Information processing device, learning model creation device, information processing method, learning model creation method and program

Abstract

Estimates are made taking into account the effects between batches. [Solution] The information processing device includes: a production condition acquisition unit that acquires nth production conditions, which are production conditions for the nth polymerization process, which is the polymerization process for the nth batch, from production conditions including indicators of raw materials for the fluorine-containing elastomer in the polymerization process and indicators of polymerization conditions for the fluorine-containing elastomer; a target value acquisition unit that acquires a target value of a production result indicator confirmed in a confirmation process for an mth batch that follows the nth batch; a production information providing unit that provides the target value of the production result indicator and the nth production condition for correspondence information that shows the correspondence between the production conditions and the production result indicator of the fluorine-containing elastomer for each batch; and a result acquisition unit that acquires the mth production condition output from the correspondence information to which the nth production condition and the target value of the production result indicator are given, as an estimated result of the production conditions for the polymerization process of the mth batch.

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, a learning model creation device, an information processing method, a learning model creation method, and a program. [Background technology]

[0002] Conventionally, there are manufacturing processes for manufacturing resins and the like through multiple steps using a so-called batch production method. It would be desirable to be able to estimate the manufacturing results of such manufacturing processes under certain manufacturing conditions. For example, Patent Document 1 discloses a technology for estimating the manufacturing results. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-080701 Summary of the Invention [Problem to be solved by the invention]

[0004] In batch production, the manufacturing conditions in the manufacturing process of the previous batch may affect the manufacturing results of the next batch. In other words, in batch production, changes in the manufacturing environment between previous and next batches may affect the manufacturing results. However, in conventional technology, the influence between batches has not been fully considered.

[0005] The present disclosure has been made in consideration of these circumstances, and aims to provide an information processing device, a learning model creation device, an information processing method, a learning model creation method, and a program that can perform estimation taking into account the influence between batches. [Means for solving the problem]

[0006] One aspect of the present disclosure is an information processing device that presents production conditions for a production process for producing a fluorine-containing elastomer for each batch, the production conditions including an index of raw materials for the fluorine-containing elastomer in the polymerization process and an index of polymerization conditions for the fluorine-containing elastomer, the information processing device including a production condition acquisition unit that acquires n-th production conditions that are production conditions for an n-th polymerization process that is the polymerization process for the n-th batch (n is a natural number), and an m-th batch (m is a natural number greater than n) that is a subsequent batch after the n-th batch. The information processing device includes: a target value acquisition unit that acquires a target value of a production result index confirmed in a process; a production information providing unit that provides the target value of the production result index acquired by the target value acquisition unit and the nth production condition acquired by the production condition acquisition unit, with respect to correspondence information indicating the correspondence between the production conditions and the production result index of a fluorinated elastomer for each batch; and a result acquisition unit that acquires the mth production condition output from the correspondence information to which the nth production condition and the target value of the production result index are assigned, as an estimated result of the production conditions for the polymerization process of the mth batch.

[0007] One aspect of the present disclosure is an information processing device that presents production conditions for a production process for producing a fluorine-containing elastomer for each batch, the production conditions including an index of raw materials for the fluorine-containing elastomer in the polymerization process, an index of polymerization conditions for the fluorine-containing elastomer, and a maintenance index showing the maintenance status of production equipment, the information processing device comprising: a production condition acquisition unit that acquires an n-th production condition that is the production condition for an n-th polymerization process that is the polymerization process for the n-th batch (n is a natural number); and a confirmation step that indicates a correspondence between the production conditions and the production result index confirmed in the confirmation step. The information processing device includes: a first manufacturing information providing unit that provides n manufacturing conditions as manufacturing information; a first result acquiring unit that acquires the manufacturing result index output from the first correspondence information to which the n manufacturing conditions are given as the nth manufacturing result index, which is the manufacturing result index of the nth batch; a second manufacturing information providing unit that provides the nth manufacturing result index acquired by the first result acquiring unit and the nth manufacturing conditions acquired by the manufacturing condition acquiring unit with respect to second correspondence information that indicates the correspondence between manufacturing conditions and manufacturing result indexes for each batch; and a second result acquiring unit that acquires the mth manufacturing conditions output from the second correspondence information to which the nth manufacturing conditions and the nth manufacturing result index are given as estimated results of manufacturing conditions for the polymerization process of the mth batch.

[0008] One aspect of the present disclosure includes a polymerization step and a confirmation step of a production result index of a fluorine-containing elastomer produced through the polymerization step, and based on a first learning dataset in which, for a production step in which a fluorine-containing elastomer is produced batch by batch, information indicating production conditions for each batch is included as learning data and the production result index confirmed in the confirmation step of a batch subsequent to the first batch is included as training data, a correlation between the production conditions for each batch and the production result index confirmed in the confirmation step of a batch subsequent to the first batch is used as an explanatory variable and the production result index confirmed in the confirmation step of a batch subsequent to the first batch is used as a target variable. a processing unit that creates a first learning model by machine learning a relationship between the manufacturing result indicators confirmed in the confirmation process of a batch subsequent to the first batch and the information indicating the manufacturing conditions of the subsequent batch, based on a second learning dataset that includes the manufacturing result indicators confirmed in the confirmation process of a batch subsequent to the first batch as learning data and information indicating the manufacturing conditions of the subsequent batch as training data, using the manufacturing result indicators confirmed in the confirmation process of a batch subsequent to the first batch as explanatory variables and the manufacturing conditions of the subsequent batch as objective variables, to create a second learning model by machine learning a relationship between the manufacturing result indicators confirmed in the confirmation process of a batch subsequent to the first batch and the information indicating the manufacturing conditions of the subsequent batch.

[0009] One aspect of the present disclosure is a learning model creation device comprising: a polymerization step; and a confirmation step of a production result index indicating the production result of a fluorine-containing elastomer produced through the polymerization step, the learning model creation device comprising: a first learning dataset containing, as learning data, production conditions for each batch, including indicators of raw materials for the fluorine-containing elastomer in the polymerization step, indicators of polymerization conditions for the fluorine-containing elastomer, and a maintenance index indicating the maintenance status of the production equipment, and the production result index confirmed in the confirmation step for the batch, as training data, to create a first learning model by machine learning the relationship between information indicating the production conditions and information indicating the production result index, with the production conditions as explanatory variables and the production result index as a response variable; and a second learning dataset containing, as learning data, the production conditions and the production result index for each batch, and the production conditions for a batch subsequent to the first batch, to create a second learning model by machine learning the relationship between the production conditions and information indicating the production result index and the production conditions, with the production conditions and the production result index as explanatory variables and the production conditions for the subsequent batch as a response variable.

[0010] One aspect of the present disclosure is an information processing method for presenting production conditions for a production process for producing a fluorine-containing elastomer for each batch, the method including a polymerization step and a confirmation step of a fluorine-containing elastomer produced through the polymerization step, the information processing method including: acquiring an n-th production condition, which is the production condition for the n-th polymerization step that is the polymerization step for the n-th batch (n is a natural number), from production conditions including indicators of raw materials of the fluorine-containing elastomer in the polymerization step and indicators of polymerization conditions for the fluorine-containing elastomer; acquiring a target value of a production result indicator confirmed in the confirmation step for an m-th batch (m is a natural number greater than n) that comes after the n-th batch; providing the acquired target value of the production result indicator and the acquired n-th production condition to correspondence information that shows the correspondence relationship between the production conditions and the production result indicator of the fluorine-containing elastomer for each batch; and acquiring the m-th production condition, output from the correspondence information to which the n-th production condition and the target value of the production result indicator are provided, as an estimated result of the production condition for the polymerization step for the m-th batch.

[0011] One aspect of the present disclosure is an information processing method for presenting production conditions for a production process for producing a fluorine-containing elastomer for each batch, the method comprising: a polymerization step; and a confirmation step of a production result index showing the production result of a fluorine-containing elastomer produced through the polymerization step, the method comprising: acquiring an n-th production condition which is the production condition for an n-th polymerization step which is the polymerization step for the n-th batch (n is a natural number) from among production conditions including an index of raw materials for the fluorine-containing elastomer in the polymerization step, an index of polymerization conditions for the fluorine-containing elastomer, and a maintenance index showing the maintenance status of production equipment; and obtaining first correspondence information showing the correspondence between the production conditions and the production result index confirmed in the confirmation step. providing the acquired nth production condition as production information to first correspondence information; acquiring the production result index output from the first correspondence information to which the nth production condition is provided as the nth production result index, which is the production result index of the nth batch; providing the acquired nth production result index and the acquired nth production condition to second correspondence information indicating the correspondence between the production conditions and the production result index for each batch; and acquiring the mth production condition output from the second correspondence information to which the nth production condition and the nth production result index are provided as an estimated result of the production condition of the polymerization step of the mth batch.

[0012] One aspect of the present disclosure includes a polymerization step and a confirmation step of a production result index of a fluorine-containing elastomer produced through the polymerization step, and based on a first learning dataset containing, as training data, information indicating the production conditions for each batch in the production step of producing a fluorine-containing elastomer for each batch and the production result index confirmed in the confirmation step of a batch subsequent to the first batch as training data, calculates the relationship between the production conditions for each batch and the production result index confirmed in the confirmation step of a batch subsequent to the first batch, using the information indicating the production conditions for each batch as an explanatory variable and the production result index confirmed in the confirmation step of a batch subsequent to the first batch as a response variable. a second learning model created by machine learning a relationship between the manufacturing result indicators confirmed in the confirmation process of a batch subsequent to the first batch and information indicating the manufacturing conditions of the subsequent batch, using the manufacturing result indicators confirmed in the confirmation process of a batch subsequent to the first batch as explanatory variables and the manufacturing conditions of the subsequent batch as objective variables, based on a second learning dataset that includes the manufacturing result indicators confirmed in the confirmation process of a batch subsequent to the first batch as learning data and information indicating the manufacturing conditions of the subsequent batch as training data;

[0013] One aspect of the present disclosure is a learning model creation method executed by a learning model creation device, comprising: a polymerization step; and a confirmation step of a production result index indicating the production result of a fluorinated elastomer produced through the polymerization step, wherein, for a production step of producing a fluorinated elastomer batch by batch, the method comprises: creating a first learning model by machine learning a relationship between information indicating the production conditions and information indicating the production result index, with the production conditions being used as explanatory variables and the production result index being used as a response variable, based on a first learning dataset including, as learning data, production conditions for each batch, including indicators of raw materials for the fluorinated elastomer in the polymerization step, indicators of polymerization conditions for the fluorinated elastomer, and a maintenance index indicating the maintenance status of the production equipment, and including the production result index confirmed in the confirmation step for the batch as training data; and creating a second learning model by machine learning a relationship between information indicating the production conditions and the production result index, with the production conditions being used as explanatory variables and the production result index being used as a response variable, based on a learning dataset including, as learning data, the production conditions and the production result index for each batch and the production conditions for a batch subsequent to the first batch, with the production conditions and the production result index being used as explanatory variables and the production conditions for the subsequent batch as the response variable.

