Abnormality determination device and abnormality determination method for manufacturing condition instruction in manufacturing process of steel material

Abstract

To provide an abnormality determination device and an abnormality determination method for a manufacturing condition instruction in a manufacturing process of a steel material, capable of determining the abnormality of the manufacturing condition instruction before the start of the manufacture of the steel material and suppressing the occurrence of a large amount of defective products by using a material prediction value.SOLUTION: An abnormality determination device for a manufacturing condition instruction in a steel material manufacturing process includes: a manufacturing information storage device that accumulates a material component actual value, a manufacturing actual value, and a material quality actual value for each product manufactured in the past in the steel material manufacturing process; a material quality prediction value output means that outputs a material quality prediction value for an instruction value of a manufacturing condition and a raw material component of a product to be newly manufactured using a material quality prediction model created based on each actual value accumulated in the manufacturing information storage device; and a manufacturing condition instruction abnormality handling means that stops actual manufacturing when it is determined that the material quality prediction value obtained by the material quality prediction value output means does not satisfy a required specification.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an apparatus and method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process. [Background technology]

[0002] Steel products are manufactured by determining the manufacturing conditions for each manufacturing process in order to conform to product standards such as the range of mechanical test characteristic values ​​and customer required specifications for the product. As a method for supporting the determination of manufacturing conditions, for example, Patent Documents 1 to 4 disclose a method for determining optimal manufacturing conditions that satisfy required specifications by using a database to predict the material properties that will be obtained when a product is manufactured based on input material components and specified values ​​of manufacturing conditions.

[0003] Furthermore, for example, Patent Document 5 discloses a method for searching for a steel type that matches the required characteristics of an ordered steel material based on the required values ​​of mechanical test properties for the ordered steel material. In the method disclosed in Patent Document 5, if a perfectly matching steel type does not exist, the actual value information of the manufacturing conditions of the steel type with the highest similarity is set as the initial information for designing manufacturing conditions for a new steel type. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-277835 [Patent Document 2] Japanese Patent Application Laid-Open No. 2003-328030 [Patent Document 3] Japanese Patent Application Laid-Open No. 2007-257621 [Patent Document 4] Japanese Patent Application Laid-Open No. 2010-191790 [Patent Document 5] Japanese Patent Application Publication No. 5-287342 Summary of the Invention [Problem to be solved by the invention]

[0005] However, when actually specifying manufacturing conditions, in order to improve yields during steel manufacturing, it is common to specify the same manufacturing conditions that satisfy all required specifications for products with similar steel types (target chemical compositions of steel) or similar ranges of mechanical test characteristics required by customers. Alternatively, in consideration of the customer's usage of steel, conditions may be specified that are limited to a specific customer or the usage of a specific steel.

[0006] In addition, the actual values ​​of steel manufacturing conditions generally have a certain degree of variation or bias relative to the specified values ​​of the manufacturing conditions, which results in errors relative to the specified values ​​of the manufacturing conditions. Factors that cause these errors include fluctuations in manufacturing capabilities determined by the capacity of production equipment, maintenance status, control accuracy, operator skills and experience, etc. To obtain consistent mechanical test value characteristics, the errors relative to the specified values ​​of the manufacturing conditions must be minimized, so the specified values ​​of the manufacturing conditions may be appropriately set according to the manufacturing capabilities at that time.

[0007] In addition, in consideration of the ever-changing manufacturing load of each process in the actual manufacturing, adjustments may be made to the conditions that best balance the manufacturing load and cost within the range that satisfies the required specifications under the given environment. In the past, it was not considered to manage the indicated values ​​of the actual manufacturing conditions using predicted material values. In such cases, the design and input of manufacturing conditions was often done manually, which sometimes resulted in incorrect design or input. Another problem was that the incorrect design or input became apparent after mass production of the product.

