Physical property prediction device, physical property prediction method, and manufacturing method
The physical property prediction device optimizes monomer selection and ratios in polymer synthesis by using ECFP and Bayesian optimization, addressing the scale issues in traditional methods and enhancing prediction accuracy.
Patent Information
- Application Number
- JP2020101108
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-06-10
- Publication Date
- 2024-11-06
- Estimated Expiration
- 2040-06-10
AI Technical Summary
Traditional Bayesian optimization methods for predicting polymer physical properties require manual rescaling based on experience, as the predicted values often have different scales or numerical ranges.
A physical property prediction device and method that uses a model for each physical property type, based on monomer information and content ratios, to calculate deviation values and determine optimal monomer and ratios for polymer synthesis, utilizing Extended Connectivity Fingerprints (ECFP) and Bayesian optimization.
Enables accurate and efficient evaluation of polymer properties by optimizing monomer selection and ratios, avoiding manual rescaling and improving the precision of property predictions.
Smart Images

Figure 0007577934000004 
Figure 0007577934000005 
Figure 0007577934000006
Abstract
Description
[Technical field]
[0001] The present invention relates to an apparatus for predicting physical properties of a polymer, a method for predicting physical properties, and a production method. [Background technology]
[0002] In recent years, artificial intelligence (AI) technology has come to be used in various technical fields. In the fields of chemistry and materials, AI technology is applied as materials informatics (MI) and is used in the research and development of new substances and materials.
[0003] For example, when synthesizing a polymer using multiple monomers, AI technology can be used to explore what ratio of these multiple monomers is necessary to synthesize a polymer with desired physical properties. The structural characteristics of compounds such as monomers are expressed according to any expression method suitable for computational processing, such as ECFP (Extended Connectivity Fingerprint), and a regression model can be constructed in which the ECFP of each monomer and its content ratio are used as explanatory variables, and a certain physical property of the polymer is used as the objective variable. By using the constructed regression model, it becomes possible to predict the physical properties of a polymer synthesized from an unknown content ratio of monomers.
[0004] It is sometimes necessary to evaluate a polymer based on multiple physical properties. In this case, a regression model is constructed using each physical property as a response variable, and multiple physical property values can be predicted from the monomer content ratio. When predicting such multiple physical property values, for example, Bayesian optimization can be used. [Prior art documents] [Non-patent literature]
[0005] [Non-Patent Document 1] David Rogers and Mathew Hahn, "Extended-Connectivity Fingerprints". J. Chem. Inf. Model, 50, 742 (2010) [Non-Patent Document 2] Fabian Pedregosa, et. al., "Scikit-learn: Machine Learning in Python", JMLR. 12, 2825 (2011) Summary of the Invention [Problem to be solved by the invention]
[0006] However, the predicted values of each physical property generally have different scales or numerical ranges, and in conventional Bayesian optimization, in order to comprehensively evaluate a polymer based on multiple physical properties, rescaling of each objective variable based on the researcher's experience, etc. was required.
[0007] In order to solve the above problems, an object of the present invention is to provide a method for evaluating the physical properties of a polymer based on a plurality of physical property values. [Means for solving the problem]
[0008] [1] A physical property prediction device having: a physical property prediction unit that obtains predicted values of multiple physical property types of a polymer using a model for each physical property type based on monomer information indicating multiple monomers and their content ratios; and a monomer determination unit that calculates a deviation value for the monomer information from the predicted values of the multiple physical property types, and determines monomers used in synthesizing the polymer and their content ratios based on the deviation value.
[0009] [2] The property prediction device described in [1], wherein the model for each property type is derived by Gaussian process regression using training data consisting of pairs of the monomer information and property values for the property type of the polymer corresponding to the monomer information.
[0010] [3] The property prediction device according to [1] or [2], wherein the monomer determination unit calculates an output value of an acquisition function of Bayesian optimization from the predicted values of the plurality of property types and calculates a deviation value of the output value.
[0011] [4] The property prediction device described in any one of [1] to [3], wherein the monomer determination unit determines the monomer information for which the deviation value is maximum or minimum as the monomers and their content ratios used in the synthesis of the polymer.
[0012] [5] Further comprising a monomer information generating unit that generates the monomer information, The property prediction device according to any one of [1] to [4], wherein the monomer information is expressed by a product of an ECFP (Extended Connectivity Fingerprint) of each monomer and a molar ratio.
[0013] [6] A method for predicting physical properties comprising the steps of: one or more processors obtaining predicted values for a plurality of physical property types of a polymer using a model for each physical property type based on monomer information indicating a plurality of monomers and their content ratios; and the one or more processors calculating a deviation value for the monomer information from the predicted values of the plurality of physical property types, and determining the monomers and their content ratios used in synthesizing the polymer based on the deviation value.
[0014] [7] The method for predicting physical properties described in [6], wherein the model for each physical property type is derived by Gaussian process regression using training data consisting of pairs of the monomer information and physical property values for the physical property type of the polymer corresponding to the monomer information.
[0015] [8] The property prediction method according to [6] or [7], wherein the determining step calculates an output value of an acquisition function of Bayesian optimization from the predicted values of the plurality of property types, and calculates a deviation value of the output value.