[0014] One aspect of the present disclosure is a program that causes a computer to function as the information processing device described above. One aspect of the present disclosure is a program that causes a computer to function as the aforementioned learning model creation device. [Effects of the Invention]

[0015] According to the information processing device, learning model creation device, information processing method, learning model creation method, and program disclosed herein, it is possible to obtain estimation results that take into account the influence between batches in a batch production system. [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 1 illustrates an example of the configuration of an information processing system according to a first embodiment. [Figure 2] 3A to 3C are diagrams illustrating an example of a manufacturing process according to the present embodiment. [Figure 3] FIG. 2 is a diagram illustrating an example of a functional configuration of the information processing apparatus according to the present embodiment. [Figure 4] FIG. 10 is a diagram illustrating an example of correspondence information according to the present embodiment. [Figure 5] FIG. 10 is a diagram illustrating an example of a search result for correspondence information according to the present embodiment. [Figure 6] FIG. 2 is a diagram illustrating an example of the flow of operations of the information processing device of the present embodiment. [Figure 7] FIG. 1 is a diagram illustrating an example of a learning model creation device according to an embodiment of the present invention. [Figure 8] 10 is a flowchart illustrating an example of the operation of the learning model creation device of the present embodiment. [Figure 9] FIG. 10 is a diagram illustrating an example of a functional configuration of an information processing device according to a second embodiment. [Figure 10] FIG. 10 is a diagram showing an example of a search result of the first correspondence information according to the present embodiment. [Figure 11] FIG. 2 is a diagram illustrating an example of the flow of operations of the information processing device of the present embodiment. [Figure 12] FIG. 1 is a diagram illustrating an example of a learning model creation device according to an embodiment of the present invention. [Figure 13] 10 is a flowchart illustrating an example of the operation of the learning model creation device of the present embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0017] [First embodiment] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. FIG. 1 is a diagram showing an example of the configuration of an information processing system 1 according to this embodiment. The information processing system 1 includes an input device 2, an information processing device 3, a data server 4, and a display device 5. The input device 2 has a function of inputting information from outside. For example, the input device 2 has an operation unit (not shown) and has a function of inputting information in response to operations performed by a user. Note that the input device 2 may also receive information input from another external device (e.g., a manufacturing management device) without the intervention of a user. The information processing device 3 is configured by, for example, a computer.

[0018] The data server 4 stores various types of information. In this embodiment, the data server 4 stores trained models, programs corresponding to trained models, tables recording past manufacturing results, and the like.

[0019] The table in which the trained model and past manufacturing results are recorded may be stored in advance in the data server 4, or may be updated externally at any time. Here, any model may be used as the trained model, and for example, a neural network model may be used. The display device 5 has a function of outputting information to the outside. For example, the display device 5 includes a display unit (not shown) and has a function of displaying and outputting information to be displayed on a screen.

[0020] [Manufacturing process] FIG. 2 is a diagram showing an example of a manufacturing process of this embodiment. The manufacturing process of this embodiment is to manufacture products (or semi-finished products; the same applies in the following explanation) in so-called batches (or lots; the same applies in the following explanation). In the following explanation, a manufacturing method for manufacturing products in batches is also simply referred to as a batch production method. In other words, the manufacturing process of this embodiment employs a batch production method. In this embodiment, a fluorine-containing elastomer is taken as an example of a product (or semi-finished product) produced in the manufacturing process, but the present invention is not limited to this. In the present disclosure, a "fluoroelastomer" is a fluoropolymer having a glass transition temperature of 20°C or lower and a melting peak (ΔH) magnitude of 4.5 [J / g] or lower, and further having at least one fluorine atom in the monomer constituting the fluoropolymer.

[0021] Examples of fluorine-containing elastomers include FFKM, FEPM, and FKM. FFKM is a copolymer containing tetrafluoroethylene (hereinafter also referred to as "TFE") units and perfluoromethylvinyl ether (hereinafter also referred to as "PMVE") units as essential units, and the total proportion of TFE units and PMVE units to the total units of the fluorine-containing elastomer is 80 mol % or more. FEPM is a copolymer containing TFE units and propylene units (hereinafter also referred to as "P") as essential units, with the total proportion of TFE units and propylene units being 50 mol% or more. FKM is a copolymer containing vinylidene fluoride (hereinafter also referred to as "VDF") and hexafluoropropylene (hereinafter also referred to as "HFP") as essential units. Among these, FFKM, FEPM, and FKM are preferred as fluorine-containing elastomers, FFKM and FEPM are more preferred, and FFKM is even more preferred.

[0022] The perfluoro(alkyl vinyl ether) (PAVE) from which the PAVE unit is derived is preferably a monomer represented by formula (1) because it has excellent polymerization reactivity when producing the specific polymer described below and can produce the present elastomer more efficiently. CF2=CF-OR f1 (1) In formula (1), R f1 R represents a perfluoroalkyl group having 1 to 10 carbon atoms. f1 In view of better polymerization reactivity, the number of carbon atoms in the perfluoroalkyl group is preferably 1 or more and 8 or less, more preferably 1 or more and 6 or less, even more preferably 1 or more and 5 or less, and particularly preferably 1 or more and 3 or less. The perfluoroalkyl group may be linear or branched.

[0023] Specific examples of PAVE include perfluoro(methyl vinyl ether) (hereinafter also referred to as "PMVE"), perfluoro(ethyl vinyl ether) (hereinafter also referred to as "PEVE"), and perfluoro(propyl vinyl ether) (hereinafter also referred to as "PPVE"), and PMVE or PPVE are preferred, with PMVE being more preferred, as this elastomer can be produced more efficiently.

[0024] The present elastomer may contain units based on monomers other than TFE units and PAVE units (hereinafter also referred to as "other monomers"). Specific examples of the other monomers include a monomer having two or more polymerizable unsaturated bonds (hereinafter also referred to as "BO"), a monomer having one or more atoms of at least one kind selected from the group consisting of a chlorine atom, a bromine atom, and an iodine atom (hereinafter also referred to as "R Hal "), a monomer having a nitrile group (hereinafter referred to as "R CN "), and the compound represented by formula (6) described below (hereinafter also referred to as "POAVE").

[0025] BO is a monomer having two or more polymerizable unsaturated bonds. Examples of the polymerizable unsaturated bond include a carbon atom-carbon atom double bond (C=C) and a carbon atom-carbon atom triple bond (C≡C). The number of polymerizable unsaturated bonds that BO has is preferably 2 or more and 6 or less, more preferably 2 or 3, and even more preferably 2, in terms of better polymerization reactivity. BO preferably contains a fluorine atom, since this reduces the compression set of the crosslinked rubber article at high temperatures.

[0026] BO is preferably a monomer represented by formula (2) in that the crosslinked rubber article has better releasability. (CR 21 R 22 =CR 23 -) a1 R 24 (2) In equation (2), R21 , R 22 , and R 23 each independently represents a hydrogen atom, a fluorine atom, a methyl group, or a trifluoromethyl group; a1 represents an integer of 2 or more and 6 or less; R 24 represents an a1-valent perfluorohydrocarbon group having 1 to 10 carbon atoms, or a group having an etheric oxygen atom at the end or between the carbon-carbon bonds of an a1-valent perfluorohydrocarbon group having 1 to 10 carbon atoms. 21 , multiple R 22 and multiple R 23 may be the same or different, and are particularly preferably the same. a1 is preferably 2 or 3, and is particularly preferably 2.

[0027] Because of the superior polymerization reactivity of BO, 21 , R 22 , and R 23 is preferably a fluorine atom or a hydrogen atom, and R 21 , R 22 , and R 23 It is more preferable that all of R are fluorine atoms or all are hydrogen atoms, and in view of better mold releasability of the crosslinked rubber article, 21 , R 22 , and R 23 It is particularly preferred that all of are fluorine atoms. R 24 R may be linear, branched, or cyclic, preferably linear or branched, and particularly preferably linear. 24 The number of carbon atoms in is preferably 2 or more and 8 or less, more preferably 3 or more and 7 or less, even more preferably 3 or more and 6 or less, and particularly preferably 3 or more and 5 or less.

[0028] R 24 may or may not have an etheric oxygen atom, but preferably has an etheric oxygen atom in view of better crosslinking reactivity and rubber physical properties. R 24The number of etheric oxygen atoms in R is preferably 1 or more and 6 or less, more preferably 1 or more and 3 or less, and even more preferably 1 or 2. 24 The etheric oxygen atom in R 24 It is preferred that the nucleotide sequence is located at the end of the nucleotide sequence.

[0029] Of the monomers represented by formula (2), specific examples of suitable monomers include the monomers represented by formula (3) and the monomers represented by formula (4). (CF2=CF-)2R 31 (3) In formula (3), R 31 represents a divalent perfluorohydrocarbon group having 1 to 10 carbon atoms, or a group having an etheric oxygen atom at the end or between the carbon-carbon bonds of a divalent perfluorohydrocarbon group having 1 to 10 carbon atoms. (CH2=CH-)2R 41 (4) In formula (4), R 41 represents a divalent perfluorohydrocarbon group having 1 to 10 carbon atoms, or a group having an etheric oxygen atom at the end or between the carbon-carbon bonds of a divalent perfluorohydrocarbon group having 1 to 10 carbon atoms.

[0030] Specific examples of the monomer represented by formula (3) include CF2=CFO(CF2)2OCF=CF2, CF2=CFO(CF2)3OCF=CF2, CF2=CFO(CF2)4OCF=CF2, and CF2=CFO(CF2)6OCF=CF 2、 CF2=CFO(CF2)8OCF=CF2, CF2=CFO(CF2)2OCF(CF3)CF2OCF=CF2, CF2=CFO(CF2)2O(CF(CF3)CF2O)2CF=CF2, CF2=CFOCF2O(CF2CF2O)2CF=CF2, CF2 =CFO(CF2O)3O(CF(CF3)CF2O)2CF=CF2, CF2=CFOCF2CF(CF3)O(CF2)2OCF(CF3)CF2OCF=CF2, and CF2=CFOCF2CF2O(CF2O)2CF2CF2OCF=CF2. Among the monomers represented by formula (3), specific examples of more suitable monomers include CF2=CFO(CF2)3OCF=CF2 (hereinafter also referred to as "C3DVE") and CF2=CFO(CF2)4OCF=CF2 (hereinafter also referred to as "C4DVE").

[0031] Specific examples of the monomer represented by formula (4) include CH2=CH(CF2)2CH=CH2, CH2=CH(CF2)4CH=CH2, and CH2=CH(CF2)6CH=CH2. Among the monomers represented by formula (4), a more preferred specific example of the monomer is CH2=CH(CF2)6CH=CH2 (hereinafter also referred to as "C6DV"). Among these, C3DVE or C4DVE is preferable for BO.

[0032] R Hal Examples of the monomer include a monomer having a bromine atom and a monomer having an iodine atom. Specific examples of monomers having a bromine atom include CF2=CFOCF2CF2CF2OCF2CF2Br, bromotrifluoroethylene, 4-bromo-3,3,4,4-tetrafluorobutene-1 (BTFB), vinyl bromide, 1-bromo-2,2-difluoroethylene, perfluoroallyl bromide, 4-bromo-1,1,2-trifluorobutene-1, 4-bromo-1,1,3,3,4,4-hexafluorobutene, 4-bromo-3-chloro-1,1,3,4,4-pentafluorobutene, 6-bromo-5,5,6,6-tetrafluorohexene, and 4-bromoperfluorobutene-1,3,3-difluoroallyl bromide.

[0033] Specific examples of the monomer having a bromine atom include 2-bromo-perfluoroethyl perfluorovinyl ether and CF2Br-R f Fluorinated compounds such as -O-CF=CF, for example, fluorovinyl ethers such as CFBrCFO-CF=CF, ROCF=CFBr, and ROCBr=CF, more specifically CHOCF=CFBr and CFCHOCF=CFBr.f is a perfluoroalkylene group, and R is a lower alkyl group or a fluoroalkyl group.

[0034] Specific examples of monomers having an iodine atom include iodinated olefins of the formula: CHR=CH-Z-CHCHR-I, where R is -H or -CH, and Z is a linear or branched C-C alkyl group optionally containing one or more ethereal oxygen atoms. 18 It is a (per)fluoroalkylene group or a (per)fluoropolyoxyalkylene group as disclosed in US Pat. No. 5,674,959.