[0008] The present invention has been made in consideration of the above, and aims to provide an apparatus and method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process, which can determine an abnormality in manufacturing condition instructions before the start of steel manufacturing by utilizing material prediction values, thereby preventing the mass production of defective products. [Means for solving the problem]

[0009] In order to solve the above-mentioned problems and achieve the object, the device for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the present invention comprises: a manufacturing information storage device that accumulates material component actual values, manufacturing actual values, and material quality actual values ​​for each product manufactured in the past in the steel manufacturing process; a material quality prediction value output means that outputs a material quality prediction value for the material component and manufacturing condition instruction values ​​of a product to be newly manufactured using a material quality prediction model created based on the actual values ​​accumulated in the manufacturing information storage device; and a manufacturing condition instruction abnormal material processing means that stops actual manufacturing when it is determined that the material quality prediction value obtained by the material quality prediction value output means does not satisfy the required specifications.

[0010] The device for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the present invention, in the above-mentioned invention, further comprises a similarity calculation means for calculating the similarity between the instruction values ​​of a product to be newly manufactured and the multiple material component actual values ​​and manufacturing actual values ​​stored in the manufacturing information storage device.

[0011] The device for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the present invention, in the above-mentioned invention, further comprises manufacturing condition instruction decision support means for comparing the material component actual values, manufacturing actual values, and material actual values ​​used in the material prediction value output means for calculating the material prediction value with the material prediction value.

[0012] In order to solve the above-mentioned problems and achieve the object, the method of the present invention for determining an abnormality in manufacturing condition instructions in a steel manufacturing process includes: a storage step of storing, in a manufacturing information storage device, material component actual values, manufacturing actual values, and material quality actual values ​​for each product manufactured in the past in the steel manufacturing process; a material quality prediction value output step of outputting material quality prediction values ​​for the material component and manufacturing condition instruction values ​​of a product to be newly manufactured, using a material quality prediction model created based on the actual values ​​stored in the manufacturing information storage device; and a manufacturing condition instruction abnormality processing step of stopping actual manufacturing when it is determined that the material quality prediction value obtained in the material quality prediction value output step does not satisfy the required specifications.

[0013] The method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the present invention, in the above-mentioned invention, further includes a similarity calculation step of calculating the similarity between the instruction values ​​of a product to be newly manufactured and the multiple material component actual values ​​and manufacturing actual values ​​stored in the manufacturing information storage device.

[0014] The method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the present invention, in the above invention, further includes a manufacturing condition instruction determination support step of comparing the material component actual values, manufacturing actual values, and material quality actual values ​​used in the material quality prediction value output step for calculating the material quality prediction value with the material quality prediction value. [Effects of the Invention]

[0015] According to the device and method for determining abnormalities in manufacturing condition instructions in a steel manufacturing process of the present invention, the predicted material values ​​obtained when the designated values ​​of the designed manufacturing conditions are given are compared with the required specifications (e.g., required values ​​of mechanical test characteristics) before manufacturing begins, and by stopping the production of products given manufacturing conditions that are determined to be abnormal, it is possible to prevent the mass production of defective products whose actual values ​​of mechanical test characteristics do not meet the required specifications. [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 1 is a block diagram showing the configuration of an apparatus for determining an abnormality in a manufacturing condition instruction in a steel manufacturing process according to an embodiment. [Figure 2] FIG. 2 is a diagram showing an example of a database constructed in a manufacturing information storage device in the device for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the embodiment. [Figure 3] FIG. 3 is a flowchart showing the flow of a method for determining an abnormality in a manufacturing condition instruction in a steel manufacturing process according to an embodiment. [Figure 4] FIG. 4 is a diagram showing an example of a material prediction model that predicts a material prediction value from case data in the vicinity of input data. [Figure 5]FIG. 5 is a diagram showing an example of the relationship between the predicted material value and the past specified conditions and actual material value in the manufacturing condition instruction decision support means. DETAILED DESCRIPTION OF THE INVENTION

[0017] An apparatus and method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to an embodiment of the present invention will be described with reference to the drawings. Note that the present invention is not limited to the following embodiments, and the components in the following embodiments include those that are easily replaceable by a person skilled in the art, or those that are substantially identical.

