WO2007080688A1 - 予測式作成装置及び予測式作成方法 - Google Patents
予測式作成装置及び予測式作成方法 Download PDFInfo
- Publication number
- WO2007080688A1 WO2007080688A1 PCT/JP2006/321450 JP2006321450W WO2007080688A1 WO 2007080688 A1 WO2007080688 A1 WO 2007080688A1 JP 2006321450 W JP2006321450 W JP 2006321450W WO 2007080688 A1 WO2007080688 A1 WO 2007080688A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- prediction
- prediction formula
- manufacturing
- result
- target
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims description 55
- 238000004519 manufacturing process Methods 0.000 claims abstract description 157
- 238000004364 calculation method Methods 0.000 claims abstract description 18
- 238000011156 evaluation Methods 0.000 claims description 43
- 238000013461 design Methods 0.000 claims description 33
- 238000012938 design process Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 description 38
- 230000006870 function Effects 0.000 description 33
- 238000005096 rolling process Methods 0.000 description 24
- 238000010586 diagram Methods 0.000 description 17
- 239000000463 material Substances 0.000 description 15
- 238000007796 conventional method Methods 0.000 description 14
- 238000005457 optimization Methods 0.000 description 11
- 238000001816 cooling Methods 0.000 description 8
- 229910000831 Steel Inorganic materials 0.000 description 7
- 238000013213 extrapolation Methods 0.000 description 7
- 239000010959 steel Substances 0.000 description 7
- 238000004088 simulation Methods 0.000 description 6
- 239000000126 substance Substances 0.000 description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 230000007547 defect Effects 0.000 description 5
- 230000007423 decrease Effects 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 239000000203 mixture Substances 0.000 description 4
- 230000000704 physical effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000010438 heat treatment Methods 0.000 description 3
- 238000002789 length control Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 229910052796 boron Inorganic materials 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 229910052804 chromium Inorganic materials 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 229910052802 copper Inorganic materials 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 229910052748 manganese Inorganic materials 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 229910052750 molybdenum Inorganic materials 0.000 description 2
- 238000000465 moulding Methods 0.000 description 2
- 229910052759 nickel Inorganic materials 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 229910052720 vanadium Inorganic materials 0.000 description 2
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- 238000012356 Product development Methods 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 239000000498 cooling water Substances 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 229910052758 niobium Inorganic materials 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 229910052717 sulfur Inorganic materials 0.000 description 1
- 229910052719 titanium Inorganic materials 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/4181—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by direct numerical control [DNC]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
Definitions
- the present invention relates to a prediction formula creation apparatus and a prediction formula creation method, and more particularly, a prediction formula suitable for use in a factory that builds product quality by heating, rolling, cooling, and heat-treating forged steel materials.
- the present invention relates to a creation device and a prediction formula creation method. Background art
- the request point When predicting the result for the value of the manufacturing condition for which the result is to be predicted (hereinafter referred to as the request point) based on the production database storing the production condition and the result, conventionally, as shown in Fig. 1, the actual database 1 A method has been proposed for calculating the degree of similarity to the required point of each sample of 0. Based on the degree of similarity, the average value calculation, regression equation creation, and neural network are used to predict the result for the required point (See Patent Documents 1 to 3).
- Patent Document 1 Japanese Patent Laid-Open No. 2 0 0 1-2 9 0 5 0 8
- Patent Document 2 Japanese Patent Laid-Open No. 2100, 2—1 5 7 5 7 2
- Patent Document 3 Japanese Patent Laid-Open No. 2 0 0 4-3 5 5 1 8 9
- the results here include dimensions (thickness, width, length, etc.), material (tensile strength / yield point, elongation, toughness, etc.), quality characteristics such as shape, and defect occurrence rate such as defect contamination rate, In addition, it indicates production process indicators such as production efficiency, lead time (time from order receipt to delivery), and manufacturing cost.
- the conventional method has good prediction accuracy in the interpolated area where the actual data exists, but the prediction accuracy in the extrapolated area where the actual data does not exist is good. Had the problem.
- the present invention has been made to solve the above-mentioned conventional problems, and an object thereof is to predict the result of extrapolation area with high accuracy.
- the present invention provides a prediction formula creation apparatus comprising the following: The manufacturing conditions of a product manufactured in the past are associated with the result information of the manufacturing, the manufacturing conditions and the A performance database that stores a plurality of information associated with the result information; and
- a similarity calculation means for comparing the manufacturing conditions stored in the record database with the manufacturing conditions to be predicted and calculating a similarity based on a plurality of comparison results
- the prediction formula creation means creates the relationship between the production condition and the production result based on the production condition and result information of the actual database, the similarity is used as the weight of the evaluation function for evaluating the modeling error.
- the prediction formula creation device of the present invention preferably has a result prediction device.
- the result prediction apparatus is a result of predicting a result for the manufacturing condition by inputting a prediction formula acquisition unit that acquires a prediction formula corresponding to the manufacturing condition of the prediction target, and the manufacturing condition of the prediction target to the prediction formula Prediction means.
- the prediction formula creation device of the present invention preferably further includes a control device for controlling manufacturing conditions.
- the control device uses a prediction formula acquisition unit that acquires a prediction formula corresponding to the manufacturing condition of the prediction target and the prediction formula, and calculates an operation amount at which the control amount becomes a target value with respect to the manufacturing condition of the prediction target. And control means for executing the control.