[0035] Specific examples of the monomer having an iodine atom include the monomer of formula I(CHCFCF) disclosed in U.S. Pat. No. 5,717,036. n OCF = CF2 and ICH2CF2O [CF(CF3)CF2O] n Examples of unsaturated ethers include CF═CF2 (wherein n is 1 or more and 3 or less). Specific examples of monomers having an iodine atom include iodoethylene, 4-iodo-3,3,4,4-tetrafluorobutene-1 (ITFB), 3-chloro-4-iodo-3,4,4-trifluorobutene, 2-iodo-1,1,2,2-tetrafluoro-1-(vinyloxy)ethane, 2-iodo-1-(perfluorovinyloxy)-1,1,-2,2-tetrafluoroethylene, 1,1,2,3,3,3-hexafluoro-2-iodo-1-(perfluorovinyloxy)propane, 2-iodoethyl vinyl ether, 3,3,4,5,5,5-hexafluoro-4-iodopentene, and iodotrifluoroethylene, all of which are disclosed in U.S. Pat. No. 4,694,045. Specific examples of the monomer having an iodine atom include allyl iodide and 2-iodo-perfluoroethyl perfluorovinyl ether.

[0036] R CNFrom the viewpoint of polymerization reactivity, it is preferable that the copolymer has a polymerizable unsaturated bond, and more preferably has one polymerizable unsaturated bond. Specific examples of the polymerizable unsaturated bond include a carbon-carbon double bond (C=C) and a carbon-carbon triple bond (C≡C). R CN is preferably a monomer represented by formula (5) in that the crosslinked rubber article has better mold releasability and heat resistance. CR 51 R 52 =CR 53 -R 54 -CN (5) In equation (5), R 51 , R 52 , and R 53 each independently represents a hydrogen atom, a fluorine atom, or a methyl group; R 54 represents a divalent perfluorohydrocarbon group having 1 to 10 carbon atoms, or a group having an etheric oxygen atom at the end or between the carbon-carbon bonds of a divalent perfluorohydrocarbon group having 1 to 10 carbon atoms.

[0037] R CN Because of its excellent polymerization reactivity, 51 , R 52 , and R 53 is preferably a fluorine atom or a hydrogen atom, and R 51 , R 52 , and R 53 It is more preferable that all of R are fluorine atoms or all of R are hydrogen atoms, and in view of the superior mold releasability and heat resistance of the crosslinked rubber article, 51 , R 52 , and R 53 It is particularly preferred that all of are fluorine atoms. R 54 R may be linear, branched, or cyclic, and is preferably linear or branched. 54 The number of carbon atoms in is preferably 2 or more and 8 or less, more preferably 3 or more and 7 or less, even more preferably 3 or more and 6 or less, and particularly preferably 3 or more and 5 or less. R 54may or may not have an etheric oxygen atom, but preferably has an etheric oxygen atom in view of better rubber properties.

[0038] R 54 The number of etheric oxygen atoms in is preferably 1 or more and 3 or less, and particularly preferably 1 or 2. Specific examples of the monomer represented by formula (5) include CF2=CFOCF2CF(CF3)OCF2CF2CN (hereinafter also referred to as "8CNVE"), CF2=CFO(CF2)5CN (hereinafter also referred to as "MV5CN"), CF2=CFOCF2CF2CF2OCF(CF3)CN, and CF2=CFO(CF2)3CN, of which 8CNVE or MV5CN are preferred because they result in better mold releasability and heat resistance of crosslinked rubber articles.

[0039] POAVE is a compound represented by formula (6). CF2=CF(OCF2CF2) n -(OCF2) m -OR f2 (6) In formula (6), R f2 represents a perfluoroalkyl group having 1 or more and 4 or less carbon atoms, n represents an integer of 0 or more and 3 or less, m represents an integer of 0 or more and 4 or less, and n+m represents an integer of 1 or more and 7 or less.

[0040] R f2 In the formula (R), the perfluoroalkyl group may be linear or branched. f2 The number of carbon atoms is preferably 1 or more and 3 or less. When n is 0, m is preferably 3 or 4. When n is 1, m is preferably an integer of 2 or more and 4 or less. When n is 2 or 3, m is preferably 0. n is preferably an integer of 1 or more and 3 or less. R f2 When the number of carbon atoms, n, and m are within the above ranges, the low-temperature properties of the crosslinked rubber article are excellent, and the productivity of the crosslinked rubber article is improved.

[0041] Specific examples of POAVE include the following: The description in parentheses after the formula is the abbreviation for the compound. CF2=CF-OCF2CF2-(OCF2)4-OCF3(C9PEVE) CF2=CF-OCF2CF2-(OCF2)2-OCF3(C7PEVE) CF2=CF-(OCF2CF2)2-OCF2CF3(EEAVE) CF2=CF-(OCF2CF2)3-OCF2CF3(EEEAVE) CF2=CF-OCF2-OCF3, CF2=CF-OCF2-OCF2-OCF3 As POAVE, C9PEVE, C7PEVE, EEAVE, or EEEAVE is preferred in terms of better low-temperature properties and productivity of crosslinked rubber articles. These compounds can be produced using the corresponding alcohols as starting materials by the method described in WO 00 / 056694.

[0042] The content of other units in the present elastomer is preferably 0.01 mol% or more and 30 mol% or less, more preferably 0.01 mol% or more and 20 mol% or less, even more preferably 0.01 mol% or more and 10 mol% or less, and most preferably 0.01 mol% or more and 5 mol% or less, based on the total content of all units in the present elastomer.

[0043] In order to achieve superior crosslinkability, the present elastomer preferably contains at least one selected from the group consisting of a polymerizable unsaturated bond, a chlorine atom, a bromine atom, an iodine atom, and a nitrile group, and more preferably contains at least one selected from the group consisting of a chlorine atom, a bromine atom, an iodine atom, and a nitrile group, and it is particularly preferred that the present elastomer has such an atom or group at at least one of its terminals and side chains.

[0044] When producing this elastomer, by using the above-mentioned other monomers in addition to TFE and PAVE, it is possible to introduce polymerizable unsaturated bonds, chlorine atoms, bromine atoms, iodine atoms, or nitrile groups into the side chains or terminals of this elastomer. Also, by polymerizing the monomers using a chain transfer agent containing iodine atoms, it is possible to introduce iodine atoms into the terminals of this elastomer. When the present elastomer contains iodine atoms, the content of iodine atoms is preferably 0.01% by mass or more and 5.00% by mass or less, more preferably 0.01% by mass or more and 2.00% by mass or less, and even more preferably 0.01% by mass or more and 1.00% by mass or less, relative to the total mass of the present elastomer.

[0045] To improve the stability of the aqueous dispersion, the surfactant may contain a compound having an ionic functional group and a polymerizable unsaturated bond. Specific examples include vinyl sulfonic acid, allyl sulfonic acid, 2-acrylamido-2-methyl-1-propanesulfonic acid, N-tigloylglycine, 6-acrylamidohexanoic acid, 1,1-difluoro-2-methyl-2-[(1-oxo-2-propen-1-yl)amino]-1-propanesulfonic acid, 3-methyl-3-[(2-methyl-1-oxo-2-propen-1-yl)amino]-2-butanesulfonic acid, 2-methacrylamido-2-methylpropanesulfonic acid, 2,3-dimethyl-3-[(1-oxo-2-propen-1-yl)amino]-2-butanesulfonic acid, and metal salts thereof.

[0046] In order to improve the stability of the aqueous dispersion, auxiliary materials may be added. Examples of auxiliary materials include a pH adjuster, an emulsifier, an alcohol, and a catalyst. An aqueous dispersion of a fluoropolymer may also be added. The aqueous dispersion of a fluoropolymer may or may not contain an emulsifier.

[0047] The manufacturing process of this embodiment includes a first process PC1, a second process PC2 which is a manufacturing process subsequent to the first process PC1, and a third process PC3. As an example, the first step PC1 includes a polymerization step PC11. The second step PC2 includes an aggregation step PC21 and a drying step PC22. The third step PC3 includes a verification step PC31. In the verification step PC31, a production result indicator of the elastomer produced through the first step PC1 and the second step PC2 is verified. The production result indicators of the fluorine-containing elastomer confirmed in the confirmation step PC31 include the production efficiency of the fluorine-containing elastomer, a processability indicator showing the processability of the fluorine-containing elastomer, an indicator showing at least one of the static properties and dynamic properties of the crosslinked rubber article, and an indicator showing the stability of the aqueous dispersion obtained in the polymerization step. The production efficiency of the fluorine-containing elastomer includes the yield, the processability index showing the processability of the fluorine-containing elastomer includes the storage modulus G' and the Mooney viscosity, the index showing at least one of the static and dynamic properties of the crosslinked rubber article includes the tensile property after crosslinking and the compression set after crosslinking, and the index showing the stability of the aqueous dispersion obtained in the polymerization step includes the particle size of the emulsion and the solids concentration of the emulsion.

[0048] That is, the production process of this embodiment produces a fluorine-containing elastomer batch by batch, and includes a first process PC1 including a polymerization process PC11, a second process PC2 which is a production process subsequent to the first process PC1, and a third process PC3 including a confirmation process PC31 of a production result indicator 440 of the fluorine-containing elastomer produced via the first process PC1 and the second process PC2. After adding a crosslinking agent to the fluorine-containing elastomer produced through the first step PC1 and the second step PC2, the mixture is molded in a mold at a predetermined reaction temperature (primary vulcanization), and after being removed from the mold, may be heated again at a predetermined temperature (secondary vulcanization). Here, the predetermined reaction temperature is 100 to 400°C. Examples of crosslinking agents include isocyanurate derivatives, bisaminophenol derivatives, and polyfunctional organic peroxides. The amount of crosslinking agent added is 0.001 to 50 grams per 100 grams of fluorine-containing elastomer. In addition to the crosslinking agent, the composition may contain additives such as reactants such as organic peroxides, acid acceptors such as magnesium oxide, reinforcing materials such as carbon black and silica, catalysts such as organophosphorus compounds and organotin compounds, and processing aids such as calcium stearate. The first step PC1 may include other steps in addition to the polymerization step PC11. The second step PC2 may include other steps in addition to the aggregation step PC21 and the drying step PC22. The third step PC3 may include other steps in addition to the confirmation step PC31.

[0049] In the following description, a manufacturing process subsequent to a certain manufacturing process will be simply referred to as a “subsequent process,” and a manufacturing process prior to a certain manufacturing process will be simply referred to as a “previous process.” For example, the second process PC2 and the third process PC3 are subsequent processes of the first process PC1. The information processing device 3 of this embodiment presents manufacturing conditions PM1 for the polymerization step PC11, in particular, included in the first step PC1, among these manufacturing steps.

[0050] In the figure, of the multiple batches in the manufacturing process, the first batch BT1 to the third batch BT3 are shown as examples, and the batches after the third batch BT3 are not shown. Each of the first batch BT1 to the third batch BT3 includes a polymerization step PC11, an aggregation step PC21, a drying step PC22, and a verification step PC31. The polymerization step PC11 of the first batch BT1 is also referred to as the first polymerization step PC111. Similarly, the aggregation step PC21 of the first batch BT1 is also referred to as the second aggregation step PC211, the drying step PC22 as the first drying step PC221, and the verification step PC31 as the third verification step PC311. The steps of the second batch BT2 and the third batch BT3 are the same as those of the first batch BT1, and therefore the explanation thereof will be omitted.