[0018] (Abnormality determination device) An abnormality determination device for specifying manufacturing conditions in a steel manufacturing process according to an embodiment (hereinafter referred to as "abnormality determination device") will be described with reference to Figures 1 and 2. The abnormality determination device 10 is used to manage manufacturing conditions for steel manufactured by, for example, heating, rolling, cooling, heat treating, etc., a cast slab. The abnormality determination device 10 is realized by, for example, a general-purpose computer such as a workstation or a personal computer, or a server located on the cloud.

[0019] 1, the abnormality determination device 10 includes a manufacturing information storage device 20, a material quality prediction value output means 30, a manufacturing condition instruction abnormal material treatment means 40, and a manufacturing condition instruction decision support means 50. The abnormality determination device 10 may be constructed within a single computer, or may be constructed using multiple computers.

[0020] The manufacturing information storage device 20 stores the actual values ​​of material components (component 1 to component X), actual values ​​of manufacturing conditions (manufacturing 1 to manufacturing Y) (manufacturing actual values), and material quality actual values ​​(material quality 1 to material quality Z) for each product manufactured in the past. These actual values ​​may be stored in separate databases, or, for example, as shown in FIG. 2, the actual values ​​may be integrated by product and stored in a database. Also, materials with similar material components or manufacturing conditions may be classified into groups, and a database may be created for each group and stored in the manufacturing information storage device 20. This makes it possible to obtain highly reliable predicted values ​​and reduce the time required for prediction.

[0021] The information to be registered in the database may be directly input by an operator, or may be stored in a storage device such as a hard disk via a removable storage medium such as a CD-ROM or DVD. Also, the information to be registered in the database may be stored in a storage device such as a hard disk via a communication network such as the Internet by wired or wireless electrical communication means.

[0022] The material prediction value output means 30 creates a material prediction model based on each actual value stored in the manufacturing information storage device 20, and uses the created material prediction model to output a material prediction value for the specified values ​​of the material components and manufacturing conditions of a product to be newly manufactured. Specifically, the material prediction value output means 30 includes an input variable limitation means 301, a material prediction value calculation means 302, and a similarity calculation means 303, as shown in FIG.

[0023] The input variable limiting means 301 stores rules for selecting input variables to be used for predicting the material quality of a product from among a large number of factors that affect the material quality of a product (hereinafter referred to as "material quality influencing factors").

[0024] Material quality influencing factors include, for example, the chemical composition of the material (slab) (contained elements, amount, etc.), heating conditions (steel extraction temperature, furnace time, etc.), rolling conditions (steel temperature history, rolling dimensions, reduction, etc.), cooling conditions (steel temperature history, cooling rate, etc.), heat treatment conditions (furnace temperature history, cooling rate, etc.). If material quality prediction is performed using such a large number of material quality influencing factors as input variables, the input space will have too many dimensions, and prediction calculations will take an extremely long time. Therefore, the input variable limitation means 301 selects input variables to be used in material quality prediction, thereby shortening the time required for material quality prediction.

[0025] For example, if material component A has no effect on the material unless it is present at a certain content a or more, a rule that "values ​​less than a will not be used as an input variable" is used. This rule can be created in various ways, for example, by inputting rules that accumulate foresight information about physical phenomena in advance, or by automatically creating rules from accumulated data.

[0026] Furthermore, the input variable limiting means 301 selects and limits the input variables to be used for material prediction, referring to the above rules, based on input information about the product whose material is to be predicted, i.e., the specified values ​​of the manufacturing conditions, and outputs the results to the material prediction value calculating means 302.

[0027] The material quality prediction value calculation means 302 extracts case data having actual values ​​close to the specified values ​​for the product from the data stored in the manufacturing information storage device 20. The number of case data to be extracted is not limited, but it is preferable to extract multiple pieces of data. The material quality prediction value calculation means 302 then calculates a material quality prediction value using the actual material values ​​of the extracted case data, and outputs these results to the similarity calculation means 303, the manufacturing condition instruction abnormal material treatment means 40, and the manufacturing condition instruction decision support means 50.