[0016] [9] The method for predicting physical properties described in any one of [6] to [8], wherein the determining step determines the monomer information for which the deviation value is maximum or minimum as the monomers and their content ratios used in the synthesis of the polymer.
[0017]
[10] The method further comprising the step of generating the monomer information, The property prediction method according to any one of [6] to [9], wherein the monomer information is expressed by the product of the ECFP (Extended Connectivity Fingerprint) and the molar ratio of each monomer.
[0018]
[11] [6] to
[10] The monomer and its content ratio are determined by the physical property prediction method according to any one of the above. Cured polyurethane A manufacturing method for producing the above. Effect of the Invention
[0019] According to the above aspect of the present invention, it is possible to provide a method for evaluating the physical properties of a polymer based on a plurality of physical property values. [Brief description of the drawings]
[0020] [Figure 1] 1 is a schematic diagram illustrating a regression model generating device and a property predicting device according to an embodiment of the present invention. [Diagram 2] 1 is a block diagram showing a hardware configuration of a regression model generating device and a property predicting device according to an embodiment of the present invention. FIG. [Diagram 3] 1 is a block diagram showing a functional configuration of a regression model generating device according to an embodiment of the present invention. [Figure 4] FIG. 2 illustrates an ECFP according to one embodiment of the present invention. [Diagram 5] 1 is a flowchart illustrating a regression model generation process according to an embodiment of the present invention. [Figure 6] 1 is a block diagram showing a functional configuration of a property prediction apparatus according to an embodiment of the present invention. [Figure 7]FIG. 13 is a diagram showing predicted values of multiple types of physical properties according to an embodiment of the present invention. [Figure 8] FIG. 13 illustrates an index according to one embodiment of the present invention. [Figure 9] FIG. 2 is a diagram showing a diol and its content ratio and a catalyst that maximizes the index according to an embodiment of the present invention. [Figure 10] FIG. 10 is a graph showing the measured values of reduced transmittance, breaking stress, and elongation of a cured polyurethane product obtained by curing a polyurethane precursor synthesized based on the diols and their contents and catalysts shown in FIG. [Figure 11] FIG. 2 is a diagram showing diols and their contents, and catalysts of Comparative Examples (1) to (10). [Figure 12] FIG. 12 is a graph showing the measured values of reduced transmittance, breaking stress, and elongation of a cured polyurethane product obtained by curing a polyurethane precursor synthesized based on the diols and their contents and catalysts shown in FIG. [Figure 13] 1 is a flowchart showing a property prediction process according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0021] The present invention will now be described in detail with reference to examples.
[0022] In an embodiment described later, a regression model generation device is disclosed that generates a regression model for each physical property type to predict physical property values for each type of polymer from monomers used in polymer synthesis and their content ratios, and a physical property prediction device that predicts physical property values for each type of polymer from monomers and their content ratios using the generated regression model for each physical property type and calculates an index for comprehensively evaluating the physical properties of the polymer from the predicted physical property values for each type.
[0023] [Summary] First, an overview of a regression model generating device and a physical property predicting device according to an embodiment of the present invention will be described with reference to Fig. 1. Fig. 1 is a schematic diagram showing a regression model generating device and a physical property predicting device according to an embodiment of the present invention.
[0024] 1, the regression model generating device 100 acquires teacher data or training data consisting of pairs of monomer information indicating monomers for synthesizing a polymer and their content ratios, and physical property values of multiple types of polymers (N types in the illustrated example) synthesized based on the monomer information, from a substance database 50. Then, the regression model generating device 100 constructs regression models for each type (estimation target regression models #1, #2, . . . , #N) that predict the physical property values of each physical property type from the acquired monomer information, and generates regression models for each type for all types.
[0025] Next, the physical property prediction device 200 acquires a predicted value for each physical property type from the monomer information using the regression model for each type (regression models #1, #2, ..., #N) previously constructed by the regression model generation device 100, and calculates an index for evaluating the overall physical properties of the polymer from the acquired multiple predicted values. The physical property prediction device 200 calculates an index for each monomer information, and determines the monomers and their content ratios (such as molar ratios) used in the synthesis of the polymer based on the index. For example, a deviation value may be used as the index, and the monomer and its content ratio for which the calculated deviation value is maximum or minimum may be selected.
[0026] The regression model generating device 100 and the property prediction device 200 may have a hardware configuration as shown in Fig. 2. That is, the regression model generating device 100 and the property prediction device 200 have a drive device 101, an auxiliary storage device 102, a memory device 103, a CPU (Central Processing Unit) 104, an input / output device 105, and a communication device 106, which are interconnected via a bus B.