- the prediction formula creation device of the present invention preferably further includes a quality design device.
- a prediction expression acquiring means for acquiring a prediction expression corresponding to the manufacturing condition of the prediction target; and an output of a prediction result obtained by inputting one or more manufacturing conditions into the prediction expression.
- Quality output auxiliary means for outputting at least one of the secondary evaluation indices calculated based on the above and assisting the quality design of the product.
- the present invention provides a prediction formula creation method comprising:
- the manufacturing conditions stored in the performance database that stores the manufacturing conditions of the products manufactured in the past and the manufacturing result information, and stores the manufacturing conditions and the information associated with the result information, and the prediction target
- Prediction formula creation process based on the manufacturing point corresponding to the manufacturing condition of the prediction target; the prediction formula creation process converts the relationship between the manufacturing condition and the manufacturing result into the manufacturing condition and result information of the actual database.
- the prediction formula creation method of the present invention preferably further includes a result prediction step.
- the result predicting step inputs a prediction formula corresponding to the manufacturing condition of the prediction target, a prediction formula acquiring step, and the manufacturing condition of the prediction target into the prediction formula to predict a result for the manufacturing condition A result prediction step.
- the prediction formula creation method of the present invention preferably further includes a control step for controlling manufacturing conditions.
- the control step uses a prediction formula acquisition step that acquires a prediction formula corresponding to the manufacturing condition of the prediction target and the prediction formula, and calculates an operation amount at which the control amount becomes a target value with respect to the manufacturing condition of the prediction target. And a control process for executing the control.
- the prediction formula creation method of the present invention preferably further includes a quality design process.
- the quality design step includes a prediction formula acquisition step of acquiring a prediction formula corresponding to the manufacturing condition to be predicted, an output of a prediction result obtained by inputting one or more manufacturing conditions into the prediction formula, the prediction result
- a quality design assistance process is provided to output at least one of the outputs of the secondary evaluation index calculated based on the above and assist the product quality design.
- the present invention provides a method for manufacturing a product made by the above method.
- the manufacturing conditions to be associated with the actual length value in the results database are as follows:
- model parameter associated with the crop length prediction value in the prediction value creation means is:
- the constraint conditions for predicting the Charbi absorption energy VE of the thick plate using the material prediction model are as follows:
- the constraint conditions for determining the plate thickness correction amount are the roll rotation speed, the maximum rolling speed of the roll diameter roll, and the constraint conditions for operating these.
- Figure 1 is a diagram showing the concept of a database-type prediction model.
- Figure 2 is a diagram showing the prediction accuracy of the conventional method.
- FIG. 3 is a flowchart showing the procedure of local regression according to the present invention.
- FIG. 4 is a flowchart showing the control means and the quality design procedure according to the present invention.
- FIG. 5 is a schematic diagram of the evaluation method in the present invention.
- FIG. 6 is a diagram showing a comparison of prediction errors of Charpy absorbed energy between the conventional method and the method of the present invention.
- FIG. 7 is also a diagram showing a comparison of prediction errors in tensile properties.
- FIG. 8 is a schematic diagram of crop length prediction and control in the third embodiment.
- Figure 9 is a diagram showing (a) crop shape and representative crop length, and (b) slab shape and plate thickness correction amount.
- FIG. 10 is a diagram showing a comparison between the actual value of the plate thickness correction amount and the value obtained by the method of the present invention and the conventional method.
- FIG. 11 is a histogram of the crop length evaluation value in the method of the present invention and the actual value.
- FIG. 12 is a block diagram showing a basic configuration in the fourth embodiment of the present invention.
- Fig. 13 is a block diagram showing the current steel plate material quality design.
- FIG. 14 is also an explanatory diagram.
- 'Fig. 15 is a diagram showing an example of change in the coefficient of influence on the strength.
- FIG. 16 is a block diagram showing a detailed configuration of the fourth embodiment.
- FIGS. 17 are diagrams showing an example of a decision support screen in the fourth embodiment.
- Fig. 18 is a diagram showing an example of the initial state of the decision support support screen in the strength design embodiment.
- FIG. 19 is a diagram showing a state where the component C concentration is lowered in the state of FIG.
- FIG. 20 is also a diagram showing a state in which the component C concentration is further lowered.
- the present invention associates the manufacturing conditions of products manufactured in the past with the manufacturing result information, compares the results database storing a plurality of these information with the manufacturing conditions stored in the results database, Similarity calculation means for calculating the similarity based on multiple comparison results, and the modeling error is evaluated when creating a prediction formula that expresses the relationship between the manufacturing conditions and manufacturing results in the vicinity of the manufacturing conditions to be predicted.
- the similarity is used as the weight of the evaluation function to be calculated, and the prediction formula is obtained by calculating the parameter so that the value of the evaluation function is minimized within the constraint condition with the physical characteristic of the prediction target as the constraint condition.
- the present invention provides a control apparatus .14 characterized by controlling an object. (See Figure 4). Furthermore, the present invention also provides a design device 16 characterized in that the manufacturing conditions of the object are designed according to the result predicted using the prediction formula creating means. It is. (See also Figure 4).
- the prediction formula obtained by the present invention guarantees the physical characteristics of the object, the prediction accuracy is improved even in the extrapolation region.