[0028] The similarity calculation means 303 calculates the similarity to the instruction value of a product to be newly manufactured for the actual values ​​of a plurality of material components and actual values ​​of manufacturing conditions stored in the manufacturing information storage device 20. Specifically, the similarity calculation means 303 evaluates and calculates the similarity between the predicted material value and the actual material value of an acquired case that is close to the input value (the actual material value corresponding to the input value), and outputs the result to the manufacturing condition instruction abnormality determination means 403.

[0029] The manufacturing condition instruction abnormal material processing means 40 stops actual manufacturing when it is determined that the material prediction value obtained by the material prediction value output means 30 does not satisfy the required specifications. Specifically, as shown in Figure 1, the manufacturing condition instruction abnormal material processing means 40 includes a material prediction result determination means 401, an instruction abnormality determination threshold input means 402, a manufacturing condition instruction abnormality determination means 403, and an abnormality processing means 404.

[0030] The material prediction result determination means 401 compares the material prediction value output from the material prediction value calculation means 302 with the required specifications (required specification values) for the material included in the input information for the product whose material is to be predicted. If the material prediction value is within the range of the required specification values, the material prediction result determination means 401 determines that the manufacturing condition instructions are normal, and outputs the determination result to the manufacturing condition instruction abnormality determination means 403. On the other hand, if the material prediction value is outside the range of the required specification values, the material prediction result determination means 401 determines that the manufacturing condition instructions are abnormal, and outputs the determination result to the manufacturing condition instruction abnormality determination means 403.

[0031] The instruction abnormality determination threshold input means 402 stores rules for determining whether a manufacturing condition instruction is abnormal. The rules for determining whether a manufacturing condition instruction is abnormal include, for example, a threshold for determining whether a manufacturing condition instruction is abnormal within a stricter range than the actual required specification value.

[0032] The manufacturing condition instruction abnormality determination means 403 determines the validity of the manufacturing condition instructions by referring to the rules of the instruction abnormality determination threshold input means 402. Then, the manufacturing condition instruction abnormality determination means 403 outputs the result of the determination that the manufacturing condition instructions are abnormal, out of the determination result and the result of the material prediction result determination means 401, to the abnormality processing means 404. Note that while the material prediction result determination means 401 determines whether the material prediction value is within the range of the required specification values, the manufacturing condition instruction abnormality determination means 403 further strictly determines the validity of the manufacturing condition instructions by, for example, narrowing the upper and lower limits of the required specification values.

[0033] The abnormality handling means 404 outputs to the manufacturing condition instruction determination support means 50 an instruction to stop manufacturing of the product having the input information in which the manufacturing condition instruction is determined to be abnormal, that is, the manufacturing condition instruction value.

[0034] The manufacturing condition instruction decision support means 50 compares the material predicted value with the material component actual value, manufacturing actual value, and material actual value used in the material predicted value output means 30 to calculate the material predicted value. Specifically, the manufacturing condition instruction decision support means 50 supports the operator in finding normal manufacturing condition instructions that satisfy the required specification values ​​by having the operator confirm the deviation between the manufacturing condition instructions of the product determined to have an abnormal manufacturing condition instruction and past manufacturing condition instructions.

[0035] (Abnormality determination method) The procedure for issuing manufacturing condition instructions that enable the production of products that satisfy required specifications using an abnormality determination method for manufacturing condition instructions in a steel manufacturing process according to an embodiment (hereinafter referred to as the "abnormality determination method") will be described with reference to Figures 3 to 5.

[0036] First, for each product manufactured in the past, the material component actual values, manufacturing actual values, and material quality actual values ​​are stored in the manufacturing information storage device 20. This corresponds to the storing step. In the storing step, not only is information about products manufactured in the past stored, but the material component actual values, manufacturing actual values, and material quality actual values ​​for each newly manufactured product may also be stored in the manufacturing information storage device 20 each time manufacturing is carried out.