[0027] Various computer programs including programs for implementing various functions and processes in the regression model generating device 100 and the physical property predicting device 200 described later may be provided by a recording medium 107 such as a CD-ROM (Compact Disk-Read Only Memory). When the recording medium 107 storing the programs is set in the drive device 101, the programs are installed in the auxiliary storage device 102 from the recording medium 107 via the drive device 101. However, the installation of the programs does not necessarily have to be performed by the recording medium 107, and may be downloaded from any external device via a network or the like. The auxiliary storage device 102 stores the installed programs and also stores necessary files and data. When an instruction to start a program is received, the memory device 103 reads out and stores the programs and data from the auxiliary storage device 102. The CPU 104 functioning as a processor executes various functions and processes of the regression model generating device 100 and the physical property predicting device 200 described later according to the programs stored in the memory device 103 and various data such as parameters required to execute the programs. The input / output device 105 provides an interface between the regression model generating device 100 and the property prediction device 200 and a user, and may be composed of, for example, input devices such as a keyboard, a mouse, a touch screen, etc., and output devices such as a display, a speaker, etc. The communication device 106 executes various communication processes for communicating with external devices.
[0028] However, the regression model generating device 100 and the property predicting device 200 are not limited to the above-mentioned hardware configurations, and may be realized by any other appropriate hardware configuration, such as a processing circuit wired to realize the functions and processing described below.
[0029] [Regression model generator] Next, a regression model generating device 100 according to an embodiment of the present invention will be described with reference to Fig. 3. Fig. 3 is a block diagram showing the functional configuration of the regression model generating device 100 according to an embodiment of the present invention.
[0030] As shown in FIG. 3, the regression model generating device 100 includes a training data acquiring unit 110 and a regression model generating unit 120.
[0031] The training data acquisition unit 110 acquires teacher data or training data consisting of pairs of monomer information and multiple types of physical property values of a polymer synthesized based on the monomer information. The monomer information is not limited, and may indicate multiple monomers used to synthesize a polymer and their content ratios.
[0032] For example, when a polyurethane is synthesized by polyaddition reaction of a diol and a diisocyanate using a catalyst, the monomers used in the monomer information are various diols, and the content ratio may be the molar ratio of the diol used. Here, each diol may be quantified according to any appropriate expression method such as ECFP (Extended Connectivity Fingerprint) based on the structural information extracted from the diol, as shown in FIG. 4. In addition, the physical property type may be a film property such as the reduced transmittance, breaking stress, and elongation of polyurethane. In addition, the number of moles may be given as training data instead of the molar ratio. In this case, the sum of the molar ratios of the monomers is normalized to be 1, and the molar ratio is calculated from the number of moles.
[0033] Here, the converted transmittance is expressed by the formula 4×(refractive index) / [(refractive index)+1] 2 (Refractive index ≧ 1) It is calculated as follows.
[0034] For example, the monomer information of sample 1 of the training data may indicate diol 1, diol 2, and diol 3 as monomers, with a molar ratio of "0.41:0.01:0.11." Furthermore, the physical property values of polyurethane synthesized based on the monomer information of sample 1 may be reduced transmittance "0.971," breaking stress "20.1," and elongation "129.2."
[0035] Here, diols 1 to 3 are as follows.
[0036] [ka] Similarly, the monomer information of Sample 2 may indicate diol 1, diol 2, and diol 3 as monomers, with a molar ratio of "0.42:0.01:0.10." Furthermore, the physical property values of polyurethane synthesized based on the monomer information of Sample 2 may be reduced transmittance "0.973," break stress "31.2," and elongation "300.3."
[0037] Each sample of such training data is acquired in advance by an experiment or the like, and is stored, for example, in the substance database 50. The training data acquisition unit 110 acquires the training data from the substance database 50, and passes it to the regression model generation unit 120.
[0038] The regression model generating unit 120 generates a regression model for each type from the training data. Specifically, the regression model generating unit 120 constructs a regression model for each type using the monomer information of the training data provided by the training data acquiring unit 110 as an explanatory variable and the physical property value of each physical property type as a response variable.
[0039] For example, when the above-mentioned samples 1, 2, ... are given, the regression model generation unit 120 first quantifies the monomers, which are non-quantified data items. For example, the regression model generation unit 120 may convert the monomers into a vector representation according to ECFP. Specifically, the regression model generation unit 120 can extract three partial structures shown in the center from the monomer structure shown on the left side of Fig. 4 according to a known ECFP derivation method, and obtain the vector representation shown on the right side.
[0040] After calculating the ECFP of each monomer in this manner, the regression model generating unit 120 multiplies the molar ratio of each sample by the vector of each ECFP. For example, when diol 1 is X1=(X 11 ,X 12 , ), and diol 2 is X2 = (X 21 ,X 22, ) and diol 3 is X3 = (X 31 ,X 32 , . . .), the regression model generating unit 120 calculates the following for the molar ratio of sample 1, “0.41× 11 ,0.41X 12 , , 0.01X 21 ,0.01X 22 , , 0.11X 31 ,0.11X 32 Similarly, the regression model generation unit 120 calculates (0.42× 11 ,0.42X 12 , , 0.01X 21 ,0.01X 22 , , 0.10X 31 ,0.10X 32 ,···).