[0051] In one example of this embodiment, the first batch BT1 to the third batch BT3 are adjacent batches on the time axis, i.e., the batch immediately following the first batch BT1 is the second batch BT2, and the batch immediately following the second batch BT2 is the third batch BT3. In the following description, batches adjacent to each other on the time axis will also be referred to as the nth batch BT(n), the n+1th batch BT(n+1), the n+2nd batch BT(n+2), etc., where n is a natural number. Any batch after the nth batch BT(n) will also be referred to as the mth batch BT(m), where m is a natural number greater than n. In particular, in relation to the nth batch BT(n) and the n+1th batch BT(n+1), the nth batch BT(n) is also called the "previous batch" and the n+1st batch BT(n+1) is also called the "next batch."

[0052] The nth batch BT(n) starts at time t1S and ends at time t1E. The n+1th batch BT(n+1) starts at time t2S and ends at time t2E. The n+2nd batch BT(n+2) starts at time t3S and ends at time t3E. The first process PC1 of the nth batch BT(n), for example, the first polymerization process PC111 shown in FIG. 2, ends at time t11E. Between time t11E when the first polymerization process PC111 ends and time t2S when the second polymerization process PC112 starts, maintenance work is performed on the manufacturing equipment used in the polymerization process PC11. Examples of maintenance work include light cleaning, heavy cleaning, replenishing or replacing consumables, and repairing or replacing broken parts. In the following description, maintenance work on manufacturing equipment is also referred to as servicing work.

[0053] In one example of this embodiment, the time t2S when the (n+1)th batch BT(n+1) starts is earlier than the time t1E when the nth batch BT(n) ends. In other words, in the batch production method of this embodiment, the next batch starts before the previous batch ends. In the following explanation, a batch production method configured to start the next batch before the previous batch ends is also referred to as a pipeline method. On the other hand, a batch production method in which the next batch is started after the previous batch is finished is also called a non-pipeline method. The pipeline method has less idle time for manufacturing equipment than the non-pipeline method. Therefore, adopting the pipeline method can improve the operating efficiency of manufacturing equipment compared to the non-pipeline method.

[0054] As described above, in the confirmation step PC31 for each batch, the production result indicators of the fluoroelastomer are confirmed. The production result indicators of the fluoroelastomer include indicators showing the production efficiency of the fluoroelastomer such as yield, indicators showing at least one of the static and dynamic properties of the crosslinked rubber article such as tensile properties after crosslinking and compression set after crosslinking, processability indicators showing the processability of the fluoroelastomer such as storage modulus G' and Mooney viscosity, and indicators showing the stability of the aqueous dispersion obtained in the polymerization step such as the particle size of the emulsion and the solids concentration of the emulsion. In one example of this embodiment, the proportion of these production result indices that are dominated by the production condition PM1 in the first process PC1 is higher than the proportion that is dominated by the indices of the production conditions in the second process PC2, which is the subsequent process. Specifically, the proportion that is dominated by the production condition PM1, such as the raw material index 420 and the production condition index 430, in the first process PC1, particularly in the polymerization process PC11, is higher than the proportion that is dominated by the indices of the production conditions in the second process PC2, which is the subsequent process. In other words, the production condition PM1 in the polymerization process PC11 has a dominant effect on these production result indices. Therefore, in order to stabilize or improve the yield and quality of each batch, it is desirable to feed back the production result index of the previous batch to the production condition PM1 of the polymerization step PC11 of the next batch.

[0055] On the other hand, if the above-mentioned pipeline method is adopted, the polymerization process PC11 of the next batch will be started before the manufacturing result indicator is obtained in the confirmation process PC31 of the previous batch. Specifically, the second polymerization process PC112 of the next batch will be started before the manufacturing result indicator is obtained in the third confirmation process PC311. 2, the time t2S when the second polymerization process PC112 of the (n+1)th batch BT(n+1) is started is earlier than the time t1E when the production result index of the nth batch BT(n) is obtained. In the example shown in the same figure, the production result index of the nth batch BT(n) is fed back to the production condition PM1 of the third polymerization process PC113, which is started at time t3S after time t1E.

[0056] In other words, when the pipeline method is adopted, the production result indicators confirmed in the confirmation process PC31 of the previous batch cannot be fed back to the production conditions PM1 of the polymerization process PC11 of the next batch. If the production result indicators of the previous batch could be fed back to the production conditions PM1 of the polymerization process PC11 of the next batch, this would be preferable because it would stabilize or improve the production efficiency and quality of the next batch. The information processing device 3 of this embodiment provides a function that can feed back the production result index of the previous batch to the production condition PM1 of the polymerization process PC11 of the next batch. The specific functional configuration of the information processing device 3 will be described below.

[0057] [Functional configuration of information processing device 3] FIG. 3 is a diagram showing an example of the functional configuration of the information processing device 3 of this embodiment. The information processing device 3 includes a manufacturing condition acquisition unit 310, a target value acquisition unit 320, a manufacturing information provision unit 330, a result acquisition unit 340, and a presentation unit 350 as its software function units (or hardware function units). Here, the explanation will be given assuming that the nth batch BT(n) is the previous batch and the (n+1)th batch BT(n+1) is the next batch.

[0058] The manufacturing condition acquisition unit 310 acquires the manufacturing conditions PM1 of the nth batch BT(n), that is, the previous batch. In the following description, the manufacturing conditions PM1 of the nth batch BT(n) will also be referred to as the nth manufacturing conditions PM1-(n). The manufacturing conditions PM1 include a raw material index 420 and a manufacturing condition index 430. The raw material indicators 420 include, for example, the amount of other monomers, the amount of polymerization initiator, etc. An example of the polymerization initiator is ammonium persulfate (APS). The production condition indicators 430 include the polymerization initiation pressure, polymerization pressure, polymerization time, the total amount of essential unit monomers supplied, inert analysis values, operating conditions of the agitator of the polymerization tank, etc. In the case of FFKM, an example of the total amount of essential unit monomers supplied is the total amount of PMVE and TFE supplied.

[0059] That is, the manufacturing condition acquisition unit 310 acquires the nth manufacturing condition PM1-(n), which is the manufacturing condition PM1 of the nth polymerization process PC11-(n), which is the polymerization process PC11 of the nth batch (n is a natural number), from the manufacturing conditions PM1 which include the resin raw material indicator 420 and the manufacturing condition indicator 430 in the polymerization process PC11. The nth manufacturing condition PM1-(n) may be provided by the user operating the input device 2, or may be provided from another external device such as a manufacturing management device without the intervention of a user. That is, the manufacturing condition acquisition unit 310 may acquire the nth manufacturing condition PM1-(n) via the user, or may acquire the nth manufacturing condition PM1-(n) from another external device without the intervention of a user.

[0060] The target value acquisition unit 320 acquires a target value for the production efficiency of the fluorine-containing elastomer produced in the mth batch BT(m), a target value for an index indicating at least one of the static properties and dynamic properties of the crosslinked rubber article, a target value for a processability index indicating the processability of the fluorine-containing elastomer, and a target value for an index indicating the stability of the aqueous dispersion obtained in the polymerization process. The production efficiency of the fluorine-containing elastomer is, for example, yield 444, and processability indices showing the processability of the fluorine-containing elastomer are, for example, storage modulus G' 445 and Mooney viscosity 446. Indices showing at least one of the static and dynamic properties of the crosslinked rubber article are tensile properties after crosslinking 447 and compression set after crosslinking 448, and indices showing the stability of the aqueous dispersion obtained in the polymerization step are, for example, emulsion particle size 442 and emulsion solids concentration 443. The mth batch BT(m) is, for example, the batch following the nth batch BT(n), i.e., the (n+1)th batch BT(n+1). That is, the target value acquisition unit 320 acquires the target value of the batch following the nth batch BT(n), which is the acquisition target of the manufacturing condition acquisition unit 310.

[0061] In other words, the target value acquisition unit 320 acquires the target value of the manufacturing result index 440 to be confirmed in the mth confirmation process PC31-(m) of the mth batch BT(m) (m is a natural number greater than n) that comes after the nth batch BT(n). The manufacturing information providing unit 330 provides the target value of the mth batch BT(m) acquired by the target value acquiring unit 320 and the nth manufacturing condition PM1-(n) acquired by the manufacturing condition acquiring unit 310 to the correspondence information 40 of the data server 4. The correspondence information 40 is information that indicates the correspondence relationship between the manufacturing conditions PM1 and the manufacturing result index 440 for each batch. An example of the correspondence information 40 will be described with reference to FIG.

[0062] 4 is a diagram showing an example of the correspondence information 40 of this embodiment. In the correspondence information 40, a batch number 410, a raw material index 420, a manufacturing condition index 430, and a manufacturing result index 440 are associated with each other for each batch. The raw material indicators 420 include an amount of elastomer raw material 421, an amount of initiator 422, and an amount of chain transfer agent 423. The manufacturing condition indicators 430 include a polymerization initiation pressure 431, a polymerization pressure 432, and a polymerization time 433. The raw material index 420 and the manufacturing condition index 430 are examples of manufacturing conditions PM1. The manufacturing result indicators 440 include emulsion particle size 442, emulsion solids concentration 443, yield 444, storage modulus G' 445, Mooney viscosity 446, tensile properties after crosslinking 447, and compression set after crosslinking 448.

[0063] In the following description, the production conditions PM1 for the kth batch, where the batch number 410 is k, will also be referred to as the kth production conditions PM1-(k). The production result index 440 for the k+1th batch, where the batch number 410 is k+1, will also be referred to as the k+1th production result index 440-(k+1). The production efficiency of the fluorine-containing elastomer for the k+1th batch, where the batch number 410 is k+1, the processability index indicating the processability of the fluorine-containing elastomer, the index indicating at least one of the static and dynamic properties of the crosslinked rubber article, and the index indicating the stability of the aqueous dispersion obtained in the polymerization step will also be referred to as the (k+1)th production efficiency of the fluorine-containing elastomer, the processability index (k+1) indicating the processability of the fluorine-containing elastomer, the index (k+1) indicating at least one of the static and dynamic properties of the crosslinked rubber article, and the index (k+1) indicating the stability of the aqueous dispersion obtained in the polymerization step, respectively.

[0064] The yield 444, tensile properties after crosslinking 447, compression set after crosslinking 448, storage modulus G' 445, Mooney viscosity 446, emulsion particle size 442, and emulsion solids concentration 443 of the (k+1)th batch, where batch number 410 is k+1, are also referred to as yield 444-(k+1), tensile properties after crosslinking 447-(k+1), compression set after crosslinking 448-(k+1), storage modulus G' 445-(k+1), Mooney viscosity 446-(k+1), emulsion particle size 442-(k+1), and emulsion solids concentration 443-(k+1), respectively. Returning to Figure 3, when the data server 4 is given the target value for the mth batch BT(m) and the nth manufacturing condition PM1-(n) from the manufacturing information provider 330, it selects the mth manufacturing condition PM1-(m) from the correspondence information 40 when the previous batch matches the nth manufacturing condition PM1-(n) and the subsequent batch matches the target value of the mth batch BT(m).

[0065] 5 is a diagram showing an example of a search result of the correspondence information 40 of this embodiment. As an example, it is assumed that the nth production conditions PM1-(n) match the kth production conditions PM1-(k), which are the production conditions PM1 of the kth batch BT(k), and the target values ​​of the mth batch BT(m) match the production result indicators 400 of the k+1th batch BT(k+1), including the emulsion particle size 442-(k+1), the emulsion solids concentration 443-(k+1), the yield 444-(k+1), the storage modulus G' 445-(k+1), the Mooney viscosity 446-(k+1), the tensile properties after crosslinking 447-(k+1), and the compression set after crosslinking 448-(k+1). In this case, the data server 4 selects the k+1-th manufacturing condition PM1-(k+1), which is the manufacturing condition PM1 of the k+1-th batch BT(k+1), from among the manufacturing conditions PM1 in the correspondence information 40.

