a predetermined range of desired characteristic values is used as an index to search a range of an index indicating the state of the metallographic structure, which is an intermediate output, and represents the state of the searched metallographic structure.
- the index range as an index, we search for multiple candidates for the composition of elements in steel and manufacturing conditions.
- the search unit 14 searches the database for the index indicating the state of the metal structure corresponding to the desired characteristic value and the component composition and manufacturing conditions of the elements in the metal corresponding to the index indicating the state of the metal structure. To do.
- the information presenting process by the presenting unit 15 is the same as that in the first embodiment, and thus the description thereof is omitted.
- the design support apparatus 1 according to the fourth embodiment can be created because the information on the state of the metal structure, which is a direct factor for revealing the characteristics of the steel material, can be used as the intermediate output.
- the accuracy of the mathematical model can be improved. Therefore, according to the design support apparatus 1 according to the fourth embodiment, it is possible to accurately obtain the component composition of the elements in the metal and the manufacturing conditions that can obtain the desired characteristic value of the characteristics of the steel material, and perform the highly accurate design. You can
- the fifth embodiment of the present invention will be described below.
- the same components as those in the first embodiment are designated by the same reference numerals and the description thereof will be omitted.
- the design support device 1 according to the fifth embodiment differs from the configuration according to the first embodiment in the input data mesh of the database created by the database creation unit 13.
- FIG. 11 shows a conceptual diagram of database creation according to the fifth embodiment.
- the database creation unit 13 defines the input data mesh by dividing the input range into a plurality of sections with the composition range of the elements in the steel and the manufacturing conditions that can be assumed as the steel material as the entire input range.
- the database creation unit 13 makes the granularity (section width) of each section of the mesh different for each input data.
- the database creation unit 13 may change the section width in advance for each item based on the metallurgical knowledge, for items that are important to be detailed and for items that are not important to be coarse.
- the database creation unit 13 may change the section width based on the data density of each item.
- the database creation unit 13 may finely set the mesh section width in the design of the carbon component composition. Alternatively, the database creation unit 13 may set the section width of each section of the mesh so that the amount of change in output is constant. In other words, the database creation unit 13 may set the section width of the mesh so that the output difference between the adjacent meshes is constant.
- the design support device 1 since only the data of the minimum necessary number of meshes is stored as a database, the calculation load in model creation, the calculation time, the search load in design, and the search time are reduced. It can be reduced. That is, it is possible to avoid a huge calculation load and search load that may occur if the numerical values of the mesh section widths are set to be relatively fine (for example, every 0.001) in all items.
- the numerical value of the mesh section width is relatively rough (for example, every 0.01) in all the items, avoiding a decrease in the design accuracy of the steel material that produces the desired characteristic value, which can occur, It is possible to efficiently design a steel material that exhibits desired characteristic values with high accuracy with a minimum load.
- Table 2 shows an example of the composition of elements in steel that affect the properties.
- Table 3 shows an example of manufacturing conditions that affect the characteristics.
- Table 4 shows the types of characteristics and characteristic values.
- 500 learning data items are used, and a machine learning method called a random forest is used to create mathematical models for predicting tensile strength and elongation as properties.
- 12 and 13 show scatter diagrams related to the actual value and the predicted value.
- the horizontal axis of the scatter diagram in FIG. 12 is the actual value of tensile strength, and the vertical axis is the predicted value of tensile strength.
- the horizontal axis of the scatter diagram of FIG. 13 is the actual value of growth, and the vertical axis is the predicted value of growth.
- the number of regression trees used in the random forest is 50 in each mathematical model.
- RMSE Root Mean Square Error
- the section width of the mesh of the component composition (unit: mass%) of C, P, Al, Sb, Ti, and Nb in the input data is set to every 0.001%
- the component composition of S, N, B, and Ca is set to every 0.0001%
- the section width of the mesh of other component composition (unit: mass%) is set to every 0.01%.