- the physical characteristics of interest are qualitative characteristics related to metallurgical phenomena (adding more chemical component C of the material increases the strength of the product but decreases the toughness. Lowering the finishing temperature during rolling reduces the product's physical properties.
- Qualitative characteristics (such as increased strength) and plastic processing reduces the roll gap of the rolling mill during rolling reduces the thickness of the product. Increase the difference between the inlet thickness and the outlet thickness of the rolling mill. This refers to characteristics derived from physical phenomena such as an increase in the load on the rolling mill. .
- the conventional method has a weak point that the prediction accuracy deteriorates sharply when the neighborhood data becomes scarce.
- the present invention does not deteriorate the prediction accuracy even if the neighborhood data becomes scarce. Good prediction accuracy can be obtained.
- control accuracy is improved because there is no operation in the wrong direction, that is, against the physical characteristics.
- quality design using this prediction formula improves the prediction accuracy, so the number of experiments can be reduced and the development cost can be reduced, and the chance loss associated with the experiment can be reduced, and the manufacturing cost can be reduced.
- the present invention as shown in Fig. 3, (1) defines a distance function, calculates the similarity between each observation data in the actual data base 10 and the requested point, and (2) weights the similarity
- a prediction formula near the required point is created by weighted regression.
- the model parameters of the prediction formula in (2) are weighted with modeling errors.
- the target physical It is obtained by solving the quadratic programming method, which is a kind of mathematical programming method, with the physical characteristics (for example, qualitative characteristics of metallurgical phenomena) as constraints.
- a regression equation is created using the given N observation data.
- the regression equation is the following linear equation. .
- the model parameters b, ai, a 2 ,..., A M are obtained by the method of least squares. Partial regression coefficient vector
- the distance L from the required point in is defined as follows. This equation is the distance function.
- the partial regression coefficient can be considered as the contribution of each input variable to the amount of change in the output variable. This is a weighted distance that takes into account its contribution.
- nth (n The distance from the required point of the observation data of 1; 2,- ⁇ ⁇ , ⁇ ) can be obtained from the following equation.
- L n LO,, ⁇ (6)
- the distance from the request point of the 1st to ⁇ th observation data is summarized and expressed as follows.
- ⁇ ⁇ represents the standard deviation of Z
- ⁇ is an adjustment parameter (initial value: 1.5).
- the similarity is 1 when the distance L is 0, that is, the manufacturing conditions are exactly the same as the required conditions, the similarity decreases as the distance increases, and the similarity decreases to 0 when the distance becomes infinite. Is defined.
- the similarity from the required point is obtained for each of the observed data.
- Similarity is an index that evaluates the closeness between required points and observation data under manufacturing conditions (input variables).
- the distance is defined, the distance between the requested point and each observation data is calculated, and the similarity is calculated based on the distance between each observation data.
- the distance function is a force using a weighted first-order norm (sum of absolute values) that takes into account the effect of each manufacturing condition on the result; the Euclidian distance, the normalized Euclidean distance, the Mahalanobis distance, Etc. may be used.
- the Gauss function is used as a function to convert distance to similarity, but it is monotonous for various distances such as ⁇ Tri-cube function. A changing continuous function may be used. JP-A-6-9 5 8 As described in 80, the value of each input variable in the condition part may be discretized into categories, and the discretized distance may be used as the similarity.
- Modeling error is the difference between the output predicted value ⁇ and the output actual value a calculated by substituting the actual value of each observation data input into the prediction equation with model parameters.
- the model parameter is obtained by formulating an optimization problem with the weighted square sum of the modeling error e as the evaluation function and the physical characteristics of the object as constraints.
- ⁇ is a diagonal matrix with similarity w.
- the target physical characteristics related to model parameters are entered as upper and lower limits as follows. '
- the decision parameter for the optimization problem is model parameter 0.
- 'Since Formula (18) is formulated as an optimization function and Formula (20) is defined as an optimization problem
- model parameter 0 is calculated by using the optimization method.
- the weighted sum of squares of the modeling error is used as the evaluation function, and the upper and lower limits of the model parameter ⁇ are used as constraints, and the formulation is formulated for the purpose of minimizing the evaluation function. Since this problem is a quadratic programming problem, model parameter 0 can be obtained by using quadratic programming.
- the optimization problem formulation method and optimization method are not limited to these.
- the evaluation function not only a weighted square sum of modeling errors but also other calculation formulas such as a sum of absolute values may be used.
- the present invention can be applied not only to the upper and lower limit values of the model parameter ⁇ but also to an equation such as an equation or an inequality that expresses a physical characteristic.
- this method is applicable even when using other mathematical programming methods (linear programming, convex programming, nonlinear programming), genetic algorithms, and simulated methods *.
- the invention can be applied.
- an operation for controlling the target result to the target value under a condition in which one of a plurality of manufacturing conditions is set as an operating variable and the manufacturing conditions other than the operating variable are given.
- the controlled variable target value, the reference value of the manipulated variable, and the actual manufacturing condition values other than the manipulated variable are externally given to the controller.
- the reference value of the variable and the actual manufacturing condition value other than the manipulated variable are given as a required point to the prediction formula creation means.
- the model parameters of the prediction formula at the required point are obtained according to the flow in Fig.