[0066] In the above example, a specific example will be described where (n = 10, m = n+1, k = 1). That is, it is assumed that the tenth production conditions PM1-10 match the production conditions PM1-1 of the first batch BT1, and the target values ​​for the eleventh batch BT11 match the production result indicators 400 of the second batch BT2, including emulsion particle size 442-2, emulsion solids concentration 443-2, yield 444-2, storage modulus G' 445-2, Mooney viscosity 446-2, post-crosslink tensile property 447-2, and post-crosslink compression set 448-2. In this case, the data server 4 selects the second manufacturing conditions PM1-2, which are the manufacturing conditions PM1 of the second batch BT2, from among the manufacturing conditions PM1 in the correspondence information 40. Note that "match" here may mean that the values ​​of the compared indices are completely the same, or that the difference between the values ​​of the compared indices is within a predetermined range.

[0067] 3, the data server 4 outputs the selected k+1th manufacturing condition PM1-(k+1) as the estimation result of the manufacturing condition PM1 to the information processing device 3. For example, the data server 4 outputs the second manufacturing condition PM1-2 to the information processing device 3 as the estimation result of the manufacturing condition PM1. The result acquisition unit 340 acquires the manufacturing conditions PM1 output by the data server 4 as the estimated results of the manufacturing conditions PM1 of the polymerization process PC11 of the m-th batch BT(m). That is, the result acquisition unit 340 acquires the mth production condition PM1-(m) output from the correspondence information 40, which is given the nth production condition PM1-(n) of the production condition PM1 and the target value, as the estimated result of the production condition PM1 of the polymerization process PC11 of the mth batch BT(m).

[0068] The presentation unit 350 outputs the inference result acquired by the result acquisition unit 340 to the display device 5. The display device 5 displays the inference result output by the presentation unit 350. In other words, the presentation unit 350 presents the inference result acquired by the result acquisition unit 340. The flow of operations of each functional unit of the information processing device 3 described above will be described with reference to FIG.

[0069] [Operation flow of information processing device 3] FIG. 6 is a diagram showing an example of the flow of operations of the information processing device 3 of this embodiment. (Step S1-1) The manufacturing condition acquisition unit 310 acquires the nth manufacturing conditions PM1-(n), which are the manufacturing conditions PM1 of the polymerization process PC11 of the nth batch BT(n). (Step S2-1) The target value acquisition section 320 acquires the target value of the production result index 440 of the fluorine-containing elastomer produced in the m-th batch BT(m).

[0070] (Step S3-1) The manufacturing information provider 330 provides the nth manufacturing condition PM1-(n) acquired in step S1-1 and the target value of the mth batch BT(m) acquired in step S2-1 to the correspondence information 40 of the data server 4. The data server 4 selects, from among the manufacturing conditions PM1 and manufacturing result indicators 440 in the correspondence information 40, the mth manufacturing condition PM1-(m) when the previous batch matches the nth manufacturing condition PM1-(n) and the subsequent batch matches the target value of the mth batch BT(m). The data server 4 outputs the selected mth manufacturing condition PM1-(m) to the information processing device 3 as an estimation result of the manufacturing condition PM1. (Step S4-1) The result acquiring unit 340 acquires the m-th manufacturing condition PM1-(m) output by the data server 4 as an estimated result of the manufacturing condition PM1. (Step S5-1) The presentation unit 350 outputs to the display device 5 the m-th manufacturing condition PM1-(m), which is the estimation result of the manufacturing condition PM1 acquired in step S4-10.

[0071] In the above-described embodiment, the manufacturing result indicators 440 may include at least one of emulsion particle size 442, emulsion solids concentration 443, yield 444, storage modulus G' 445, Mooney viscosity 446, tensile properties after crosslinking 447, and compression set after crosslinking 448. As described above, the information processing system 1 of this embodiment estimates the production conditions PM1 for the polymerization process PC11 of the next batch based on the production result index for the polymerization process PC11 of the previous batch and the target value for the next batch in a pipeline production process. Therefore, according to the information processing system 1 of this embodiment, the production result index confirmed in the confirmation process PC31 of the previous batch can be fed back to the production conditions PM1 for the polymerization process PC11 of the next batch before the previous batch is completed. The information processing system 1 configured in this way can detect quality fluctuations in manufactured products such as resins at an early stage and reflect these changes in the next batch at an early stage. In other words, the information processing system 1 allows for timely recipe changes.

[0072] Furthermore, the information processing system 1 includes the amount 423 of the chain transfer agent as a production condition PM1 in the correspondence information 40. When methanol is used as the chain transfer agent in the resin polymerization step PC11, a slight change in the concentration of methanol may have a significant effect on at least one of the items included in the production result index 440 of the fluorine-containing elastomer. In such a case, the amount of, for example, methanol as a chain transfer agent included in the production conditions PM1 in the polymerization step PC11 has a greater effect on at least one of the items included in the production result index 440 of the fluorine-containing elastomer than various production conditions in subsequent steps. According to the information processing system 1 of this embodiment, the production conditions PM1 for the next batch are estimated taking into account the amount 423 of the chain transfer agent, thereby making it possible to further stabilize or improve the items included in the production result index 440 of the fluorine-containing elastomer for the next batch.

[0073] [Variations] The information processing device 3 may estimate the production conditions PM1 for the polymerization process PC11 of the next batch, taking into account the maintenance index PM2. Here, the maintenance index PM2 is an index that indicates the maintenance status of manufacturing equipment. For example, the maintenance index PM2 includes the implementation status of maintenance activities such as equipment cleaning, and the replacement status of consumables such as packing, filters, belts, valves, desiccants, and oil. Here, the implementation status of maintenance activities includes, for example, the number of batches after heavy cleaning work, and the replacement status of consumables includes, for example, the number of batches after valve or packing replacement.

[0074] That is, the manufacturing condition PM1 includes a maintenance index PM2 that indicates the maintenance status of the manufacturing equipment. In this case, the correspondence information 40 indicates the correspondence relationship between the n-th manufacturing condition PM1-(n) that includes the maintenance index PM2, the m-th manufacturing condition PM1-(m), and the m-th manufacturing condition PM1-(m). The manufacturing condition acquisition unit 310 acquires the n-th manufacturing condition PM1-(n) including the maintenance index PM2. The manufacturing information providing unit 330 provides the target value and the n-th manufacturing condition PM1-(n) including the maintenance index PM2 to the correspondence information 40. That is, the manufacturing information providing unit 330 provides the n-th manufacturing condition PM1-(n) taking the maintenance index PM2 into consideration to the correspondence information 40.

[0075] As a result, the result acquisition unit 340 acquires the estimation result of the manufacturing condition PM1 taking the maintenance index PM2 into consideration. The presentation unit 350 outputs the estimation result of the manufacturing condition PM1 taking the maintenance index PM2 into consideration to the display device 5. Generally, when maintenance of manufacturing equipment is performed, such as deep cleaning or replacement of consumables, the manufacturing environment may change significantly between the previous batch and the next batch, or the manufacturing environment may change discontinuously. Furthermore, the manufacturing environment may change gradually with each batch since the maintenance of the manufacturing equipment.

[0076] The information processing system 1 configured as in this modified example can obtain the estimated results of the production conditions PM1 that take into account the maintenance status of the production equipment. Therefore, even if the production environment changes due to maintenance of the production equipment, the information processing system 1 can stabilize or improve the items included in the production result index 440 of the fluorine-containing elastomer for the next batch.

[0077] [Variation 2] In the above embodiment, the correspondence information 40 is described as information in a table format in which the manufacturing conditions PM1 and the manufacturing result index 440 are associated with each other for each batch, but the present invention is not limited to this. The correspondence information 40 may be any information that indicates the correspondence relationship between the nth manufacturing condition PM1-(n), the mth manufacturing condition PM1-(m), and the target value of the mth batch BT(m). Specifically, the correspondence information 40 may be any information that indicates the correspondence relationship between the manufacturing condition PM1 of the previous batch and the manufacturing condition PM1 and target value of the next batch. For example, the correspondence information 40 may be a so-called learned model in which the correspondence relationship between the manufacturing condition PM1 of the previous batch and the manufacturing condition PM1 and target value of the next batch is learned by machine learning.

[0078] For example, the trained model includes a first training model created by machine learning the relationship between the manufacturing conditions of a batch and the manufacturing result indicators confirmed in the confirmation process for subsequent batches, based on a first training dataset, with information indicating the manufacturing conditions of the batch as explanatory variables and manufacturing result indicators confirmed in the confirmation process for subsequent batches as target variables. Here, the first training dataset includes information indicating the manufacturing conditions of the batch as training data and the manufacturing result indicators confirmed in the confirmation process for subsequent batches as training data.

[0079] Furthermore, the trained model includes a second learning model created by machine learning the relationship between the manufacturing conditions of a batch and the manufacturing result indicators confirmed in the confirmation process for subsequent batches and information indicating the manufacturing conditions of subsequent batches, based on the second learning dataset, with the manufacturing conditions of the batch and the manufacturing result indicators confirmed in the confirmation process for subsequent batches as explanatory variables and the manufacturing conditions of subsequent batches as objective variables. Here, the second learning dataset includes the manufacturing conditions of the batch and the manufacturing result indicators confirmed in the confirmation process for subsequent batches as training data, and information indicating the manufacturing conditions of subsequent batches as training data.

[0080] In this case, the data server 4 stores correspondence information 40 as a learned model in which the correspondence relationship between the manufacturing conditions PM1 of the previous batch and the manufacturing conditions PM1 and target values ​​of the next batch is learned by machine learning. The combination of explanatory variables and response variables can be set as appropriate. For example, the explanatory variable may be at least one of the amount of chain transfer agent, polymerization temperature, amount of initiator, and polymerization pressure, and the response variable may be the storage modulus G'. In this case, for example, if the storage modulus G' is too high, it can be adjusted by performing at least one of increasing the amount of chain transfer agent, raising the polymerization temperature, increasing the amount of initiator, and lowering the polymerization pressure. Alternatively, for example, the explanatory variable may be at least one of the amount of initiator, polymerization temperature, and auxiliary materials, and the objective variable may be productivity such as yield, etc. Alternatively, for example, the explanatory variable may be at least one of the amount of chain transfer agent, the amount of initiator, polymerization temperature, and raw material composition, and the objective variable may be compression set after crosslinking. The information processing system 1 configured in this manner can use the results of learning large amounts of information efficiently or with high precision by machine learning, and can therefore further stabilize or improve the productivity of the fluorine-containing elastomer in the next batch, the processability index showing the processability of the fluorine-containing elastomer, the index showing at least one of the static properties and dynamic properties of the crosslinked rubber article, and the index showing the stability of the aqueous dispersion obtained in the polymerization step. Furthermore, the correspondence information 40 may be implemented by a conditional branching program that implements the correspondence between input values ​​and output values ​​corresponding to the trained model described above.

[0081] [Learning model creation device] A method for creating a trained model will be described. The trained model is created by a learning model creation device. That is, the learning model creation device creates the trained model. Note that the information processing device 3 may include the learning model creation device. That is, the information processing device 3 may create the trained model. Also, the data server 4 may include the learning model creation device. That is, the data server 4 may create the trained model.