- the desired characteristic values of the characteristics of the steel material used when searching for the design conditions including the learned mathematical models and the section width of the mesh, are complete, the desired characteristic value of the steel material is revealed. It is possible to obtain the component composition of elements in steel and the manufacturing conditions in multiple steps.
- Table 6 shows the input (component composition of elements in steel and manufacturing conditions in multiple steps) obtained by the search.
- the steel products manufactured under these design conditions have a tensile strength of 1200 MPa and an elongation of 12.0%, which means that a steel material exhibiting desired characteristic values could be designed.
- Table 7 shows the search results when the interval width of the mesh of all component compositions in the input data is set to 0.01%.
- the steel product manufactured under the design conditions in this example has a tensile strength of 1240 MPa and an elongation of 11.5%. That is, it is understood that Example 1 is more preferable from the viewpoint of designing a steel material that exhibits desired characteristic values.
- Table 9 and Table 10 show the search results when the desired characteristic values of the characteristics of the steel material are set as shown in Table 8 and read as an index.
- the composition of the elements in the steel and the production conditions for producing the desired characteristic values are not searched, and instead the composition and the production conditions satisfying the characteristics of either tensile strength or elongation are obtained.
- Two candidates are presented. Steel materials can be designed using these candidates as reference values.
- the present invention can also be realized as a program describing the processing content for realizing each function of the above-described design support apparatus 1 or a storage medium recording the program. It should be understood that these are also included in the scope of the present invention.
- the design support device 1 has been shown as an example including the data aggregating unit 11 and the model creating unit 12, but these may be realized by another information processing device.
- the information processing apparatus aggregates the actual data required to create the mathematical model and creates the mathematical model.
- the information processing device transmits the created mathematical model to the design support device 1.
- another information processing apparatus may include a database creation unit 13 in addition to the data aggregation unit 11 and the model creation unit 12. In this case, such an information processing device may create a database and transmit the database to the design support device 1.
Abstract
Description