- (iv) is calculated as follows.
- Equation (12) Equation (12)
- Xi 1 / ai X ⁇ Y — (b + a 2 ⁇ X 2 + ⁇ ⁇ + a M ⁇ XM) ⁇ ⁇ ⁇ ⁇ (23)
- Y on the right side of this equation is the control target value
- [b, ai, a2, ⁇ , a M ] is the model parameter of the prediction formula
- [X 2 , ⁇ , X M ] is the production other than the manipulated variable
- a! Is determined so as to satisfy the physical characteristics. Therefore, the amount of change in the manipulated variable is also calculated so as to satisfy the physical characteristics. Especially, the actual data does not exist in the vicinity of the required point. In the extrapolation area where the calculation accuracy of the model parameters is not good, the accuracy of the manipulated variable change is improved.
- the designer when supporting quality design, the designer inputs the manufacturing conditions and calculates and displays the predicted quality values for the manufacturing conditions by a computer. Based on the display result, the designer repeatedly inputs the manufacturing conditions and obtains the manufacturing conditions that make the quality a predetermined value.
- the present invention is used for this work, as shown in Fig. 4, (i) the value of the manufacturing condition is input to the design device, and (ii) the value of the manufacturing condition is given to the prediction formula creation means as a required point. (iii) Based on the constraint on the model parameters based on the value of the requested point and the physical characteristics of the object input from the outside, the model parameter of the prediction formula in the request is obtained according to the flow of Fig. 3, and it is returned to the controller.
- (Iv) Calculate the predicted value of the result using Equation (12) based on the model parameters of the prediction equation and the manufacturing conditions, and then calculate and output the secondary evaluation index to the designer. Display.
- secondary evaluation indexes are results other than quality (manufacturing cost, quality defect occurrence rate, production efficiency / lead time, risk, etc.).
- the prediction formula creation means, control device, design device, results database, and constraint condition creation means in Fig. 4 consist of computers, each of which includes an arithmetic processing unit (consisting of a CPU, work RAM, ROM, etc.), various programs, and various data.
- Storage unit for example, HDD (Hard Disk Drive), etc.
- operation unit for inputting operation instructions from the user (for example, keyboard, mouse, etc.)
- display unit for displaying information such as images and characters (for example, , Liquid crystal display, etc.)
- a communication unit that controls the communication state between devices via a network (LAN (Local Area Network), WAN (Wide Area Network), Intranet, etc.).
- LAN Local Area Network
- WAN Wide Area Network
- Intranet a network
- Each of these means and devices may be configured to be connected via a network as a computer as an independent hardware, and a plurality of these devices' devices may function in each computer. It may exist as Information transmission between computers is not limited to the configuration via a network, but may be via a storage medium (USB memory, CD-ROM, floppy disk, etc.).
- Example 1 Example 1
- This example is an example of a method for creating a prediction formula for Charpy absorbed energy, which is a kind of material with a quality characteristic value, for a kind of thick steel plate. Compared with the conventional method, the present invention shows that the prediction accuracy is improved.
- the number of observation data stored in the performance database is 10 3 2
- the output variable is Charpy absorbed energy
- the input variable is 27 other than the constant terms shown in Table 1.
- 'In order to evaluate the prediction accuracy we used the cross-validation method as shown in Fig. 5.
- One data is arbitrarily extracted as evaluation data from the results database, and a prediction formula is created using the other data as model data. Calculate the predicted value by substituting the value of the input variable of the evaluation data into the prediction formula. Since the value of the output variable of the evaluation data, that is, this is the actual value, these differences are prediction errors. The above is performed for all data 10 3 2 and the prediction error is statistically evaluated.
- the constraint parameters derived from the physical properties of the elephant were given to the model parameters for.
- Table 1 shows the constraints imposed on the model parameters for each manufacturing condition.
- the constraints in Table 1 include LOW and UP items, which represent the lower limit and upper limit constraints, respectively.
- the first is that there are no restrictions. For example, when explaining the plate thickness, 0 is entered for LOW and 1 for UP. This indicates that the model parameter value corresponding to the plate thickness has a lower limit of 0 and no upper limit. This is a constraint derived from the physical properties of the object that the Charpy absorbed energy increases as the plate thickness increases.
- the prediction error standard deviation in the extrapolated region is 3% according to the present invention, compared with the case where the prediction formula is obtained by the conventional method without giving the constraint condition of the target physical characteristics. 3% smaller and improved.
- the difference between the prediction error of the invention method and the prediction error of the conventional method was tested, and there was a significant difference at a significance level of 5%. It can be said that there is. ,
- This example is an example of a method for creating a prediction formula for tensile strength, which is a kind of material with a quality characteristic value, for a kind of steel plate. It shows that the prediction accuracy of the present invention is improved compared to the conventional method. ,
- the number of observation data stored in the results database is 2608, the output variable is the tensile strength, and the input variable is 26, excluding the specimen temperature from the manufacturing conditions of Example 1.
- the cross-validation method was used as shown in Fig. 5.
- One data is arbitrarily extracted as evaluation data from the performance database, and data is removed from the other data that has a high degree of similarity from the value (requirement point) of the manufacturing conditions of the evaluation data.
- Create a prediction formula as model data In other words, the extrapolated area is created by removing the data near the request point.