[0082] 7 is a diagram showing an example of a learning model creation device 200 according to this embodiment. The learning model creation device 200 is realized by a device such as a personal computer, a server, a smartphone, a tablet computer, or an industrial computer. The learning model creation device 200 creates a first learning model by machine learning the relationship between the manufacturing conditions for each batch and the manufacturing result index confirmed in the confirmation process for a subsequent batch, based on the first learning dataset, using information indicating the manufacturing conditions for each batch as explanatory variables and manufacturing result indexes confirmed in the confirmation process for a subsequent batch as target variables. The first learning dataset includes information indicating the manufacturing conditions for each batch as learning data and manufacturing result indexes confirmed in the confirmation process for a subsequent batch as training data.

[0083] The learning model creation device 200 creates a second learning model by machine learning the relationship between the manufacturing conditions of a batch and the manufacturing result indicators confirmed in the confirmation process for subsequent batches, and information indicating the manufacturing conditions of subsequent batches, based on the second learning dataset, using the manufacturing conditions of the batch and the manufacturing result indicators confirmed in the confirmation process for subsequent batches as explanatory variables and the manufacturing conditions of subsequent batches as target variables. The second learning dataset includes the manufacturing conditions of the batch and the manufacturing result indicators confirmed in the confirmation process for subsequent batches as learning data, and information indicating the manufacturing conditions of subsequent batches as training data. That is, the learning model creation device 200 trains learning models including a first learning model and a second learning model using training data such as a first learning data set and a second learning data set, and creates a trained model. Here, the learning model is a model that is the basis of the trained model.

[0084] For example, the learning model creation device 200 constructs a learned model using algorithms such as a convolution neural network (CNN), a recurrent neural network (RNN), a long short-term memory (LSTM), a random forest, a support vector machine (SVM), or a neural network. An input sample is data that is input to an input layer when training a learning model. An output sample is data such as training data that serves as a correct answer for comparison with an output value output from an output layer when training a learning model.

[0085] The learning model creation device 200 includes an input unit 202 , a receiving unit 204 , a processing unit 206 , an output unit 208 , and a storage unit 210 . The input unit 202 inputs information. As an example, the input unit 202 may have an operation unit such as a keyboard and a mouse. In this case, the input unit 202 inputs information according to an operation performed by a user on the operation unit. As another example, the input unit 202 may input information from an external device. The external device may be, for example, a portable storage medium. A training dataset is input to the input unit 202.

[0086] The receiving unit 204 acquires a training dataset from the input unit 202 and accepts the acquired training dataset. The training dataset includes a first training dataset and a second training dataset. The training dataset includes input samples and output samples, and the input samples and output samples are paired. The training dataset is made up of a plurality of pairs.

[0087] The processing unit 206 includes a learning model 207, which includes a first learning model and a second learning model. For all pairs of the first learning dataset, the processing unit 206 calculates the error between the output value output from the output layer when an input sample is input to the input layer of the first learning model and the output sample corresponding to the input sample, and changes the parameters of the first learning model so that the error is as small as possible. For all pairs of the second learning dataset, the processing unit 206 calculates the error between the output value output from the output layer when an input sample is input to the input layer of the second learning model and the output sample corresponding to the input sample, and changes the parameters of the second learning model so that the error is as small as possible. The processing unit 206 creates a learned model by changing the parameters of the first learning model and the second learning model. Here, the output sample is an example of training data. The learning model 207 is trained by changing the parameters of the learning model 207. The trained model created as described above is received by the data server 4 from the output unit 208 via a network or a medium. When the learning model creation device 200 is included in the information processing device 3, the result acquisition unit 340 may acquire the trained model from the learning model creation device 200.

[0088] All or part of the input unit 202, the reception unit 204, the processing unit 206 and the output unit 208 are functional units (hereinafter referred to as software functional units) that are realized, for example, by a processor such as a CPU executing a program stored in the memory unit 210. In addition, all or part of the input unit 202, the reception unit 204, the processing unit 206 and the output unit 208 may be realized by hardware such as an LSI, an ASIC or an FPGA, or may be realized by a combination of software functional units and hardware.

[0089] (Operation of the learning model creation device 200) FIG. 8 is a flowchart showing an example of the operation of the learning model creation device 200 of this embodiment. (Step S1-2) The input unit 202 acquires a learning dataset. (Step S2-2) The receiving unit 204 acquires the training data set from the input unit 202 and accepts the acquired training data set.

[0090] (Step S3-2) The processing unit 206 acquires the learning dataset from the receiving unit 204. The processing unit 206 trains the parameters of the learning model 207 based on the acquired learning dataset, and creates the learning model. (Step S4-2) The output unit 208 acquires the learning model 207 from the processing unit 206. The output unit 208 outputs the acquired learning model 207 as a trained model. The learning method for creating a trained model is not limited to supervised learning, but may be any of unsupervised learning, semi-supervised learning, reinforcement learning, or deep learning, or may be a combination of these learning methods; any learning method for machine learning is acceptable.

[0091] As described above, the learning model creation device 200 of this embodiment creates a learning model based on the first learning data set and the second learning data set. With this configuration, the learning model creation device 200 can create a learned model. Therefore, by using the learned model created by the learning model creation device 200, the information processing device 3 can shorten the time required to determine the manufacturing conditions for the subsequent batch and improve the accuracy of determining the manufacturing conditions for the subsequent batch.

[0092] [Second embodiment] This embodiment differs from the first embodiment in that it includes an information processing device 31 instead of or in addition to the information processing device 3 in the first embodiment. The other configurations are the same as those in the first embodiment, and therefore description thereof will be omitted. FIG. 9 is a diagram showing an example of the functional configuration of an information processing device 31 according to the second embodiment. The information processing device 31 has, as its functional parts, a manufacturing condition acquisition part 311, a first manufacturing information provision part 331, a second manufacturing information provision part 332, a first result acquisition part 341, a second result acquisition part 342, an index presentation part 351, and a result presentation part 352.

[0093] The manufacturing condition acquisition unit 311 acquires the n-th manufacturing condition PM1-(n), which is the manufacturing condition PM1 of the n-th polymerization step PC11-(n). The first manufacturing information providing unit 331 provides the n-th manufacturing condition PM1-(n) acquired by the manufacturing condition acquiring unit 311 to the first correspondence information 41 as manufacturing information. The first correspondence information 41 is information indicating the correspondence relationship between the manufacturing conditions PM1 for each batch and the manufacturing result indicators 440 confirmed in the confirmation process PC31. In one example of this embodiment, the first correspondence information 41 has the same configuration as the correspondence information 40 described above.

[0094] When the data server 4 receives the n-th manufacturing condition PM1-(n) from the first manufacturing information providing unit 331, it searches the first correspondence information 41 and selects a manufacturing result index 440 that matches the n-th manufacturing condition PM1-(n). The search result of the manufacturing result index 440 by the data server 4 will be described with reference to FIG. 10 is a diagram showing an example of a search result of the first correspondence information 41 of this embodiment. As an example, it is assumed that the nth manufacturing condition PM1-(n), which is the manufacturing condition PM1 of the nth batch BT(n), matches the kth manufacturing condition PM1-(k), which is the manufacturing condition PM1 of the kth batch BT(k). In this case, the data server 4 selects, from the first correspondence information 41, the kth manufacturing result index 440-(k), which is the manufacturing result index 440 of the kth batch BT(k).

[0095] A specific example where (n=10, m=n+1, k=1) in the above example will be described. That is, it is assumed that the tenth manufacturing condition PM1-(10) matches the manufacturing condition PM1-1 of the first batch BT1. In this case, the data server 4 selects the first manufacturing result index 440-1, which is the manufacturing result index 440 of the first batch BT1, from the first correspondence information 41. As in the first embodiment, "match" here may mean that the values ​​of the compared indices are completely identical, or that the difference between the values ​​of the compared indices is within a predetermined range.

[0096] Returning to FIG. 9, the data server 4 outputs the selected manufacturing result index 440 to the information processing device 31. The first result acquisition unit 341 acquires the manufacturing result index 440 output from the first correspondence information 41 given the nth manufacturing condition PM1-(n) as the nth manufacturing result index 440-(n), which is the manufacturing result index 440 of the nth batch BT(n). The index presenting unit 351 presents the n-th manufacturing result index 440-(n) acquired by the first result acquiring unit 341 by outputting it to the display device 5.

[0097] The second manufacturing information providing unit 332 provides the second correspondence information 42 with the nth manufacturing result index 440(n) acquired by the first result acquiring unit 341 and the nth manufacturing condition PM1-(n) acquired by the manufacturing condition acquiring unit 311. The second correspondence information 42 is information indicating the correspondence relationship between the manufacturing conditions PM1 for each batch and the manufacturing result index 440. In one example of this embodiment, the second correspondence information 42 has the same configuration as the correspondence information 40 described above.

[0098] When the data server 4 receives the nth manufacturing condition PM1-(n) and the nth manufacturing result index 440-(n) from the second manufacturing information provider 332, it searches the second correspondence information 42 and selects the (n+1)th manufacturing condition PM1-(n+1), which is the manufacturing condition PM1 of the next batch of the batch that matches the nth manufacturing condition PM1-(n) and the nth manufacturing result index 440-(n). The search result of the data server 4 for the manufacturing condition PM1 of the next batch will be described with reference to FIG. As an example, suppose that the nth manufacturing conditions PM1-(n) of the nth batch BT(n) match the kth manufacturing conditions PM1-(k) of the kth batch BT(k), and the nth manufacturing result index 440-(n) of the nth batch BT(n) matches the kth manufacturing result index 440-(k) of the kth batch BT(k). In this case, the data server 4 selects, from the second correspondence information 42, the k+1st manufacturing condition PM1-(k+1), which is the manufacturing condition PM1 of the k+1st batch BT(k+1).

[0099] In the above example, a specific example will be described where (n=10, m=n+1, k=1). That is, it is assumed that the tenth production condition PM1-10 matches the first production condition PM1-1 of the first batch BT1, and the tenth production result index 440-10 matches the first production result index 440-1 of the first batch BT1. In this case, the data server 4 selects, from the second correspondence information 42, the second manufacturing result index 440-2, which is the manufacturing result index 440 of the batch next to the first batch BT1, that is, the second batch BT2. As in the first embodiment, "match" here may mean that the values ​​of the compared indices are completely identical, or that the difference between the values ​​of the compared indices is within a predetermined range.

[0100] The second result acquisition unit 342 acquires the mth production condition PM1-(m) output from the second correspondence information 42 given the nth production condition PM1-(n) and the nth production result index 440(n) as an estimated result of the production condition PM1 of the polymerization process PC11 of the mth batch BT(m). The result presenting unit 352 presents the estimation result acquired by the second result acquiring unit 342 by outputting it to the display device 5. The flow of operations of each functional unit of the information processing device 3 described above will be described with reference to FIG.

[0101] [Operation flow of information processing device 3] FIG. 11 is a diagram showing an example of the flow of operations of the information processing device 31 of this embodiment. (Step S1-3) The manufacturing condition acquisition unit 311 acquires the nth manufacturing conditions PM1-(n), which are the manufacturing conditions PM1 of the polymerization process PC11 of the nth batch BT(n). (Step S2-3) The first manufacturing information providing unit 331 provides the n-th manufacturing condition PM1-(n) acquired in step S1-3 to the first correspondence information 41 of the data server 4. The data server 4 selects the nth manufacturing result index 440-(n), which is the manufacturing result index 440 of the batch that matches the nth manufacturing condition PM1-(n), from the manufacturing conditions PM1 in the first correspondence information 41. The data server 4 outputs the selected nth manufacturing result index 440-(n) to the information processing device 31 as an estimated result of the manufacturing result index 440.