計算機により金属材料の設計を支援する設計支援方法であって、
金属中元素の成分組成及び製造条件を含む入力情報と前記金属材料の特性値を含む出力情報とを関係付けた少なくとも1つの数理モデルを用いて作成されたものであり、かつ前記入力情報に対応する入力の範囲を複数区間で区切ったメッシュ毎の入力データに対する前記数理モデルの出力データが前記入力データと関係付けて蓄積されているデータベースに対して、所望の特性値を入力し、金属中元素の成分組成及び製造条件を検索する検索ステップと、
前記検索ステップにより検索された、前記所望の特性値に対応する金属中元素の成分組成及び製造条件を提示する提示ステップと
を含む。
金属材料の設計を支援する設計支援装置であって、
金属中元素の成分組成及び製造条件を含む入力情報と前記金属材料の特性値を含む出力情報とを関係付けた少なくとも1つの数理モデルを用いて作成されたものであり、かつ前記入力情報に対応する入力の範囲を複数区間で区切ったメッシュ毎の入力データに対する前記数理モデルの出力データが前記入力データと関係付けて蓄積されているデータベースに対して、所望の特性値を入力し、金属中元素の成分組成及び製造条件を検索する検索部と、
前記検索部により検索された、前記所望の特性値に対応する金属中元素の成分組成及び製造条件を提示する提示部と
を有する。
以下、本発明の実施形態1について説明する。本実施形態において設計する金属材料は、鉄鋼である例について説明する。しかしながら、金属材料は鉄鋼に限られず、任意の金属に適用可能である。
図1は本発明の実施形態1に係る設計支援装置1の概要図である。図1に示すように、本発明の一実施形態に係る設計支援装置1は、データ集約部11と、モデル作成部12と、データベース作成部13と、検索部14と、提示部15とを備える計算機である。
・パターン1:加熱工程→熱間圧延工程→冷却工程
・パターン2:加熱工程→熱間圧延工程→冷却工程→酸洗工程→冷間圧延工程
・パターン3:加熱工程→熱間圧延工程→冷却工程→酸洗工程→冷間圧延工程→焼鈍工程
・パターン4:加熱工程→熱間圧延工程→冷却工程→酸洗工程→冷間圧延工程→焼鈍工程→鍍金工程→熱処理工程
・加熱工程:加熱温度、加熱時間
・熱間圧延工程:板厚、板幅、累積圧下率、圧延開始温度、圧延終了温度、
巻取温度、冷却速度
・冷却工程:冷却開始温度、冷却速度
・酸洗工程:酸洗薬液濃度、酸洗薬液温度、酸洗速度
・冷間圧延工程:板厚、板幅、圧下率
・焼鈍工程:加熱速度、加熱温度、保持時間、冷却速度、冷却方法
・鍍金工程:溶融鍍金温度、鍍金調整ガス吹付量
・熱処理工程:加熱速度、加熱温度
以下、本発明の実施形態2について説明する。実施形態1と同一の構成については同一の符号を付し、説明は省略する。実施形態2に係る設計支援装置1は、実施形態1に係る構成と比較して、データ集約部11が集約する実績データの内容が相違する。
以下、本発明の実施形態3について説明する。実施形態1と同一の構成については同一の符号を付し、説明は省略する。実施形態3に係る設計支援装置1は、実施形態1に係る構成と比較して、モデル作成部12により特性値毎に数理モデルを作成する点が相違する。
以下、本発明の実施形態4について説明する。実施形態1と同一の構成については同一の符号を付し、説明は省略する。実施形態4に係る設計支援装置1は、実施形態1に係る構成と比較して、データ集約部11が集約する実績データの内容、及びモデル作成部12が作成する数理モデルの構成が相違する。
以下、本発明の実施形態5について説明する。実施形態1と同一の構成については同一の符号を付し、説明は省略する。実施形態5に係る設計支援装置1は、実施形態1に係る構成と比較して、データベース作成部13により作成するデータベースの入力データメッシュが相違する。
以下、自動車用冷延鋼板についての鉄鋼材料の設計の例を示す。本実施例では鉄鋼材料の特性として引張強度と伸びを選択し、所望の特性値を現出するような設計条件を検索する。
11 データ集約部
12 モデル作成部
13 データベース作成部
14 検索部
15 提示部
Claims (8)
- 計算機により金属材料の設計を支援する設計支援方法であって、
金属中元素の成分組成及び製造条件を含む入力情報と前記金属材料の特性値を含む出力情報とを関係付けた少なくとも1つの数理モデルを用いて作成されたものであり、かつ前記入力情報に対応する入力の範囲を複数区間で区切ったメッシュ毎の入力データに対する前記数理モデルの出力データが前記入力データと関係付けて蓄積されているデータベースに対して、所望の特性値を入力し、金属中元素の成分組成及び製造条件を検索する検索ステップと、
前記検索ステップにより検索された、前記所望の特性値に対応する金属中元素の成分組成及び製造条件を提示する提示ステップと
を含む、設計支援方法。 - 前記入力情報は、金属組織の状態を表す指標を含む、請求項1に記載の設計支援方法。
- 前記データベースは複数の数理モデルを用いて作成されたものであり、
前記複数の数理モデルは、前記金属材料の特性の種類毎にそれぞれ作成される、請求項1又は2に記載の設計支援方法。 - 計算機により金属材料の設計を支援する設計支援方法であって、
金属中元素の成分組成及び製造条件を含む入力情報と金属組織の状態を表す指標を含む中間出力情報とを関係付けた少なくとも1つの第1数理モデルと、前記中間出力情報と前記金属材料の特性値を含む出力情報とを関係付けた少なくとも1つの第2数理モデルとを用いて作成されたものであり、かつ前記入力情報に対応する入力の範囲を複数区間で区切ったメッシュ毎の入力データに対する前記第1数理モデルの中間出力データ及び前記第2数理モデルの出力データが前記入力データと関係付けて蓄積されているデータベースに対して、所望の特性値を入力し、所望の特性値に対応する金属組織の状態を表す指標と、該金属組織の状態を表す指標に対応する金属中元素の成分組成及び製造条件とを前記データベースに基づき検索する検索ステップと、
前記所望の特性値に対応する金属中元素の成分組成及び製造条件を提示する提示ステップと
を含む、設計支援方法。 - 前記メッシュの各区間の区間幅を入力情報毎に異ならせる、請求項1乃至4のいずれか一項に記載の設計支援方法。
- 前記メッシュの各区間の区間幅を出力の変化量が一定になるように定める、請求項1乃至4のいずれか一項に記載の設計支援方法。
- 前記入力情報の範囲を予め定めた条件に基づき所定の範囲に制限する、請求項1乃至6のいずれか一項に記載の設計支援方法。
- 金属材料の設計を支援する設計支援装置であって、