[0037] First, initial input values ​​of the instruction values ​​of the manufacturing conditions for a product having certain required specifications are input to the abnormality determination device 10 (step S1). The input method in step S1 is not particularly limited, and the input may be made from another computer or by an operator.

[0038] Next, the process proceeds to a material prediction value output step. In the material prediction value output step, the material prediction value output means 30 calculates a material prediction value corresponding to the initially input instruction value of the manufacturing condition for the product, using a material prediction model, based on the initially input instruction value of the manufacturing condition for the product. In step S2, the material prediction value output means 30 refers to the rules stored in the input variable limitation means 301, and selects input variables that have a large influence on the material, based on the instruction value for the product input in step S1 (step S2).

[0039] The above-mentioned "indication values ​​related to the product" include, for example, material components and their contents, the steel material extraction temperature and residence time in a heating furnace, the rolling temperature and reduction rate in hot rolling, etc. Then, the material quality prediction value output means 30 outputs the selected input variables and the input values ​​(indication values) corresponding to these input variables to the material quality prediction value calculation means 302.

[0040] Next, the material prediction value calculation means 302 calculates the distance between the actual value of each case and the designated value of the manufacturing condition for the input product (step S3). In step S3, the distance is calculated using the input variables selected in step S2, the designated values ​​of the manufacturing conditions, and the actual values ​​of the material components and the actual values ​​of the manufacturing conditions corresponding to the selected input variables from the data stored in the manufacturing information storage device 20.

[0041] The distance calculation in step S3 uses a distance function that defines the distance between the indicated values ​​of the material components and the indicated values ​​of the manufacturing conditions and the actual values ​​of the material components and the manufacturing conditions stored in the material prediction value calculation means 302. There are various methods for this, but for example, the method disclosed in Japanese Patent No. 4365600 can be used. In step S3, distances are calculated as many times as the number of stored cases.

[0042] 4, for example, to obtain data on cases in which the actual values ​​of the manufacturing conditions are close to the output value (predicted material value) for the input product from the data stored in the manufacturing information storage device 20. There are various methods for this, but for example, among the data stored in the manufacturing information storage device 20, the N cases (N is any natural number) with the smallest distance calculated in step S3 can be defined as cases in the vicinity of the output value (predicted value).

[0043] Next, the material quality prediction value calculation means 302 calculates a material quality prediction value for the specified value for the input product using actual material quality values ​​related to the material among the acquired case data that are near the specified value for the input product (step S5). There are various methods for this, but for example, the method disclosed in Japanese Patent No. 4365600 can be used. Furthermore, as the output value (predicted material quality value), output variables that represent the material, such as tensile strength, yield point, elongation, Charpy absorbed energy, etc., can be used. Furthermore, the material quality prediction value calculation means 302 outputs the calculated predicted material quality value to the material quality prediction result determination means 401.

[0044] Next, the similarity calculation means 303 calculates the similarity between the predicted material value calculated in step S5 and a case in which the actual material value is in the vicinity, using, for example, a method described in Japanese Patent Laid-Open No. 6-95880 (step S6). This process corresponds to the similarity calculation step. The similarity calculation means 303 outputs the calculated similarity to the manufacturing condition instruction abnormality determination means 403.

[0045] Next, the material prediction result determination means 401 determines whether the material prediction value calculated by the material prediction value calculation means 302 is within the allowable range of the required specifications, i.e., whether it satisfies the following formula (1) (step S7), and outputs the result to the manufacturing condition instruction abnormality determination means 403.

[0046] Lower limit of required specifications ≦ Estimated material value ≦ Upper limit of required specifications (1)

[0047] Next, the manufacturing condition instruction abnormality determination means 403 determines whether the manufacturing condition instruction is normal or abnormal (step S8) by referring to the rules stored in the instruction abnormality determination threshold input means 402. Methods for determining whether the manufacturing condition instruction is normal or abnormal include a method that takes into account a prediction error as in the following formula (2), and a method that uses the similarity calculated by the similarity calculation means 303 as in the following formula (3).