[0102] (Step S3-3) The first result acquisition unit 341 acquires the n-th manufacturing result index 440-(n) output by the data server 4 as an estimated result of the manufacturing result index 440. (Step S4-3) The second manufacturing information providing unit 332 provides the second correspondence information 42 with the n-th manufacturing condition PM1-(n) acquired in step S1-3 and the n-th manufacturing result index 440-(n) acquired in step S3-3.

[0103] The data server 4 selects the mth manufacturing condition PM1-(m) of the mth batch BT(m), which corresponds to a batch in which the manufacturing condition PM1 in the second correspondence information 42 matches the nth manufacturing condition PM1-(n) and the manufacturing result index 440 in the second correspondence information 42 matches the nth manufacturing result index 440-(n). For example, in the above-mentioned case (m=n+1), the data server 4 selects the (n+1)th batch BT(n+1), i.e., the (n+1)th manufacturing condition PM1-(n+1), which is the manufacturing condition PM1 of the batch next to the nth batch BT(n). The data server 4 outputs the selected (n+1)th manufacturing condition PM1-(n+1) to the information processing device 31 as the estimated result of the manufacturing condition PM1.

[0104] (Step S5-3) The second result acquisition unit 342 acquires the estimation result of the manufacturing condition PM1 output in step S4-3. (Step S6-3) The result presenting unit 352 presents the m-th manufacturing condition PM1-(m), which is the estimation result of the manufacturing condition PM1 acquired in step S5-3, by outputting it to the display device 5.

[0105] As described above, the information processing system 1 of this embodiment includes the information processing device 31. In a pipeline-type manufacturing process, the information processing device 31 estimates the manufacturing result index 440 of a batch based on the manufacturing result index of the polymerization process PC11 of the previous batch. That is, the information processing device 31 can estimate the manufacturing result of the batch after the polymerization process PC11 is completed and before the batch is completed. Specifically, the information processing device 31 can estimate the manufacturing result of the batch after the polymerization process PC11 is completed and before the second process PC2, for example, is started.

[0106] Furthermore, the information processing device 31 of this embodiment estimates the production conditions PM1 for the polymerization process PC11 of the next batch based on the result of the polymerization process PC11 of the previous batch and the estimated production result at the end of the previous batch. That is, the information processing device 31 estimates the production result of the previous batch based on the production result index of the polymerization process PC11 of the previous batch, and estimates the production conditions PM1 for the next batch based on the estimated production result. Therefore, according to the information processing device 31 of this embodiment, the production result index confirmed in the confirmation step PC31 of the previous batch can be fed back to the production condition PM1 of the polymerization step PC11 of the next batch before the previous batch is completed. The information processing system 1 configured in this manner can detect quality fluctuations in the manufactured product (e.g., fluorine-containing elastomer) at an early stage and reflect the changes in the next batch at an early stage. In other words, the information processing system 1 can make timely recipe changes.

[0107] The information processing device 31 of this embodiment also includes an index presentation unit 351. The index presentation unit 351 presents a manufacturing result index 440 of the previous batch that the information processing device 31 is using for estimation. According to the information processing system 1 configured in this manner, it is possible to present to the user the estimated production result index 440 at the time of the yield of the previous batch at the time of completion of the polymerization step PC11 of the previous batch. Therefore, it is possible to quickly detect any fluctuations in the quality of the product being manufactured (e.g., a fluorine-containing elastomer) and reflect them in the next batch. In other words, the information processing system 1 allows timely recipe changes.

[0108] [Variations] Note that, similarly to the information processing device 3 described above, the information processing device 31 may estimate the production conditions PM1 of the polymerization process PC11 of the next batch, taking into account the maintenance index PM2. The information processing device 31 configured in this manner can obtain an estimated result of the manufacturing conditions PM1 that takes into account the maintenance status of the manufacturing equipment. Therefore, even if the manufacturing environment changes due to maintenance of the manufacturing equipment, the information processing system 1 can stabilize or improve the processability of the next batch and the index that indicates at least one of the static and dynamic properties of the crosslinked rubber article.

[0109] Furthermore, the first correspondence information 41 and the second correspondence information 42 may be a so-called trained model that has been machine-learned, similar to the correspondence information 40 described above. For example, the trained model includes a first trained model and a second trained model. The first trained model may be created by machine learning the relationship between the manufacturing condition PM1 and the manufacturing result index 440 for each batch, with the manufacturing condition PM1 for each batch as an explanatory variable and the manufacturing result index 440 as an objective variable. The second trained model may be created by machine learning the relationship between the manufacturing condition PM1 and the manufacturing result index 440 for a batch and the manufacturing condition PM1 for the next batch, with the manufacturing condition PM1 and the manufacturing result index 440 for a batch as explanatory variables and the manufacturing condition PM1 for the next batch as an objective variable. The information processing device 31 configured in this manner can use the results of learning large amounts of information efficiently or with high precision through machine learning, thereby making it possible to further stabilize or improve processability indicators that indicate the processability of the fluoroelastomer, such as the production efficiency of the next batch and the storage modulus G'. Furthermore, the first correspondence information 41 and the second correspondence information 42 may be information in which the correspondence between input values ​​and output values ​​corresponding to the trained model described above is implemented by a conditional branching program.

[0110] [Learning model creation device] A method for creating a trained model will be described. The trained model is created by a training model creation device. That is, the training model creation device creates the trained model. Note that the information processing device 31 may include the training model creation device. That is, the information processing device 31 may create the trained model. Also, the data server 4 may include the training model creation device. That is, the data server 4 may create the trained model.

[0111] 12 is a diagram showing an example of a learning model creation device 250 of this embodiment. The learning model creation device 250 is realized by a device such as a personal computer, a server, a smartphone, a tablet computer, or an industrial computer. The learning model creation device 250 trains a first learning model using training data such as a first learning dataset in which information indicating the manufacturing conditions PM1 for each batch is used as an input sample and information indicating a manufacturing result index is used as an output sample.The learning model creation device 250 trains a second learning model using training data such as a second learning dataset in which information indicating the manufacturing conditions PM1 for each batch and information indicating a manufacturing result index are used as input samples and information indicating the manufacturing conditions for the batch following that batch is used as an output sample.The learning model creation device 250 creates a learning model by training the first learning model and the second learning model.Here, the learning model is a model that serves as the basis for the trained model.

[0112] For example, the learning model creation device 250 constructs a learned model using algorithms such as CNN, RNN, LSTM, random forest, SVM, and neural network. An input sample is data input to an input layer when training a learning model. An output sample is data such as training data that serves as a correct answer for comparison with an output value output from an output layer when training a learning model. The learning model creation device 250 includes an input unit 252 , a receiving unit 254 , a processing unit 256 , an output unit 258 , and a storage unit 260 . The input unit 252 may be implemented as the input unit 202. A training dataset is input to the input unit 252. The training dataset includes a first training dataset and a second training dataset. The receiving unit 254 may be implemented as the receiving unit 204.

[0113] Processing unit 256 includes learning model 257, which includes a first learning model and a second learning model. For all pairs of the first learning dataset, processing unit 256 inputs an input sample of the first learning dataset into the input layer of the first learning model, calculates the error between the output value output from the output layer and the output sample corresponding to the input sample, and changes the parameters of the first learning model so as to minimize the error. For all pairs in the second learning dataset, the processing unit 256 inputs an input sample into the input layer of the second learning model, calculates the error between the output value output from the output layer and the output sample corresponding to the input sample, and changes the parameters of the second learning model so that the error is as small as possible. The processing unit 256 creates a learned model by changing the parameters of the first learning model and the second learning model. Here, the output sample is an example of training data. The learning model 257 is trained by changing the parameters of the first learning model and the second learning model included in the learning model 257.

[0114] The trained model created as described above is received by the data server 4 from the output unit 258 via a network or a medium. When the learning model creation device 250 is included in the information processing device 31, the first result acquisition unit 341 may acquire the first trained model from the learning model creation device 250, and the second result acquisition unit 342 may acquire the second trained model from the learning model creation device 250.

[0115] All or part of the input unit 252, the reception unit 254, the processing unit 256 and the output unit 258 are functional units (hereinafter referred to as software functional units) that are realized, for example, by a processor such as a CPU executing a program stored in the memory unit 260. In addition, all or part of the input unit 252, the reception unit 254, the processing unit 256, and the output unit 258 may be realized by hardware such as an LSI, an ASIC, or an FPGA, or may be realized by a combination of software functional units and hardware.

[0116] (Operation of the learning model creation device 250) FIG. 13 is a flowchart showing an example of the operation of the learning model creation device 250 of this embodiment. (Step S1-3) The input unit 252 acquires a learning dataset. (Step S2-3) The receiving unit 254 acquires the training data set from the input unit 252 and accepts the acquired training data set.

[0117] (Step S3-3) The processing unit 256 acquires the learning dataset from the receiving unit 254. The processing unit 256 trains the parameters of the learning model 257 based on the acquired learning dataset, and creates the learning model. (Step S4-3) The output unit 258 acquires the learning model 257 from the processing unit 256. The output unit 258 outputs the acquired learning model 257 as a trained model. The learning method for creating a trained model is not limited to supervised learning, but may be any of unsupervised learning, semi-supervised learning, reinforcement learning, or deep learning, or may be a combination of these learning methods; any learning method for machine learning is acceptable.

[0118] As described above, the learning model creation device 250 of this embodiment creates a learning model based on the first learning data set and the second learning data set. With this configuration, the learning model creation device 250 can create a learned model. Therefore, by using the learned model created by the learning model creation device 250, the information processing device 31 can shorten the time required to determine the manufacturing conditions for the subsequent batch and improve the accuracy of determining the manufacturing conditions for the subsequent batch. Furthermore, the information processing system 1 may be configured by appropriately combining the functions of the information processing device 3 of the first embodiment and the information processing device 31 of the second embodiment.

[0119] A program for implementing the functions of any of the components of any of the above-described devices may be recorded on a computer-readable recording medium and loaded into a computer system for execution. The term "computer system" as used herein includes hardware such as an operating system or peripheral devices. The term "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and compact discs (CDs) and read-only memories (ROMs), as well as storage devices such as hard disks built into computer systems. The term "computer-readable recording medium" also includes devices that retain a program for a certain period of time, such as volatile memory within a computer system that acts as a server or client when a program is transmitted over a network such as the Internet or a communication line such as a telephone line. Such volatile memory may be, for example, random access memory (RAM). The recording medium may also be, for example, a non-transitory recording medium.

[0120] The above program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by transmission waves in the transmission medium. Here, the "transmission medium" that transmits the program refers to a medium that has the function of transmitting information, such as a network such as the Internet or a communication line such as a telephone line. The above program may also be one that realizes part of the above-mentioned functions. Furthermore, the above program may be a so-called differential file that can realize the above-mentioned functions in combination with a program already recorded in a computer system. A differential file may also be called a differential program.

[0121] Furthermore, the functions of any of the components in any of the above-described devices may be implemented by a processor. For example, each process in the embodiments may be implemented by a processor operating based on information such as a program and a computer-readable recording medium storing information such as the program. Here, the functions of each unit of the processor may be implemented by, for example, individual hardware, or may be implemented by integrated hardware. For example, the processor may include hardware, and the hardware may include at least one of a circuit for processing digital signals and a circuit for processing analog signals. For example, the processor may be configured using one or more circuit devices mounted on a circuit board, or one or both of one or more circuit elements. An integrated circuit (IC) or the like may be used as the circuit device, and a resistor or a capacitor may be used as the circuit element.