[0041] Next, the regression model generation unit 120 calculates the monomer information of sample 1: (0.41X 11 ,0.41X 12 , , 0.01X 21 ,0.01X 22 , , 0.11X 31 ,0.11X 32 ,···);Physical properties: Converted transmittance 0.971, breaking stress 20.1, elongation 129.2"Subsample 1-1 by physical properties"Monomer information: (0.41X 11 ,0.41X 12 , , 0.01X 21 ,0.01X 22 , , 0.11X 31 ,0.11X 32 ,···);Physical properties: Reduced transmittance 0.971", Subsample 1-2" Monomer information: (0.41X 11 ,0.41X 12 , , 0.01X 21 ,0.01X 22 , , 0.11X 31 ,0.11X 32 ,···);Physical properties: Breaking stress 20.1", and subsample 1-3" Monomer information: (0.41X11 ,0.41X 12 , , 0.01X 21 ,0.01X 22 , , 0.11X 31 ,0.11X 32 ,···);Physical properties:Elongation 129.2".
[0042] Similarly, the regression model generation unit 120 generates sample 2' monomer information: (0.42X 11 ,0.42X 12 , , 0.01X 21 ,0.01X 22 , , 0.10X 31 ,0.10X 32 ,···);Physical properties: Converted transmittance 0.973, breaking stress 31.2, elongation 300.3"From subsample 2-1 by physical properties type"Monomer information: (0.42X 11 ,0.42X 12 , , 0.01X 21 ,0.01X 22 , , 0.10X 31 ,0.10X 32 ,···);Physical properties: Reduced transmittance 0.973", Subsample 2-1" Monomer information: (0.42X 11 ,0.42X 12 , , 0.01X 21 ,0.01X 22 , , 0.10X 31 ,0.10X 32 ,···);Physical properties: Breaking stress 31.2", and subsample 2-32-1" Monomer information: (0.42X 11 ,0.42X 12 , , 0.01X 21 ,0.01X 22 , , 0.10X 31 ,0.10X 32 ,...); physical property value: elongation "300.3" is generated. In the same manner, the regression model generation unit 120 generates sub-samples i-1, i-2, and i-3 for each sample i.
[0043] Then, the regression model generation unit 120 extracts subsamples 1-1, 2-1, ... and performs regression analysis on these subsamples i-1 in order to construct a regression model #1 for each type with the monomer information as the explanatory variable and the converted transmittance as the objective variable. Specifically, the regression model generation unit 120 constructs a regression model #1 for each type so as to reproduce the subsamples 1-1, 2-1, ... according to Gaussian process regression.
[0044] Here, Gaussian process regression is a function y from the input variable x to the output variable y. i =f(x i ) One of its features is its nonlinearity, making it effective when linear regression does not provide a good fit. Another important feature is that it uses Bayesian estimation. The estimated function is not a single function, but is obtained as a distribution of functions, making it possible to express the uncertainty of the estimation.
[0045] And for each y i The relationship between the y i Normal distribution of each y i Therefore, the covariance describing the relationship between y i Since the values of have a normal distribution, for example, if we take a 95% confidence interval, we have an uncertainty of 2σ i It becomes possible to predict the value of
[0046] Similarly, the regression model generation unit 120 extracts subsamples 1-2, 2-2, ... and performs regression analysis on these subsamples i-2 to construct a regression model #2 for each type with monomer information as an explanatory variable and breaking stress as a response variable. Specifically, the regression model generation unit 120 constructs a regression model #2 for each type so as to reproduce the subsamples 1-2, 2-2, ... according to Gaussian process regression.
[0047] Furthermore, the regression model generation unit 120 extracts subsamples 1-3, 2-3, ... and performs regression analysis on these subsamples i-3 in order to construct a regression model #3 for each type with monomer information as an explanatory variable and elongation as a response variable. Specifically, the regression model generation unit 120 constructs the regression model #3 for each type so as to reproduce the subsamples 1-3, 2-3, ... according to Gaussian process regression.
[0048] In this way, when the regression models #1, #2, and #3 for each type are constructed, the regression model generating unit 120 passes the constructed regression models #1, #2, and #3 to the property prediction device 200.
[0049] [Regression model generation process] Next, a regression model generation process according to an embodiment of the present invention will be described with reference to Fig. 5. The regression model generation process may be executed by, for example, the above-described regression model generation device 100, and more specifically, may be realized by a processor of the regression model generation device 100. Fig. 5 is a flowchart showing the regression model generation process according to an embodiment of the present invention.
[0050] 5, in step S101, the regression model generating device 100 acquires training data. For example, when polyurethane is synthesized as a polymer, the training data is composed of pairs of monomer information indicating monomers (e.g., various diols, etc.) and their content ratios (e.g., number of moles, molar ratio, etc.) and multiple types of physical property values of the polymer (e.g., reduced permeability, breaking stress, elongation, etc.), and is acquired in advance by an experiment or the like.
[0051] In step S102, the regression model generation device 100 calculates a partial structure and a molar ratio from the monomer information for each sample of the acquired training data. Specifically, the regression model generation device 100 calculates the ECFP of each monomer in the monomer information and converts each monomer into a vector representation. In addition, the regression model generation device 100 calculates the molar ratio from the number of moles.
[0052] In step S103, the regression model generating device 100 calculates explanatory variable values of the regression model from the calculated partial structure and molar ratio. Specifically, the regression model generating device 100 calculates a product vector of the ECFP calculated in step S102 and the molar ratio, and sets the calculated product vector as an explanatory variable value of the regression model.
[0053] In step S104, the regression model generating device 100 associates the explanatory variable values calculated in step S103 with the physical property values of each type of the training data acquired in step S101, and constructs a regression model for each type.