[0048] Lower limit of required specifications + threshold ≦ predicted material value ≦ upper limit of required specifications - threshold (2) Lower limit of similarity≦similarity≦upper limit of similarity (3)

[0049] The "threshold value" in the above formula (2) and the "lower limit value of similarity" and "upper limit value of similarity" in the above formula (3) are input from the instruction abnormality determination threshold input means 402. Furthermore, the determination in step S8 may be performed using a single determination means for material properties such as tensile strength, yield point, elongation, and Charpy absorbed energy, or different determination means may be used for each material property. The determination result in step S7 is output to the manufacturing condition instruction abnormality determination means 403, but the determination result in step S8 is output to the manufacturing condition instruction decision support means 50 if the manufacturing condition instructions are determined to be normal, or to the abnormality treatment means 404 if the manufacturing condition instructions are determined to be abnormal.

[0050] Next, if the predicted material value is within the range determined using the threshold input from the instruction abnormality determination threshold input means 402, the manufacturing condition instruction abnormality determination means 403 determines that the manufacturing condition instruction is normal (Yes in step S9) and proceeds to step S10. On the other hand, if the predicted material value is outside the range determined using the threshold input from the instruction abnormality determination threshold input means 402, the manufacturing condition instruction abnormality determination means 403 determines that the manufacturing condition instruction is abnormal (No in step S9) and proceeds to step S11.

[0051] In step S10, an instruction is issued to manufacture the product according to the input manufacturing condition instructions (step S10). Meanwhile, in step S11, the abnormality handling means 404 outputs an instruction to stop manufacturing the product having input information (manufacturing condition instruction values) that is determined to be abnormal in the manufacturing condition instructions (step S11). This process corresponds to the manufacturing condition instruction abnormality material handling step.

[0052] Next, the manufacturing condition instruction determination support means 50 compares the predicted material value obtained from the manufacturing instruction determined to be abnormal with the past instruction conditions and actual material value (step S12). The comparison result of step S12 can be, for example, the relationship between the composition or manufacturing parameters and the material as shown in FIG.

[0053] Next, it is determined whether the manufacturing condition instruction determined to be abnormal in step S9 was actually correct (step S13) based on the comparison result shown in Fig. 5. "It is actually correct manufacturing condition instruction" means, for example, that the manufacturing conditions designed to meet new required specifications were actually correct but were derived with a low degree of similarity because they were not similar to manufacturing conditions used in past manufacturing.

[0054] In step S13, if it is determined that the manufacturing condition instructions are incorrect, i.e., that the manufacturing condition instructions need to be revised (No in step S13), the process returns to step S1, and new instruction values ​​for the manufacturing condition instructions are set. On the other hand, in step S13, if it is determined that the manufacturing condition instructions are correct, i.e., that the manufacturing condition instructions do not need to be revised (Yes in step S13), the process proceeds to step S10, and an instruction is issued to carry out manufacturing according to the input manufacturing condition instructions. This series of processes from step S12 to step S13 corresponds to the manufacturing condition instruction determination support step.

[0055] According to the device and method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the embodiment described above, the predicted material value obtained when the designated values ​​of the designed manufacturing conditions are given is compared with the required specifications (e.g., required values ​​of mechanical test characteristics) before manufacturing begins, and by stopping the production of products given manufacturing conditions that are determined to be abnormal, it is possible to prevent the mass production of defective products whose actual values ​​of mechanical test characteristics do not meet the required specifications.

[0056] Furthermore, according to the device and method for determining an abnormality in a manufacturing condition instruction in a steel manufacturing process according to the embodiment, it is possible to suppress overdetection and overreaction in abnormality determination by evaluating the similarity between past actual values ​​of the manufacturing conditions and the manufacturing condition instruction values. Furthermore, according to the device and method for determining an abnormality in a manufacturing condition instruction in a steel manufacturing process according to the embodiment, when an incorrect manufacturing condition instruction value is given, it is possible to derive a correct manufacturing condition instruction value in a short period of time by comparing the material composition actual values, manufacturing actual values, and material quality actual values ​​used to calculate the material quality predicted value with the material quality predicted value.