金属中元素の成分組成及び製造条件を含む入力情報と前記金属材料の特性値を含む出力情報とを関係付けた少なくとも1つの数理モデルを用いて作成されたものであり、かつ前記入力情報に対応する入力の範囲を複数区間で区切ったメッシュ毎の入力データに対する前記数理モデルの出力データが前記入力データと関係付けて蓄積されているデータベースに対して、所望の特性値を入力し、金属中元素の成分組成及び製造条件を検索する検索部と、
前記検索部により検索された、前記所望の特性値に対応する金属中元素の成分組成及び製造条件を提示する提示部と
を有する、設計支援装置。
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
MX2021008618A MX2021008618A (es) | 2019-01-17 | 2019-02-19 | Metodo auxiliar de dise?o y dispositivo auxiliar de dise?o para material metalico. |
EP19910610.5A EP3913635A4 (en) | 2019-01-17 | 2019-02-19 | METALLIC MATERIAL DESIGN AIDING METHOD AND DESIGN AIDING DEVICE |
BR112021013351-8A BR112021013351A2 (pt) | 2019-01-17 | 2019-02-19 | Método de auxílio de desenho de material metálico e dispositivo de auxílio de desenho |
CN201980089027.3A CN113330440A (zh) | 2019-01-17 | 2019-02-19 | 金属材料的设计辅助方法和设计辅助装置 |
US17/423,566 US20220083700A1 (en) | 2019-01-17 | 2019-02-19 | Design aid method and design aid device for metallic material |
KR1020217025533A KR20210110716A (ko) | 2019-01-17 | 2019-02-19 | 금속 재료의 설계 지원 방법 및 설계 지원 장치 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2019006145A JP6617842B1 (ja) | 2019-01-17 | 2019-01-17 | 金属材料の設計支援方法及び設計支援装置 |
JP2019-006145 | 2019-01-17 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020148918A1 true WO2020148918A1 (ja) | 2020-07-23 |
Family
ID=68836093
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2019/006147 WO2020148918A1 (ja) | 2019-01-17 | 2019-02-19 | 金属材料の設計支援方法及び設計支援装置 |
Country Status (8)
Country | Link |
---|---|
US (1) | US20220083700A1 (ja) |
EP (1) | EP3913635A4 (ja) |
JP (1) | JP6617842B1 (ja) |
KR (1) | KR20210110716A (ja) |
CN (1) | CN113330440A (ja) |
BR (1) | BR112021013351A2 (ja) |
MX (1) | MX2021008618A (ja) |
WO (1) | WO2020148918A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022117664A1 (en) * | 2020-12-04 | 2022-06-09 | Thiry Cedric Robert | Computer implemented engineering materials mechanical property based search method |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7200982B2 (ja) * | 2020-09-14 | 2023-01-10 | Jfeスチール株式会社 | 材料特性値予測システム及び金属板の製造方法 |
JPWO2023002951A1 (ja) | 2021-07-21 | 2023-01-26 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS565090B1 (ja) | 1969-10-03 | 1981-02-03 | ||
JPH08255190A (ja) * | 1995-03-17 | 1996-10-01 | Sekisui Chem Co Ltd | 粘着剤及び粘着テープの設計支援システム |
JP2003328030A (ja) * | 2002-03-08 | 2003-11-19 | Jfe Steel Kk | 鋼材の製品品質設計装置及び最適品質設計支援装置 |
JP2007047872A (ja) * | 2005-08-05 | 2007-02-22 | Mitsubishi Heavy Ind Ltd | 材料の選定方法及び化学成分判定方法とそれらのシステム |
JP4393586B2 (ja) | 1996-08-08 | 2010-01-06 | 株式会社ブリヂストン | 多成分系材料の設計方法、最適化解析装置及び多成分系材料の最適化解析プログラムを記録した記録媒体 |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4189908B2 (ja) * | 2002-04-26 | 2008-12-03 | 株式会社神戸製鋼所 | 溶接部の材質予測方法 |
JP4009670B2 (ja) * | 2002-08-02 | 2007-11-21 | 独立行政法人科学技術振興機構 | 成分配合設計方法、成分配合設計プログラム及びそのプログラムを記録した記録媒体 |