[0054] Specifically, the regression model generating device 100 associates the vector of each sample of monomer information as the explanatory variable value calculated in step S104 with the converted transmittance for that sample, and constructs a regression model #1 for each type that predicts the converted transmittance from the vector of monomer information.
[0055] In addition, the regression model generating device 100 associates the vector of each sample of monomer information as the explanatory variable value calculated in step S104 with the breaking stress for that sample, and constructs a regression model #2 for each type that predicts the breaking stress from the vector of monomer information.
[0056] Furthermore, the regression model generating device 100 associates the vector of each sample of monomer information as the explanatory variable value calculated in step S104 with the elongation for that sample, and constructs a regression model #3 for each type that predicts the elongation from the vector of monomer information.
[0057] In one embodiment, the regression models #1, #2, and #3 for each type may be constructed using Gaussian process regression, as described above.
[0058] In one embodiment, the regression models #1, #2, and #3 for each type may use the type of catalyst as an explanatory variable in addition to the monomer information. Specifically, the type of catalyst may be quantified by any appropriate method such as binary notation and added to the explanatory variables of the regression models #1, #2, and #3 for each type.
[0059] After constructing the regression models #1, #2, and #3 for each type, the regression model generating device 100 provides the constructed regression models #1, #2, and #3 for each type to the property prediction device 200.
[0060] In the above-described embodiment, the regression model for each type is constructed according to Gaussian process regression, but the regression model for each type according to the present invention is not limited to this and may be constructed according to any other appropriate regression analysis method such as multiple regression analysis.
[0061] [Physical property prediction device] Next, a physical property prediction device 200 according to an embodiment of the present invention will be described with reference to Figs. 6 to 8. The physical property prediction device 200 uses the regression models for each type generated by the regression model generation device 100, calculates predicted values for multiple physical property types of a polymer from monomer information as test data, and evaluates the overall physical properties of the polymer based on the calculated predicted values of multiple physical properties. Fig. 6 is a block diagram showing the functional configuration of the physical property prediction device 200 according to an embodiment of the present invention.
[0062] As shown in FIG. 6, the property prediction apparatus 200 includes a monomer information generation unit 210, a property prediction unit 220, and a monomer determination unit 230.
[0063] The monomer information generating unit 210 generates monomer information. Specifically, the monomer information generating unit 210 comprehensively generates combinations of monomers and the content ratios of each monomer as test data for the monomer information, and passes the generated monomer information to the property predicting unit 220.
[0064] For example, when the polymer is polyurethane, the monomer information generating unit 210 selects a predetermined number of monomers from a plurality of candidate monomers, and further sets the molar ratio of each monomer in the combination of the selected monomers. Here, the monomer information may be expressed by setting the molar ratio of the unselected monomers to 0, or the monomer information may be expressed only by the molar ratio of the selected monomers.
[0065] The molar ratio may be comprehensively selected according to restrictive conditions. For example, the molar ratio may be comprehensively set in increments of 0.01 so that the sum of the molar ratios of each diol and diisocyanate is 1.0. However, the molar ratio may be selected so that (the total number of moles of OH groups in the diol) / (the total number of moles of NCO groups in the diisocyanate)=1.1 to 1.2. However, the present invention is not limited to such restrictive conditions, and appropriate restrictive conditions may be set according to the embodiment.
[0066] Then, the monomer information generating unit 210 calculates the ECFP of the selected monomer, multiplies the vector calculated by the ECFP by the content ratio, and passes the calculated product vector to the property predicting unit 220 as test data.
[0067] The physical property prediction unit 220 inputs monomer information indicating a plurality of monomers and their content ratios into a model for each physical property type, and obtains predicted values for a plurality of physical property types. Specifically, the physical property prediction unit 220 inputs vectors representing the monomer information generated as test data by the monomer information generation unit 210 into the regression model for each type as explanatory variable values, and obtains corresponding predicted physical property values from each regression model for each type as objective variable values. For example, when three regression models for each type corresponding to three objective variables of reduced transmittance, breaking stress, and elongation are available, the physical property prediction unit 220 obtains predicted values of reduced transmittance, breaking stress, and elongation for each vector.
[0068] In one embodiment, the physical property prediction unit 220 uses an acquisition function for Bayesian optimization. Specifically, the physical property prediction unit 220 may input the predicted value of each physical property type to the acquisition function and determine the quality of the physical property based on the output value of the acquisition function. In this embodiment, the acquisition function a may be of the UCB (Upper Confidence Bound) type, a(m,s)=m+2×s may be used. Here, m is the predicted value, and s is the standard deviation of each sample of the test data that indicates the uncertainty of the prediction. Gaussian process regression may be used to obtain the predicted value m and the uncertainty s. Gaussianprocessregressor from scikit-learn may be used to execute Gaussian process regression. There is no particular limitation on the kernel, but the product of ConstantKernel and Matern plus WhiteKernel may be used. When Gaussian process regression is used, a predicted distribution is output for each sample. The mean and standard deviation of the predicted distribution may be referred to as the predicted value m and the uncertainty s, respectively. The prediction uncertainty s is used when calculating the acquisition function a of each physical property.