- the predicted value is calculated by substituting the value of the input variable of the evaluation data into the prediction formula. Since the value of the output variable of the evaluation data, that is, this is the actual value, the difference between these becomes the prediction error.
- the above is performed for all data 2 6 0 8 and the prediction error is statistically evaluated.
- the prediction error is evaluated by changing the neighborhood data removal rate when creating model data from 50% to 95%.
- a quality design device is configured using this prediction formula creation means, the designer can obtain a quality prediction value with an accuracy that may be in the extrapolation range. As a result, the number of experiments can be reduced, the development cost can be reduced, and the chance loss associated with the experiment can be reduced, thereby reducing the manufacturing cost.
- the present invention is limited to prediction of Charpy absorbed energy and tensile strength of a thick plate, but the application target of the present invention is not limited to this.
- Examples include yield stress (YP), yield ratio (YR), and ductility (EL).
- the method of the present invention was applied to a model for predicting the tip length at the tip and tail ends of the planar shape after rolling the thick steel plate and the simulation of the tip length control.
- FIG. 8 shows an overview of crop length prediction and crop length control in this embodiment.
- JIT model Just-In-Time model
- the model of crop length is constructed by (Model creation means 20).
- the crop length prediction value is obtained from the obtained model (Kupp length predictor, step 22).
- the plate thickness correction amount that shortens the crop length is determined by a quadratic programming problem with constraints (optimum control amount calculation means 24 and 26).
- optimum control amount calculation means 24 and 26 Apply the obtained thickness correction amount to the actual process (manufacturing process 28).
- V The results obtained are stored in a database, and the model is further modified.
- the prediction formula formula ⁇ 3 ⁇ 4 means of the method of the present invention was applied to the loop length prediction model.
- the model construction means 20 and the mouth length prediction means 22 as the mouth length prediction model will be described in detail.
- the crop length which is the objective variable (output variable, dependent variable) of the local neighborhood regression model, is the crop length L cr O, L at the representative position divided by 16 in the plate width direction, as shown in Fig. 9 (a). crl, L cr 2, L cr 4, and L cr 8 were used.
- Crop length explanatory variables are based on physical knowledge such as forming amount, slab shape, shape after pressing, plate thickness correction amount, etc.
- the crop length prediction formula is a line like Expressed as a format.
- Thickness correction amount dh is as shown in Fig. 9 (b) at the representative position of 16 equally divided in the longitudinal direction of rolling.
- the plate thickness correction amount dh ′ 0 dh 2 dh 8 was used.
- the plate thickness ⁇ is the plate thickness difference with dh 4 as the reference position.
- the plate thickness correction amount influence coefficient ana 12 a 3 with respect to the plate thickness correction amount dh 0 dh 2 dh 8 is limited as shown in the following equation.
- the constraint of the sign of whether the influence coefficient a of the plate thickness correction amount a i, 2 a 13 is positive or negative is used as the constraint. These are given from the physical foresight of the object.
- the model parameter 0 was determined using the attached JIT model (invention method). Note that this embodiment is not limited to the JIT model.
- the accuracy of the model is based on these data.
- the local regression coefficient ⁇ which is a model parameter, is calculated under the constraints.
- the obtained local regression coefficient ⁇ and the required point data, ie, evaluation data Calculate the crop length predicted value from the explanatory variables such as the amount of molding using Equation (4), and (iii) use this crop length predicted value as the objective variable of the crop length actual value in the evaluation data. Compare and evaluate.
- Table 2 shows the standard deviation ⁇ of the predicted value relative to the actual value of the crop length at typical positions in the sheet width direction.
- the plate thickness correction amount which is an operation amount for controlling the control length so as to minimize the clip length
- each crop length L cr O, L crl, L cr 2, L cr 4, L cr 8 is model parameter ⁇ as a factor, forming amount, slab shape, shape after rolling, and plate thickness correction amount It can be expressed as a linear combination of dh (Equation (4)) and the crop length evaluation function ⁇
- the plate thickness correction amounts d h O, d h 2 and d h 8 that minimize (Equation (6)) are determined.
- the plate thickness correction amount constraint is obtained based on physical characteristics and operational constraints such as the roll rotation speed, the diameter of the nozzle, and the maximum rolling speed of the roll. For example, as a restriction on rolling speed
- the optimal plate thickness correction amount that is, the plate thickness correction amount after the operation
- the optimal plate thickness correction amount is calculated by solving the constrained quadratic programming problem with the objective function crop length evaluation function ⁇ .
- the same data as the simulation of the above-mentioned prediction model for 692 items was used.
- the evaluation data arbitrarily extracted from the database 10 and the corresponding evaluation data are provided for each evaluation data by simulation of the cup length prediction model.
- the estimated coefficient of influence a, i, a] 2 , a, 3 and the loop length L cr 0, L crl, L cr 2, L cr 4, L cr 8 are used.
- Restrictions include roll speed, roll diameter, maximum rolling speed
- the maximum manipulated variable is obtained for each evaluation data from the length and the plate thickness correction amount change section length, and is set as a constraint condition. From these data, the constrained optimization problem was solved for each evaluation data, and the plate thickness correction amount after operation was calculated and compared with the actual plate thickness correction amount.