DE10339595A1 (de) * | 2003-08-26 | 2005-04-07 | Siemens Ag | Verfahren zur Vorhersage und Steuerung der Vergießbarkeit von Flüssigstahl |
CN100334240C (zh) * | 2005-08-05 | 2007-08-29 | 武汉大学 | 镍基高温合金成分的优化设计方法 |
US20080255811A1 (en) * | 2007-04-13 | 2008-10-16 | Zi Qiang Sheng | System and Method for Determining Surface Roughness |
US8137483B2 (en) * | 2008-05-20 | 2012-03-20 | Fedchun Vladimir A | Method of making a low cost, high strength, high toughness, martensitic steel |
JP2014038595A (ja) * | 2012-07-20 | 2014-02-27 | Jfe Steel Corp | 鋼材の材質予測装置及び材質制御方法 |
EP3055802B1 (en) * | 2013-10-10 | 2023-12-06 | Oerlikon Metco (US) Inc. | Methods of selecting material compositions and designing materials having a target property |
DE102014224461A1 (de) * | 2014-01-22 | 2015-07-23 | Sms Siemag Ag | Verfahren zur optimierten Herstellung von metallischen Stahl- und Eisenlegierungen in Warmwalz- und Grobblechwerken mittels eines Gefügesimulators, -monitors und/oder -modells |
GB2536939A (en) * | 2015-04-01 | 2016-10-05 | Isis Innovation | Method for designing alloys |
JP6756997B2 (ja) * | 2016-05-26 | 2020-09-16 | ファイフィット株式会社 | 有限要素法解析方法、有限要素法解析装置、解析サービスシステムおよび有限要素法解析プログラムを記録した記録媒体 |
WO2020090848A1 (ja) * | 2018-10-30 | 2020-05-07 | 昭和電工株式会社 | 材料設計装置、材料設計方法、及び材料設計プログラム |
-
2019
- 2019-01-17 JP JP2019006145A patent/JP6617842B1/ja active Active
- 2019-02-19 EP EP19910610.5A patent/EP3913635A4/en active Pending
- 2019-02-19 MX MX2021008618A patent/MX2021008618A/es unknown
- 2019-02-19 WO PCT/JP2019/006147 patent/WO2020148918A1/ja unknown
- 2019-02-19 BR BR112021013351-8A patent/BR112021013351A2/pt unknown
- 2019-02-19 US US17/423,566 patent/US20220083700A1/en active Pending
- 2019-02-19 KR KR1020217025533A patent/KR20210110716A/ko not_active Application Discontinuation
- 2019-02-19 CN CN201980089027.3A patent/CN113330440A/zh active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS565090B1 (ja) | 1969-10-03 | 1981-02-03 | ||
JPH08255190A (ja) * | 1995-03-17 | 1996-10-01 | Sekisui Chem Co Ltd | 粘着剤及び粘着テープの設計支援システム |
JP4393586B2 (ja) | 1996-08-08 | 2010-01-06 | 株式会社ブリヂストン | 多成分系材料の設計方法、最適化解析装置及び多成分系材料の最適化解析プログラムを記録した記録媒体 |
JP2003328030A (ja) * | 2002-03-08 | 2003-11-19 | Jfe Steel Kk | 鋼材の製品品質設計装置及び最適品質設計支援装置 |
JP2007047872A (ja) * | 2005-08-05 | 2007-02-22 | Mitsubishi Heavy Ind Ltd | 材料の選定方法及び化学成分判定方法とそれらのシステム |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022117664A1 (en) * | 2020-12-04 | 2022-06-09 | Thiry Cedric Robert | Computer implemented engineering materials mechanical property based search method |
Also Published As
Publication number | Publication date |
---|---|
EP3913635A1 (en) | 2021-11-24 |
JP2020115258A (ja) | 2020-07-30 |
JP6617842B1 (ja) | 2019-12-11 |
EP3913635A4 (en) | 2022-03-02 |
BR112021013351A2 (pt) | 2021-09-14 |
MX2021008618A (es) | 2021-11-04 |
KR20210110716A (ko) | 2021-09-08 |