[0069] The acquisition function a is not limited to the above, and may be of EI (Expected Improvement) type or PI (Probability of Improvement) type.
[0070] For example, as shown in Fig. 7(a), when the test data 1 has diols 1, 2, and 3 as monomers and the content ratio thereof is 0.41:0.01:0.11, the physical property prediction unit 220 uses a regression model to obtain a predicted value of the reduced transmittance of 0.41, a predicted value of the breaking stress of 33, and a predicted value of the elongation of 500, as shown in Fig. 7(b). Furthermore, the physical property prediction unit 220 inputs each predicted value into the acquisition function a, and obtains an output value of the reduced transmittance of 0.49, an output value of the breaking stress of 33.6, and an output value of the elongation of 504.
[0071] Similarly, when the test data 2 has diols 1, 2, and 3 as monomers and their content ratio is 0.42:0.01:0.10, the physical property prediction unit 220 uses a regression model to obtain a predicted value of the reduced transmittance of 0.42, a predicted value of the breaking stress of 21, and a predicted value of the elongation of 501, as shown in Fig. 7(b). Furthermore, the physical property prediction unit 220 inputs each predicted value to the acquisition function a, and obtains an output value of the reduced transmittance of 0.48, an output value of the breaking stress of 21.4, and an output value of the elongation of 507.
[0072] The advantage of using such an acquisition function is that it is easier to reach the global optimum solution. It is known that in the search for the optimum solution based on the predicted value, it is easy to recommend data around the data with good performance in the past data (teaching data) as candidates, and there is a possibility of falling into a local optimum solution. If the truly best performing data is different from the known teaching data, the search for the optimum solution based on the predicted value will be trapped in a local optimum solution, making it difficult to reach the global optimum solution. On the other hand, by using an acquisition function, it is possible to prevent being trapped in a local optimum solution. The acquisition function takes into account not only the predicted value m but also the uncertainty s of the prediction. The uncertainty s of the prediction has the tendency to become larger the more different the features of the data are from the teaching data. Therefore, by recommending as candidates those with large predicted values m and prediction uncertainty s, it is possible to recommend "things that look good and have not been explored", and it is possible to efficiently find the optimal one from various variations, making it easier to reach the global optimum solution.
[0073] The property prediction unit 220 passes the predicted value and / or the acquisition function value of each property type calculated for each test data to the monomer determination unit 230.
[0074] The monomer determination unit 230 calculates a deviation value for the content ratio from the predicted values of a plurality of physical property types, and determines the content ratio of the monomers used in synthesizing the polymer to be synthesized based on the deviation value. Specifically, the monomer determination unit 230 first indexes the predicted values and / or acquisition function values of each physical property type acquired from the physical property prediction unit 220 using a deviation value in order to evaluate the overall physical properties of the polymer. For example, when indexing the predicted physical property values using the standard deviation of the acquisition function values, the monomer determination unit 230
[0075]
number
[0076] For example, for the above-mentioned test data 1 and 2, the monomer determination section 230 can calculate indices with standard deviations of 0.64 and 0.71, respectively, as shown in FIG.
[0077] The deviation value is not limited to being calculated based on the acquisition function value, but may be calculated from a predicted value.
[0078] When the deviation value for evaluating the overall physical properties of the polymer is obtained in this manner, the monomer determination unit 230 determines the monomers and their content ratios for producing the polymer based on the obtained deviation value. For example, when it is possible to evaluate that the physical properties are better as the various property values used to calculate the deviation value are larger, the monomer determination unit 230 may determine that the polymer produced based on the monomers and their content ratios with the largest deviation value has the best physical properties. On the other hand, when it is possible to evaluate that the physical properties are better as the various property values used to calculate the deviation value are smaller, the monomer determination unit 230 may determine that the polymer synthesized based on the monomers and their content ratios with the smallest deviation value has the best physical properties.
[0079] The monomer determination unit 230 outputs the monomer with the maximum or minimum deviation value and its content ratio, or the index of the corresponding test data (see, for example, FIG. 9).
[0080] Here, FIG. 9(a) shows the diol and its content ratio and catalyst with the maximum deviation value. Here, diol 4 is 2-methyl-2,4-pentanediol, and catalyst 1 is a silica-supported molybdenum catalyst. Also, FIG. 9(b) shows the predicted values, prediction uncertainty, and acquisition function of the reduced transmittance, breaking stress, and elongation of a polyurethane cured material obtained by curing a polyurethane precursor synthesized based on the diol and its content ratio and catalyst in FIG. 9(a). Furthermore, FIG. 9(c) shows the deviation value calculated based on the acquisition function value in FIG. 9(b).
[0081] FIG. 10 shows the measured values of the reduced transmittance, breaking stress, and elongation of a cured polyurethane product obtained by curing a polyurethane precursor synthesized based on the diols and their contents and catalysts shown in FIG. 9(a).
[0082] Here, the polyurethane precursor was synthesized as follows using dicyclohexylmethane 4,4'-diisocyanate as the diisocyanate.