[0122] Here, the processor may be, for example, a CPU. However, the processor is not limited to a CPU, and various types of processors may be used, such as a GPU (Graphics Processing Unit) or a DSP (Digital Signal Processor). Furthermore, the processor may be, for example, a hardware circuit based on an ASIC (Application Specific Integrated Circuit).

[0123] The processor may be configured with, for example, multiple CPUs, or may be configured with a hardware circuit using multiple ASICs. The processor may also be configured with, for example, a combination of multiple CPUs and a hardware circuit using multiple ASICs. The processor may also include, for example, one or more of an amplifier circuit or a filter circuit that processes analog signals. The embodiments of this disclosure have been described in detail above with reference to the drawings, but the specific configuration is not limited to this embodiment, and includes designs within the scope that do not deviate from the gist of this disclosure. [Explanation of symbols]

[0124] 1...information processing system, 2...input device, 3...information processing device, 4...data server, 5...display device, 200, 250...learning model creation device, 202, 252...input unit, 204, 254...receiving unit, 206, 256...processing unit, 207, 257...learning model, 208, 258...output unit, 210, 260...storage unit, 310...manufacturing condition acquisition unit, 320...target value acquisition unit, 330...manufacturing information providing unit, 340...result acquisition unit, 350...presentation unit, 40...correspondence information, 31...information processing device, 311...manufacturing condition acquisition unit, 331...first manufacturing information providing unit, 332...second manufacturing information providing unit, 341...first result acquisition unit, 342...second result acquisition unit, 351...index presentation unit, 352...result presentation unit, 41...first correspondence information, 42...second correspondence information,

Claims

1. 1. An information processing device which includes a polymerization step and a confirmation step of a fluorine-containing elastomer produced through the polymerization step, and which presents production conditions for a production step of a fluorine-containing elastomer for each batch, a production condition acquisition unit that acquires n-th production conditions, which are production conditions for an n-th polymerization step that is the polymerization step of an n-th batch (n is a natural number), from production conditions including indicators of raw materials of the fluorine-containing elastomer in the polymerization step and indicators of polymerization conditions for the fluorine-containing elastomer; a target value acquiring unit that acquires a target value of a manufacturing result indicator confirmed in the confirmation step for an mth batch (m is a natural number greater than n) that is subsequent to the nth batch; a production information providing unit that provides the target value of the production result index acquired by the target value acquiring unit and the nth production condition acquired by the production condition acquiring unit in response to correspondence information indicating the correspondence between the production conditions and the production result index of the fluorine-containing elastomer for each batch; a result acquisition unit that acquires the m production condition output from the correspondence information to which the n production condition and the target value of the production result index are given as an estimated result of the production condition of the polymerization step of the m batch; An information processing device comprising:

2. The manufacturing conditions include a maintenance index indicating the maintenance status of the manufacturing equipment, the correspondence information indicates a correspondence relationship between the nth production condition including the maintenance index, the mth production condition, and a production result index of the mth batch of fluorine-containing elastomer, the manufacturing condition acquisition unit acquires the nth manufacturing condition including the maintenance index, The manufacturing information providing unit provides the target value of the manufacturing result index and the nth manufacturing condition including the maintenance index to the corresponding information. The information processing device according to claim 1 .

3. 1. An information processing device which includes a polymerization step and a confirmation step of a production result index showing the production result of a fluorine-containing elastomer produced through the polymerization step, and which presents production conditions for a production step of producing a fluorine-containing elastomer for each batch, a production condition acquisition unit that acquires an nth production condition, which is the production condition for an nth polymerization step that is the polymerization step of an nth batch (n is a natural number), from production conditions including an index of raw materials of the fluorine-containing elastomer in the polymerization step, an index of polymerization conditions for the fluorine-containing elastomer, and a maintenance index that indicates the maintenance status of production equipment; a first manufacturing information providing unit that provides the nth manufacturing condition acquired by the manufacturing condition acquiring unit as manufacturing information in response to first correspondence information indicating a correspondence relationship between the manufacturing conditions and manufacturing result indicators confirmed in the confirmation step; a first result acquisition unit that acquires the manufacturing result index output from the first correspondence information to which the nth manufacturing condition is given as an nth manufacturing result index that is a manufacturing result index of the nth batch; a second manufacturing information providing unit that provides the nth manufacturing result index acquired by the first result acquiring unit and the nth manufacturing condition acquired by the manufacturing condition acquiring unit with respect to second correspondence information that indicates a correspondence relationship between manufacturing conditions and manufacturing result indexes for each batch; a second result acquisition unit that acquires the m production conditions output from the second correspondence information to which the n production conditions and the n production result index are given as estimated results of production conditions for the polymerization step of the m batch; An information processing device comprising:

4. 4. The information processing device according to claim 1, wherein the production result indicators of the fluorine-containing elastomer include at least one of a production efficiency of the fluorine-containing elastomer, a processability indicator showing at least one of the static properties and the dynamic properties of the crosslinked rubber article, and an indicator showing the stability of the aqueous dispersion obtained in the polymerization step.

5. a polymerization step and a confirmation step of a production result index of a fluorine-containing elastomer produced through the polymerization step, wherein, for a production step of producing a fluorine-containing elastomer on a batch-by-batch basis, information indicating production conditions for each batch is included as learning data and the production result index confirmed in the confirmation step of a batch subsequent to the first batch is included as training data, using the information indicating the production conditions for each batch as an explanatory variable and the production result index confirmed in the confirmation step of a batch subsequent to the first batch as a target variable, to create a first learning model by machine learning the relationship between the production conditions for each batch and the production result index confirmed in the confirmation step of a batch subsequent to the first batch; a processing unit that creates a second learning model by machine learning the relationship between the manufacturing result indicators confirmed in the confirmation step of a batch subsequent to the first batch and information indicating the manufacturing conditions of the subsequent batch, using the manufacturing result indicators confirmed in the confirmation step of a batch subsequent to the first batch as explanatory variables and the manufacturing conditions of the subsequent batch as objective variables, based on a second learning dataset that includes the manufacturing result indicators confirmed in the confirmation step of a batch subsequent to the first batch as learning data and information indicating the manufacturing conditions of the subsequent batch as training data. A learning model creation device comprising:

6. a production process for producing a fluorine-containing elastomer batch by batch, the production process comprising a polymerization step and a confirmation step of a production result index showing the production result of a fluorine-containing elastomer produced through the polymerization step, the production conditions for each batch including an index of raw materials of the fluorine-containing elastomer in the polymerization step, an index of polymerization conditions for the fluorine-containing elastomer, and a maintenance index showing the maintenance status of production equipment, are included as learning data, and the production result index confirmed in the confirmation step of the batch is included as training data, to create a first learning model by machine learning the relationship between information showing the production conditions and information showing the production result index, with the production conditions as explanatory variables and the production result index as a response variable; a processing unit that creates a second learning model by machine learning the relationship between information indicating the manufacturing conditions and the manufacturing result index and the manufacturing conditions, using the manufacturing conditions and the manufacturing result index as explanatory variables and the manufacturing conditions of the subsequent batch as objective variables, based on a second learning dataset that includes the manufacturing conditions and the manufacturing result index for each batch as learning data and the manufacturing conditions of the subsequent batch as training data; A learning model creation device comprising:

7. 1. An information processing method for presenting production conditions for a production process for producing a fluorine-containing elastomer for each batch, comprising: a polymerization step; and a confirmation step of a fluorine-containing elastomer produced through the polymerization step, acquiring n-th production conditions which are production conditions for an n-th polymerization step which is the polymerization step of an n-th batch (n is a natural number) from production conditions including indicators of raw materials of the fluorine-containing elastomer in the polymerization step and indicators of polymerization conditions for the fluorine-containing elastomer; acquiring a target value of a manufacturing result indicator to be confirmed in the confirmation step for an mth batch (m is a natural number greater than n) subsequent to the nth batch; providing the acquired target value of the production result index and the acquired nth production condition to correspondence information indicating the correspondence relationship between the production conditions and the production result index of the fluorine-containing elastomer for each batch; acquiring the m production condition output from the correspondence information in which the n production condition and the target value of the production result index are given as an estimated result of the production condition of the polymerization step of the m batch; An information processing method including:

8. 1. An information processing method for presenting production conditions for a production process for producing a fluorine-containing elastomer for each batch, comprising: a polymerization step; and a confirmation step of a production result index showing the production result of a fluorine-containing elastomer produced through the polymerization step, acquiring n-th production conditions which are production conditions for an n-th polymerization step which is the polymerization step of the n-th batch (n is a natural number) from among production conditions which include an index of raw materials of the fluorine-containing elastomer in the polymerization step, an index of polymerization conditions for the fluorine-containing elastomer, and a maintenance index which indicates the maintenance status of production equipment; providing the acquired nth manufacturing condition as manufacturing information to first correspondence information indicating a correspondence relationship between the manufacturing condition and a manufacturing result index confirmed in the confirmation step; acquiring the manufacturing result index output from the first correspondence information to which the nth manufacturing condition is given as an nth manufacturing result index that is a manufacturing result index of the nth batch; providing the acquired nth manufacturing result index and the acquired nth manufacturing condition to second correspondence information indicating a correspondence relationship between the manufacturing condition and the manufacturing result index for each batch; acquiring the m production condition output from the second correspondence information to which the n production condition and the n production result index are given as an estimated result of the production condition of the polymerization step of the m batch; An information processing method including:

9. a first learning model is created by machine learning the relationship between the production conditions for each batch and the production result index confirmed in the confirmation step of a batch subsequent to the first batch, using the information indicating the production conditions for each batch as an explanatory variable and the production result index confirmed in the confirmation step of a batch subsequent to the first batch as a target variable, based on a first learning dataset which includes, for a production process of producing a fluorine-containing elastomer for each batch, information indicating the production conditions for each batch as learning data and the production result index confirmed in the confirmation step of a batch subsequent to the first batch as training data; A learning model creation method executed by a learning model creation device, which creates a second learning model by machine learning the relationship between the manufacturing result indicators confirmed in the confirmation process of batches subsequent to the batch in question and the information indicating the manufacturing conditions of the subsequent batch, using the manufacturing result indicators confirmed in the confirmation process of batches subsequent to the batch in question as explanatory variables and the manufacturing conditions of the subsequent batch as target variables, based on a second learning dataset which includes the manufacturing result indicators confirmed in the confirmation process of batches subsequent to the batch in question as learning data and information indicating the manufacturing conditions of the subsequent batch as training data.

10. a production process for producing a fluorine-containing elastomer batch by batch, the production process comprising a polymerization step and a confirmation step of a production result index showing the production result of a fluorine-containing elastomer produced through the polymerization step, the production conditions for each batch including an index of raw materials of the fluorine-containing elastomer in the polymerization step, an index of polymerization conditions for the fluorine-containing elastomer, and a maintenance index showing the maintenance status of production equipment, are included as learning data, and the production result index confirmed in the confirmation step of the batch is included as training data, to create a first learning model by machine learning the relationship between information showing the production conditions and information showing the production result index, with the production conditions as explanatory variables and the production result index as a response variable; A learning model creation method executed by a learning model creation device, which creates a second learning model by machine learning the relationship between information indicating the manufacturing conditions and manufacturing result indexes and the manufacturing conditions, using the manufacturing conditions and manufacturing result indexes as explanatory variables and the manufacturing conditions of the subsequent batch as target variables, based on a learning dataset that includes the manufacturing conditions and manufacturing result indexes for each batch as learning data and the manufacturing conditions of the subsequent batch as training data.

11. A program that causes a computer to function as the information processing device according to any one of claims 1 to 4.

12. A program that causes a computer to function as the learning model creation device according to claim 5 or 6.

Images