Parameter generation device, system, method and program

Abstract

An input means 81 receives an input of a constraint and a first objective function defining a combination of factors relating to the production of a material. An objective function generation means 82 generates a second objective function that defines a stochastic fluctuation for a parameter of the first objective function. An optimisation processing means 83 optimises a model containing the second objective function and the constraint. The output means 84 sets the values of variables of the second objective function obtained by the optimisation as a parameter set and outputs the same.

Description

Parameter generation device, system, method and program

[0001] The present invention relates to a parameter generation device, a parameter generation system, a parameter generation method, and a parameter generation program for generating desired parameters.

[0002] In research to discover new materials that will produce the desired performance, researchers try out a huge number of combinations of factors such as the type and amount of material, processing temperature, pressure, time, etc. However, the number of combinations is astronomical, making it impossible to try every possible pattern.

[0003] Generally, knowledge is accumulated from past experience and simulation results about combinations of conditions that are expected to produce good results and combinations of conditions that are expected to produce the opposite results. Therefore, in research sites, new combinations of elements are determined from within the range that satisfies the conditions indicated by this accumulated knowledge, and the results are repeatedly verified through prototyping, simulations, etc.

[0004] Furthermore, in order to derive a combination of elements that satisfies desired conditions, a mathematical programming solver may be used to derive the optimal combination based on an objective function designed by an engineer or the like and constraints that define the conditions to be satisfied (i.e., a mathematical optimization problem).

[0005] Patent Document 1 describes a design support system that reduces the number of numerical simulations required to consider design parameters for achieving a design goal. The design support system described in Patent Document 1 performs a direct analysis using initial settings for the design parameters, and then performs an inverse analysis based on the analysis results, which is an adjoint numerical analysis, thereby performing a sensitivity analysis for the design goal.

[0006] Re-toku WO2007 / 122677

[0007] When considering a method for manufacturing a desired new material, it is necessary to create a combination of elements for manufacturing the new material. One method for doing this is to create the combination based on the experience and intuition of an expert. However, this method is highly dependent on the individual, and there is a problem that the work cannot be carried out efficiently without a specific expert.

[0008] Another method for creating combinations of elements is to randomly select combinations of elements that satisfy certain conditions. However, as the conditions become more complex, the probability of generating a combination of elements that satisfies the conditions decreases, making this method inefficient.

[0009] On the other hand, it is possible to derive an optimal solution to a designed mathematical optimization problem by using a mathematical programming solver. However, the resulting optimal solution is unique to each designed objective function. Normally, when searching for new materials, a certain number of combinations of elements are required as candidate candidates for manufacturing the material. Therefore, it can be said to be inefficient for engineers to design a mathematical optimization problem every time a combination of elements is derived.

[0010] Furthermore, the method described in Patent Document 1 aims to reduce the number of simulations when considering design parameters for achieving a design goal, and is not intended to derive combinations of multiple elements.

[0011] Furthermore, the objective function designed by engineers and others may not always be accurate. In such cases, the obtained optimal solution may not necessarily be the solution for producing the desired material. Therefore, it is desirable to be able to efficiently find multiple methods for producing the desired material.

[0012] Therefore, an object of the present invention is to provide a parameter generation device, a parameter generation system, a parameter generation method, and a parameter generation program that are capable of discovering multiple methods for producing a desired material.

[0013] The parameter generation device according to the present invention is characterized by comprising: an input means for receiving input of a first objective function and constraint conditions that define a combination of elements related to material production; an objective function generation means for generating a second objective function in which stochastic fluctuations are set for the parameters of the first objective function; an optimization processing means for optimizing a model including the second objective function and the constraint conditions; and an output means for outputting the values ​​of the variables of the second objective function obtained by the optimization as a parameter set.

[0014] The parameter generation system according to the present invention includes a prediction model generation device that uses past experimental data as training data to train a prediction model in which materials are used as explanatory variables and characteristic values ​​indicating the properties of the material are used as objective variables; a first objective function generation device that uses the prediction model to generate a first objective function that defines a combination of elements related to the production of the material; and a parameter generation device that generates a parameter set using the first objective function, wherein the first objective function generation device generates a first objective function that includes a linear sum of characteristic values ​​indicated by the objective variables as a combination of elements and inputs the first objective function to the parameter generation device, and the parameter generation device includes input means for accepting input of the first objective function and constraint conditions, objective function generation means for generating a second objective function in which stochastic fluctuations are set for the parameters of the first objective function, optimization processing means for optimizing a model including the second objective function and the constraint conditions, and output means for outputting the values ​​of the variables of the second objective function obtained by the optimization as a parameter set.

[0015] The parameter generation method according to the present invention is characterized in that a computer receives input of a first objective function and constraint conditions that define a combination of elements related to the production of a material, generates a second objective function in which stochastic fluctuations are set for the parameters of the first objective function, optimizes a model including the second objective function and the constraint conditions, and outputs the values ​​of the variables of the second objective function obtained by the optimization as a parameter set.

[0016] The parameter generation program according to the present invention is characterized in that it causes a computer to execute an input process for accepting input of a first objective function and constraint conditions that define a combination of elements related to material production; an objective function generation process for generating a second objective function in which stochastic fluctuations are set for the parameters of the first objective function; an optimization process for optimizing a model including the second objective function and the constraint conditions; and an output process for outputting the values ​​of the variables of the second objective function obtained by the optimization as a parameter set.