[0083] In a 100 mL three-neck flask equipped with a Dimroth condenser, a thermometer, and a stirring blade, placed in an oil bath, the monomers (diol 1, diol 2, diol 4) in the content ratio shown in FIG. 9(a) and butyl acetate weighed out so that the final concentration of the entire reaction liquid was 20 mass % were charged and stirred to dissolve. At this time, it was confirmed that the above monomers had dissolved and become a homogeneous solution. Next, dicyclohexylmethane 4,4'-diisocyanate was weighed and added so that the molar ratio of [OH] / [NCO]=1.2, and the mixture was heated with continued stirring until the liquid temperature reached 90°C.
[0084] When the liquid temperature was stabilized at 90°C, 5% by mass of catalyst 1 was added to the substrate (diol and diisocyanate) to initiate the reaction. Every hour, 0.1 mL of the reaction liquid in the system was collected through a Pasteur pipette, filtered through a syringe filter with a pore size of 0.25 μm, and then analyzed by a liquid chromatograph set to GPC (gel permeation chromatogram) mode. Here, the system was heated until the peak derived from the diol disappeared and the elution time of the polyurethane peak top position became constant. After confirming the end point of the reaction by the above operation, the reaction liquid was cooled to room temperature. Next, the catalyst was removed through a membrane filter with a pore size of 50 μm, and then the polyurethane was precipitated by pouring into 300 mL of n-hexane. Next, the precipitate was washed with 50 mL of n-hexane, and the solvent was removed in a vacuum drying device to obtain a polyurethane precursor.
[0085] Catalyst 1 (silica-supported molybdenum catalyst) was prepared from ammonium molybdate (manufactured by Nippon Inorganic Chemical Industry Co., Ltd.) and silica gel (manufactured by Fuji Silysia Chemical Co., Ltd.).
[0086] The reduced transmittance, breaking stress and elongation of the cured polyurethane were measured as follows.
[0087] (Converted refractive index of cured polyurethane) 2-2.5g of polyurethane precursor was placed in a 50mL glass screw-cap bottle, and then about 6g of ethyl acetate was added and dissolved by a rotor-type stirring device. Next, in accordance with JISK1557-6:2009, an isocyanurate-type hexamethylene diisocyanate trimer was added as a 50% by mass ethyl acetate solution so that the [OH] amount [mol] was the same as the [NCO] amount [mol] calculated from the hydroxyl value of the polyurethane precursor that was previously measured, and mixed for about 1 hour. The solution obtained here was mixed with TPX TM The polyurethane cured plate was obtained by spreading it in a resin petri dish, air-drying it for 15 hours, and then heating it in a thermostatic chamber maintained at 90°C for 10 hours. The polyurethane cured plate obtained was cut into a shape of 10 mm x 20 mm using a super straight cutter (manufactured by Dumbbell) to obtain a test specimen.
[0088] The refractive index of the cured polyurethane was measured using an Abbe refractometer (Atago, DR-A1-Plus) with 1-bromonaphthalene as an intermediate liquid.
[0089] After measuring the refractive index of the cured polyurethane, 4×(refractive index) / [(refractive index)+1] 2 (Refractive index ≧ 1) The converted transmittance of the cured polyurethane product was calculated based on the above.
[0090] (Stress at break and elongation of cured polyurethane) After obtaining a polyurethane cured product plate in the same manner as above, a dumbbell-shaped No. 7 test piece according to JIS K 6251:2010 was cut using a super straight cutter (manufactured by Dumbbell), and the breaking stress and elongation of the polyurethane cured product were measured using a tabletop tensile compression tester (manufactured by A&D, MCT-2150).
[0091] Fig. 11 shows the diols and their content ratios and catalysts of Comparative Examples (1) to (10). Here, diol 5 is 2-butyl-2-ethyl-1,3-propanediol, catalyst 2 is dibutyltin dilaurate, and OH [mmol] means the number of moles of OH groups in each monomer. Fig. 12 shows the measured values of reduced transmittance, breaking stress, and elongation of the cured polyurethane synthesized based on the diols and their content ratios of Fig. 11.
[0092] Here, the polyurethane precursors of Comparative Examples (1) to (9) were synthesized in the same manner as described above, except that dicyclohexylmethane 4,4'-diisocyanate was used as the diisocyanate and catalyst 2 was used as the catalyst.
[0093] The polyurethane precursor of Comparative Example (10) was synthesized in the same manner as described above, except that dicyclohexylmethane 4,4'-diisocyanate was used as the diisocyanate.
[0094] Furthermore, in [Formula 1], x ij Instead of (acquisition function value), actual measured values were used to calculate the index of actual measured values.
[0095] [Property prediction processing] Next, a property prediction process according to an embodiment of the present invention will be described with reference to Fig. 13. The property prediction process may be executed by, for example, the above-described property prediction device 200, and more specifically, may be realized by a processor of the property prediction device 200. Fig. 13 is a flowchart showing the property prediction process according to an embodiment of the present invention.
[0096] As shown in FIG. 13, in step S201, the property prediction device 200 obtains explanatory variable values to be input to the regression model for each type. Specifically, the property prediction device 200 generates monomer information as test data to be input to the regression model for each type provided by the regression model generation device 100. For example, the property prediction device 200 selects a predetermined number of monomers from a plurality of monomers, and further comprehensively sets various molar ratios for the selected monomers. For each test data generated in this way, the property prediction device 200 calculates the ECFP of each monomer, multiplies the calculated ECFP vector by the molar ratio, and calculates a product vector.