- Fig. 10 shows an example of the case where the operation amount (plate thickness correction amount) is insufficient or too effective when comparing the post-operation plate thickness correction amount and the plate thickness correction amount actual value for the actual data in the extrapolated area.
- the horizontal axis is the rolling direction position when it is divided into 16 equal parts in the longitudinal direction
- the vertical axis is the thickness correction amount when the plate thickness correction amount d h 4 is used as a reference.
- Quality design must be based on past manufacturing results and cost information. At present, designers make decisions by looking at forms, but there is no method for quantitatively evaluating risks (deviating from past manufacturing performance) and costs, and the designed manufacturing conditions are appropriate. I can't evaluate. Therefore, in the present embodiment, as shown in FIG. 12, the quality DB 30 storing the values of each manufacturing condition manufactured in the past and the quality characteristic value (actual value) at that time, and the unit amount of each manufacturing condition Based on the unit price information of each manufacturing condition obtained from the cost DB 3 2 in the cost 40 2
- the above objective function here, manufacturing cost and proximity from past cases
- two or more objective functions in the figure, the amount of deviation from the past results and the cost
- the final decision is made by the designer and contributes to decision support.
- the current quality design of thin steel sheets is based on past similar cases and designs according to the required specifications such as thickness, width, target strength, and toughness, as shown in Fig. 18. Based on know-how, the initial values of design values such as chemical components other than component A, component B, and component C, heating conditions, rolling conditions, and cooling conditions are determined by thickness, width, target strength, and toughness.
- the predicted strength value becomes the target value based on the records of past similar properties (production conditions, average strength results). The manufacturing conditions are changed by trial and error.
- the cost per strength is high for component C, components A and B are about the same, and it is desirable to increase the strength with components A and B, and make up for the shortage with component C.
- component B should be less than a certain allowable value so that the slab can be diverted.
- I would like to do the same as in the past.
- the user selects the manufacturing conditions arbitrarily, and inputs the value 4.10 and the manufacturing condition calculation that calculates the manufacturing conditions other than the selected manufacturing conditions that satisfy the required material property values Means 4 1 2 and influence coefficient calculation means 4 1 4 for calculating the local influence coefficient in the vicinity of the production conditions from the material DB 3 0 when the production condition values are given, and the above production condition calculation means 4 From 1 2 and influence coefficient calculation means 4 14, there is provided support screen creation means 4 1 6 for creating a screen 50 for supporting designer decision making as illustrated in FIG. ..
- the support screen creation means 4 1 6 satisfies the value of the current manufacturing condition, the contour line of the cost, and the required quality characteristic value in the selected manufacturing condition space.
- the contour lines of the manufacturing condition values other than the selected manufacturing conditions, the limit values of the respective manufacturing conditions, and the past actual values of the selected manufacturing conditions are displayed.
- the change direction of the manufacturing condition and the past results are displayed at the same time from the current design value.
- the contour line of the value of the component ⁇ ⁇ ⁇ that makes the intensity level the same and the contour line of the cost ⁇ ⁇ at that time are displayed.
- the objective function the component cost and the risk of deviating from the past manufacturing set value are taken into account, but the number and types of objective functions are not limited to this.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Computing Systems (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Automation & Control Theory (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Control Of Metal Rolling (AREA)