CN113330440A (zh) | 2021-08-31 |
US20220083700A1 (en) | 2022-03-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xie et al. | Online prediction of mechanical properties of hot rolled steel plate using machine learning | |
WO2020148918A1 (ja) | 金属材料の設計支援方法及び設計支援装置 | |
KR101011546B1 (ko) | 예측식 작성장치 및 예측식 작성방법 | |
JP4855353B2 (ja) | 製品の品質改善条件解析装置、解析方法、コンピュータプログラム、及びコンピュータ読み取り可能な記録媒体 | |
JP7028316B2 (ja) | 金属材料の設計支援方法、予測モデルの生成方法、金属材料の製造方法、及び設計支援装置 | |
CN114611844B (zh) | 一种转炉出钢过程合金加入量的确定方法和系统 | |
CN106345823B (zh) | 基于热轧钢卷生产流程的在线实时预测机械性能的方法 | |
CN106802977A (zh) | 一种用于烧结矿性能指标预测及综合质量评价方法 | |
WO2021004198A1 (zh) | 一种板材性能的预测方法及装置 | |
Mohanty et al. | Online mechanical property prediction system for hot rolled IF steel | |
Dobrzański et al. | Application of neural networks for the prediction of continuous cooling transformation diagrams | |
Feng et al. | Endpoint temperature prediction of molten steel in RH using improved case-based reasoning | |
Geng et al. | A data-driven machine learning approach to predict the hardenability curve of boron steels and assist alloy design | |
Zhao et al. | Prediction of mechanical properties of cold rolled strip based on improved extreme random tree | |
Gupta et al. | A machine learning model for multi-class classification of quenched and partitioned steel microstructure type by the k-nearest neighbor algorithm | |
JP5682131B2 (ja) | 鋼材の材質予測装置 | |
CN116469481B (zh) | 一种基于XGBoost算法的LF精炼钢水成分预报方法 | |
CN115456264B (zh) | 一种中小型转炉的终点碳含量和终点温度预测方法 | |
Dong et al. | Just-in-time learning-based soft sensor for mechanical properties of strip steel via multi-block weighted semisupervised models | |
JP2007122127A (ja) | 鉄鋼製品の生産計画装置及び方法 | |
Peet et al. | Neural network modelling of hot deformation of austenite | |
Kudrya et al. | On necessity of taking into account statistical nature of the objects using Big Data in metallurgy | |
Vijay Reddy et al. | Influence of carbon equivalent content on phase transformation during inter-critical heating of dual phase steels using discrete micro-scale cellular automata model | |
US20230145099A1 (en) | Product information determining method, manufacturing method, system and product information determining device | |
Jun et al. | Prediction of hot metal temperature based on data mining |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
REG | Reference to national code |
Ref country code: BR Ref legal event code: B01A Ref document number: 112021013351 Country of ref document: BR |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 20217025533 Country of ref document: KR Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2019910610 Country of ref document: EP Effective date: 20210817 |
|
ENP | Entry into the national phase |
Ref document number: 112021013351 Country of ref document: BR Kind code of ref document: A2 Effective date: 20210706 |