[0057] Furthermore, according to the embodiment, the device and method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process determine whether the manufacturing condition instruction values ​​satisfy the required specifications even when errors in the manufacturing actual values ​​and errors in the material model are taken into account relative to the predicted material value, and if the required specifications are not satisfied, it is possible to stop the manufacturing of abnormal material according to the manufacturing condition instruction. Therefore, it is no longer possible to erroneously specify manufacturing conditions that have no margin for the upper and lower limits of the required specification range or that do not satisfy the required specification range, and it is possible to reduce the frequency of defects occurring due to the material actual values ​​deviating from the required specifications.

[0058] Furthermore, the device and method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to the embodiment can provide support when new manufacturing condition instructions need to be considered by comparing predicted material properties values ​​obtained for input values ​​with past actual material properties values. Therefore, regardless of the skill level of the person examining the manufacturing conditions, the manufacturing condition instructions can be reset to the correct ones in a short time.

[0059] Although the present invention has been described above as an embodiment, the present invention is not limited to the description and drawings that form part of the disclosure of the present invention. In other words, other embodiments, examples, and operational techniques that can be made by those skilled in the art based on the present invention are all included in the scope of the present invention. [Explanation of symbols]

[0060] 10 Abnormality determination device 20 Manufacturing information storage device 30 Material prediction value output means 301 Input variable limitation means 302 Material prediction value calculation method 303 Similarity calculation means 40 Manufacturing condition instructions and measures for dealing with abnormal materials 401 Material prediction result judgment means 402 Instruction abnormality determination threshold input means 403 Manufacturing condition instruction abnormality determination means 404 Abnormality Handling Procedures 50 Manufacturing condition instruction determination support means

Claims

1. a manufacturing information storage device that stores material composition actual values, manufacturing actual values, and material quality actual values ​​for each product manufactured in the past in a steel manufacturing process; a material prediction value output means for outputting a material prediction value for the specified values ​​of material components and manufacturing conditions of a product to be newly manufactured, using a material prediction model created based on each actual value stored in the manufacturing information storage device; a manufacturing condition instruction abnormal material processing means for stopping actual manufacturing when it is determined that the material prediction value obtained by the material prediction value output means does not satisfy the required specifications; An abnormality determination device for manufacturing condition instructions in a steel manufacturing process, comprising:

2. 2. The device for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to claim 1, further comprising a similarity calculation means for calculating a similarity between a plurality of material component actual values ​​and manufacturing actual values ​​stored in the manufacturing information storage device and the instruction values ​​of a product to be newly manufactured.

3. 3. The apparatus for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to claim 1, further comprising a manufacturing condition instruction decision support means for comparing the material component actual value, manufacturing actual value, and material actual value used in the material prediction value output means for calculating the material prediction value with the material prediction value.

4. a storage step of storing, in a manufacturing information storage device, actual material composition values, actual manufacturing values, and actual material quality values ​​for each product manufactured in the past in a steel manufacturing process; a material prediction value output step for outputting a material prediction value for the specified values ​​of material components and manufacturing conditions of a product to be newly manufactured, using a material prediction model created based on each actual value stored in the manufacturing information storage device; a manufacturing condition instruction abnormal material processing step for stopping actual manufacturing when it is determined that the material prediction value obtained in the material prediction value output step does not satisfy the required specifications; A method for determining abnormalities in manufacturing condition instructions in a steel manufacturing process, including:

5. 5. The method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to claim 4, further comprising a similarity calculation step of calculating a similarity between a plurality of material component actual values ​​and manufacturing actual values ​​stored in the manufacturing information storage device and the instruction values ​​of a product to be newly manufactured.

6. 6. The method for determining an abnormality in manufacturing condition instructions in a steel manufacturing process according to claim 4 or claim 5, further comprising a manufacturing condition instruction decision support step of comparing the material component actual values, manufacturing actual values, and material actual values ​​used in the material prediction value output step for calculating the material prediction value with the material prediction value.

Images