- Feedback Control In General (AREA)
- General Factory Administration (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2006800510296A CN101361085B (zh) | 2006-01-13 | 2006-10-20 | 预测公式生成装置及预测公式生成方法 |
EP06822419A EP1975860A4 (en) | 2006-01-13 | 2006-10-20 | DEVICE AND METHOD FOR CREATING A PREDICTION FORMULA |
CA2636898A CA2636898C (en) | 2006-01-13 | 2006-10-20 | Apparatus and method for constructing prediction model |
US12/087,568 US8374981B2 (en) | 2006-01-13 | 2006-10-20 | Apparatus and method for constructing prediction model |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2006-006127 | 2006-01-13 | ||
JP2006006127A JP2006309709A (ja) | 2005-03-30 | 2006-01-13 | 結果予測装置、制御装置及び品質設計装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2007080688A1 true WO2007080688A1 (ja) | 2007-07-19 |
Family
ID=38256102
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2006/321450 WO2007080688A1 (ja) | 2006-01-13 | 2006-10-20 | 予測式作成装置及び予測式作成方法 |
Country Status (6)
Country | Link |
---|---|
US (1) | US8374981B2 (ja) |
EP (1) | EP1975860A4 (ja) |
KR (1) | KR101011546B1 (ja) |
CN (1) | CN101361085B (ja) |
CA (1) | CA2636898C (ja) |
WO (1) | WO2007080688A1 (ja) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009167481A (ja) * | 2008-01-17 | 2009-07-30 | Jfe Steel Corp | インペラー脱硫制御装置及び方法 |
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 |
JP2020119085A (ja) * | 2019-01-21 | 2020-08-06 | 株式会社日立製作所 | 計算機システム及び対象に関する目的を達成するために有用な情報の提示方法 |
WO2022018912A1 (ja) * | 2020-07-21 | 2022-01-27 | 株式会社日立製作所 | 予測システム、予測方法、ならびに表示装置 |
TWI792035B (zh) * | 2019-09-03 | 2023-02-11 | 財團法人工業技術研究院 | 製作產品之材料推薦系統與材料推薦方法 |
Families Citing this family (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120150576A1 (en) * | 2010-12-08 | 2012-06-14 | Sap Ag | Integrating simulation and forecasting modes in business intelligence analyses |
US9373087B2 (en) * | 2012-10-25 | 2016-06-21 | Microsoft Technology Licensing, Llc | Decision tree training in machine learning |
JP5999265B2 (ja) * | 2013-08-02 | 2016-09-28 | 東芝三菱電機産業システム株式会社 | 省エネルギー操業リコメンドシステム |
ZA201407620B (en) * | 2014-08-26 | 2015-11-25 | Musigma Business Solutions Pvt Ltd | Systems and methods for creating and evaluating experiments |
KR101897022B1 (ko) * | 2014-10-10 | 2018-09-10 | 제이에프이 스틸 가부시키가이샤 | 재료 특성값 추정 방법, 재료 특성값 추정 장치 및 강대의 제조 방법 |
US10073439B1 (en) * | 2014-10-31 | 2018-09-11 | Desprez, Llc | Methods, systems, and software for processing expedited production or supply of designed products |
TWI712899B (zh) * | 2015-07-28 | 2020-12-11 | 香港商阿里巴巴集團服務有限公司 | 資訊查詢方法及裝置 |
US11144842B2 (en) | 2016-01-20 | 2021-10-12 | Robert Bosch Gmbh | Model adaptation and online learning for unstable environments |
DE102017103588A1 (de) * | 2016-02-24 | 2017-08-24 | Jtekt Corporation | Analysevorrichtung und analysesystem |
JP6978224B2 (ja) * | 2017-05-30 | 2021-12-08 | 株式会社Uacj | 材料組織計算装置および制御プログラム |
KR102293631B1 (ko) * | 2018-01-05 | 2021-08-24 | 제이에프이 스틸 가부시키가이샤 | 금속 재료의 부식량 예측 방법, 금속 재료의 선정 방법 및 금속 재료의 부식량 예측 장치 |
CN108628956B (zh) * | 2018-04-12 | 2019-02-15 | 北京亿维讯同创科技有限公司 | 产品创新设计的方法及系统 |
WO2020110953A1 (ja) * | 2018-11-28 | 2020-06-04 | 昭和電工株式会社 | 技術予測装置、方法、およびプログラム |
JP6617842B1 (ja) * | 2019-01-17 | 2019-12-11 | Jfeスチール株式会社 | 金属材料の設計支援方法及び設計支援装置 |
WO2020152750A1 (ja) * | 2019-01-21 | 2020-07-30 | Jfeスチール株式会社 | 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 |
JP2020125595A (ja) * | 2019-02-01 | 2020-08-20 | 株式会社小松製作所 | 建設機械の制御システム、建設機械、及び建設機械の制御方法 |
CA3145393A1 (en) * | 2019-07-24 | 2021-01-28 | Ashley Jensen | Local productivity prediction and management system |
CN111212434B (zh) * | 2019-12-31 | 2023-06-13 | 宁波奥克斯电气股份有限公司 | 一种wifi模组质量预测方法、装置、设备及存储介质 |
CN117687861A (zh) * | 2022-08-29 | 2024-03-12 | 英业达科技有限公司 | 内存计算的仿真器的设定方法 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0695707A (ja) * | 1992-09-11 | 1994-04-08 | Toshiba Corp | モデル予測制御装置 |
JPH0695880A (ja) | 1992-09-14 | 1994-04-08 | Yamatake Honeywell Co Ltd | 事例ベース推論装置 |
JP2001290508A (ja) | 2000-04-04 | 2001-10-19 | Nippon Steel Corp | 結果予測装置、方法、及びコンピュータ読み取り可能な記憶媒体 |
JP2002157572A (ja) | 2000-11-17 | 2002-05-31 | Nippon Steel Corp | 結果予測装置、方法、及びコンピュータ読み取り可能な記憶媒体 |
JP2003305508A (ja) * | 2002-04-08 | 2003-10-28 | Jfe Steel Kk | 圧延要領決定方法 |
JP2004355189A (ja) | 2003-05-28 | 2004-12-16 | Jfe Steel Kk | 結果予測装置 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6157899A (en) * | 1998-10-21 | 2000-12-05 | Bethlehem Steel Corporation | Prediction of responses to different powdering tests on a galvanneal-coated steel substrate or different tests on other substrates, using computer-based systems and methods |