[0097] In step S202, the property prediction device 200 uses a regression model for each type to calculate a predicted value of each property as a response variable from the monomer information. Specifically, the property prediction device 200 inputs the product vector acquired in step S201 into the regression model for each type and calculates a corresponding predicted value of the property. At this time, the property prediction device 200 may input the calculated predicted value of the property into a predetermined acquisition function and calculate an acquisition function value for the predicted value of each property type. For example, when the acquisition function is of UCB type, the uncertainty s may be set as the standard deviation of each sample of the test data.
[0098] In step S203, the property prediction device 200 calculates a deviation value as an index for comprehensively evaluating the polymer from the predicted value and / or the acquisition function value for each property type. Specifically, when the predicted property value is indexed by the standard deviation of the acquisition function value, the property prediction device 200 calculates a deviation value as an index for comprehensively evaluating the polymer from the predicted property value and / or the acquisition function value.
[0099]
number
[0100] In step S204, the physical property prediction device 200 determines the monomers and their content ratios used in the synthesis of the polymer based on the deviation value. Specifically, the physical property prediction device 200 may output the monomers and their content ratios that maximize or minimize the deviation value, or an index of the test data.
[0101] In the above-described examples, attention is focused on the diol as a monomer of the polyurethane precursor, and the present invention is applied to a case where only the diol is changed. However, the present invention is not limited to this, and can also be applied to a case where attention is focused on the diisocyanate as a monomer of the polyurethane precursor, and only the diisocyanate is changed, or a case where both the diol and the diisocyanate are changed.
[0102] In addition, in the above-described embodiment, the present invention is applied to a case where the physical property type of a cured thermosetting polyurethane is used as the physical property type of the polymer, but the present invention is not limited to this, and can be applied, for example, to a case where the physical property type of not only a thermosetting resin but also a non-thermosetting resin such as a thermoplastic resin is used.
[0103] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the specific embodiments described above, and various modifications and variations are possible within the scope of the gist of the present invention as defined in the claims. [Explanation of symbols]
[0104] 50 Substance Database 100 Regression model generator 110 Training Data Acquisition Unit 120 Regression model generation unit 200 Physical Properties Prediction Device 210 Monomer Information Generation Unit 220 Physical Properties Prediction Department 230 Monomer Determinant
Claims
1. a property prediction unit that obtains predicted values of a plurality of property types of a polymer by a model for each property type based on monomer information indicating a plurality of monomers and their content ratios; a monomer determination unit that calculates a deviation value for the monomer information from the predicted values of the plurality of physical property types, and determines the monomers to be used in synthesizing the polymer and their content ratios based on the deviation value; A physical property prediction device having the above structure.
2. 2. The physical property prediction device according to claim 1, wherein the model for each physical property type is derived by Gaussian process regression using training data consisting of pairs of the monomer information and physical property values of the physical property type of the polymer corresponding to the monomer information.
3. 3. The property prediction apparatus according to claim 1, wherein the monomer determination unit calculates output values of an acquisition function of Bayesian optimization from the predicted values of the plurality of property types, and calculates a deviation value of the output values.
4. 4. The physical property prediction device according to claim 1, wherein the monomer determination unit determines the monomer information for which the deviation value is maximum or minimum as the monomers used in synthesis of the polymer and their content ratios.
5. Further comprising a monomer information generating unit that generates the monomer information, 5. The physical property prediction apparatus according to claim 1, wherein the monomer information is expressed by a product of an Extended Connectivity Fingerprint (ECFP) of each monomer and a molar ratio.
6. A step in which one or more processors obtain predicted values of a plurality of physical property types of a polymer using a model for each physical property type based on monomer information indicating a plurality of monomers and their content ratios; The one or more processors calculate deviation values for the monomer information from the predicted values of the plurality of physical property types, and determine the monomers and their content ratios used in synthesizing the polymer based on the deviation values; The physical property prediction method has the following features:
7. 7. The method for predicting physical properties according to claim 6, wherein the model for each physical property type is derived by Gaussian process regression using training data consisting of pairs of the monomer information and physical property values of the physical property type of the polymer corresponding to the monomer information.
8. 8. The method for predicting physical properties according to claim 6, wherein the determining step calculates output values of an acquisition function of Bayesian optimization from the predicted values of the plurality of physical property types, and calculates a deviation value of the output values.
9. 9. The method for predicting physical properties according to claim 6, wherein the determining step determines the monomer information for which the deviation value is maximum or minimum as the monomers and their content ratios used in the synthesis of the polymer.
10. generating said monomer information; 10. The method for predicting physical properties according to claim 6, wherein the monomer information is expressed by a product of an Extended Connectivity Fingerprint (ECFP) of each monomer and a molar ratio.
Citation Information
Patent Citations
Information processing apparatus, method and program
JP2019082790A
Design support apparatus and design support method
JP2019086817A
Method and device for predicting physical property data
JP2020038493A
Polymer design device, program, and method
JP2020071827A
JPP6633820B