TWI267012B (en) * | 2004-06-03 | 2006-11-21 | Univ Nat Cheng Kung | Quality prognostics system and method for manufacturing processes |
-
2006
- 2006-10-20 EP EP06822419A patent/EP1975860A4/en not_active Ceased
- 2006-10-20 WO PCT/JP2006/321450 patent/WO2007080688A1/ja active Application Filing
- 2006-10-20 CA CA2636898A patent/CA2636898C/en not_active Expired - Fee Related
- 2006-10-20 CN CN2006800510296A patent/CN101361085B/zh not_active Expired - Fee Related
- 2006-10-20 US US12/087,568 patent/US8374981B2/en active Active
- 2006-10-20 KR KR1020087015314A patent/KR101011546B1/ko active IP Right Grant
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0695707A (ja) * | 1992-09-11 | 1994-04-08 | Toshiba Corp | モデル予測制御装置 |
JPH0695880A (ja) | 1992-09-14 | 1994-04-08 | Yamatake Honeywell Co Ltd | 事例ベース推論装置 |
JP2001290508A (ja) | 2000-04-04 | 2001-10-19 | Nippon Steel Corp | 結果予測装置、方法、及びコンピュータ読み取り可能な記憶媒体 |
JP2002157572A (ja) | 2000-11-17 | 2002-05-31 | Nippon Steel Corp | 結果予測装置、方法、及びコンピュータ読み取り可能な記憶媒体 |
JP2003305508A (ja) * | 2002-04-08 | 2003-10-28 | Jfe Steel Kk | 圧延要領決定方法 |
JP2004355189A (ja) | 2003-05-28 | 2004-12-16 | Jfe Steel Kk | 結果予測装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP1975860A4 |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009167481A (ja) * | 2008-01-17 | 2009-07-30 | Jfe Steel Corp | インペラー脱硫制御装置及び方法 |
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 |
JP2020119085A (ja) * | 2019-01-21 | 2020-08-06 | 株式会社日立製作所 | 計算機システム及び対象に関する目的を達成するために有用な情報の提示方法 |
JP7051724B2 (ja) | 2019-01-21 | 2022-04-11 | 株式会社日立製作所 | 計算機システム及び対象に関する目的を達成するために有用な情報の提示方法 |
TWI792035B (zh) * | 2019-09-03 | 2023-02-11 | 財團法人工業技術研究院 | 製作產品之材料推薦系統與材料推薦方法 |
WO2022018912A1 (ja) * | 2020-07-21 | 2022-01-27 | 株式会社日立製作所 | 予測システム、予測方法、ならびに表示装置 |
JP2022021042A (ja) * | 2020-07-21 | 2022-02-02 | 株式会社日立製作所 | 予測システム、予測方法、ならびに表示装置 |
JP7436313B2 (ja) | 2020-07-21 | 2024-02-21 | 株式会社日立製作所 | 予測システム、予測方法、ならびに表示装置 |
Also Published As
Publication number | Publication date |
---|---|
CN101361085B (zh) | 2012-07-04 |
US20090083206A1 (en) | 2009-03-26 |
US8374981B2 (en) | 2013-02-12 |
CA2636898A1 (en) | 2007-07-19 |
EP1975860A1 (en) | 2008-10-01 |
KR20080071607A (ko) | 2008-08-04 |
CN101361085A (zh) | 2009-02-04 |
CA2636898C (en) | 2013-07-30 |
EP1975860A4 (en) | 2011-05-04 |
KR101011546B1 (ko) | 2011-01-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2007080688A1 (ja) | 予測式作成装置及び予測式作成方法 | |
JP2008112288A (ja) | 予測式作成装置、結果予測装置、品質設計装置、予測式作成方法及び製品の製造方法 | |
TW201609283A (zh) | 壓延模擬裝置 | |
Li et al. | Ensemble machine learning systems for the estimation of steel quality control | |
JP2014038595A (ja) | 鋼材の材質予測装置及び材質制御方法 | |
CN112949108B (zh) | 热轧高强钢板形缺陷全流程预测方法及图形用户界面装置 | |
JP2012104042A (ja) | 均一伸びの予測方法および均一伸びの予測プログラム | |
WO2020152993A1 (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 | |
JP2017120640A (ja) | 電磁鋼板の磁気特性予測装置及び磁気特性制御装置 | |
JP5853723B2 (ja) | リン濃度予測装置及び吹錬制御方法 | |
JP5962290B2 (ja) | 鋼材の熱伝達係数予測装置及び冷却制御方法 | |
Kujawińska et al. | Assessment of ductile iron casting process with the use of the DRSA method | |
JP4932294B2 (ja) | 製造仕様決定支援システム、製造仕様決定支援方法、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 | |
Fan et al. | Prediction algorithm for springback of frame-rib parts in rubber forming process by incorporating Sobol within improved grey relation analysis | |
Zhao et al. | Prediction of mechanical properties of cold rolled strip based on improved extreme random tree | |
Tan et al. | Creep lifetime prediction of 9% Cr martensitic heat-resistant steel based on ensemble learning method | |
JP5271136B2 (ja) | 出力値予測方法、該装置および該方法のプログラム | |
JP5195331B2 (ja) | 製造プロセスにおける品質予測装置、予測方法、プログラム及びコンピュータ読み取り可能な記録媒体 | |
JP2006309709A (ja) | 結果予測装置、制御装置及び品質設計装置 | |
Schowtjak et al. | Prediction of ductile damage evolution based on experimental data using artificial neural networks | |
JP5821656B2 (ja) | 生石灰濃度予測装置及び吹錬制御方法 | |
Choudhury | Prediction and analysis of mechanical properties of low carbon steels using machine learning | |
CN116484745A (zh) | 物理冶金指导工业大数据挖掘的热轧低合金钢设计方法 | |
JP5857759B2 (ja) | 変形抵抗予測装置及び圧延機の制御方法 | |
Peng et al. | Knowledge-driven material design platform based on the whole-process simulation and modeling |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 1020087015314 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12087568 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2636898 Country of ref document: CA |
|
WWE | Wipo information: entry into national phase |
Ref document number: 200680051029.6 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2006822419 Country of ref document: EP |