[0017] The present invention allows for the discovery of multiple ways to produce desired materials.

[0018] It is a block diagram showing an example of the configuration of an embodiment of a simulation system of the present invention.It is an explanatory diagram showing an example of the operation of a parameter generation device.It is a block diagram showing an overview of a parameter generation device according to the present invention.It is a block diagram showing an overview of a simulation system according to the present invention.It is a schematic block diagram showing the configuration of a computer according to at least one embodiment.

[0019] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0020] 1 is a block diagram showing an example of the configuration of an embodiment of a simulation system according to the present invention. The simulation system 100 of this embodiment includes a predictive model generation device 10, a first objective function generation device 20, a parameter generation device 30, an optimization processing device 40, and a simulator 50.

[0021] The predictive model generation device 10 is a device that generates a predictive model that predicts the effect of the type and amount of raw materials on the characteristics of a product (e.g., material) from past experimental data. Specifically, the predictive model generation device 10 learns a predictive model that predicts values ​​indicating the characteristics of a material (hereinafter referred to as characteristic values) based on past experimental data. Note that characteristic values ​​can also be called performance indicators.

[0022] The prediction model generating device 10 includes a storage unit 11 , a learning unit 12 , and a model output unit 13 .

[0023] The storage unit 11 stores training data used for learning by the learning unit 12. The training data is, for example, data that associates multiple materials used in manufacturing a material with characteristic values ​​that indicate the hardness, toughness, heat resistance, and other material characteristics when using those materials. The storage unit 11 is realized, for example, by a magnetic disk or the like.

[0024] The learning unit 12 uses past experimental data as training data to learn a prediction model in which the materials are used as explanatory variables and the characteristic values ​​are used as target variables. Note that the method by which the learning unit 12 learns the prediction model is arbitrary, and any method such as machine learning may be used.

[0025] The model output unit 13 outputs the prediction model generated by the learning unit 12. The model output unit 13 may input the prediction model to the first objective function generation device 20.

[0026] The first objective function generator 20 generates an objective function that defines a combination of elements related to the production of a material to obtain a target characteristic value. Here, the elements related to the production of a material refer to the details that must be specified in the material production method, specifically, the type and amount of raw material, as well as the processing temperature, pressure, processing time, etc. The objective function is defined using parameters such as weights (coefficients) and biases that are set for each element.

[0027] That is, the first objective function generating device 20 generates an objective function (hereinafter referred to as the first objective function) used to derive the optimal combination of the type and amount of material, processing method, etc. that will achieve the target value, using elements related to the production of the material and the parameters described above.

[0028] The first objective function generation device 20 may generate the first objective function using the prediction model generated by the learning unit 12. For example, ｉ The prediction model predicts j elements x as an example of the combination of elements as shown in the following formula 1. ｊ Here, x bar (superscript bar of x) is the mean value of the element, and σ is the standard deviation of the element, which is a value calculated when creating a prediction model.

[0029]

[0030] At this time, the first objective function generating device 20 calculates each characteristic value y ｉ For example, a first objective function may be generated that includes a linear sum of each characteristic value y ｉ The weight for ｉ In this case, the first objective function generating device 20 may generate a first objective function such as the following formula 2. Formula 2 is a linear sum of the squares of the differences of the characteristic values ​​from the target median value. ｉ is the characteristic value y ｉ The target median of the weight W ｉ is determined by engineers, etc. ｉ The designation may be accepted by the first objective function generation device 20 or by the parameter generation device 30 described later.

[0031]

[0032] Furthermore, as exemplified by the following Equation 3, the first objective function may be expressed in a form obtained by expanding Equation 2 above.

[0033] In the examples of the above formulas 1 to 3, for example, a ｉｊ , b ｉ , Lmed ｉ , W ｉ , Q ｉｊ , L ｉ The above-mentioned parameters are as follows. That is, the parameters shown in this embodiment include not only parameters at the time of formulation, but also parameters obtained during formulation.

[0034] In the above description, the first objective function is configured as a linear sum of squares of differences of objective variables (characteristic values) from a target median value. However, the content included in the first objective function is not limited to characteristic values. The first objective function may also include elements other than characteristic values ​​(e.g., a processing method). The first objective function generation device 20 inputs the generated first objective function to the parameter generation device 30.

[0035] In addition, in the present embodiment, the first objective function generation device 20 is realized as an independent device. However, the first objective function generation device 20 may be realized integrally with another device, for example, may be included in the parameter generation device 30.

[0036] The parameter generating device 30 is a device that generates parameters to be input to the simulator 50, and is connected to the optimization processing device 40 and the simulator 50. The simulator 50 is a device that performs trials based on the generated parameters. The simulator 50 may take any form, and any known device may be used.

[0037] The optimization processing device 40 is a device that executes optimization processing based on the model generated by the parameter generation device 30. The optimization processing device 40 may be realized by a (classical) computer that executes a mathematical programming solver. The optimization processing device 40 may also be a device dedicated to finding the ground state of the Hamiltonian of the Ising model. In this case, the optimization processing device 40 is realized as, for example, a device that executes annealing based on the Ising model generated by the parameter generation device 30.

[0038] The parameter generating device 30 includes an input unit 31, an objective function generating unit 32, an optimization processing unit 33, and an output unit 34.

[0039] The input unit 31 accepts input of the above-described first objective function. The input unit 31 also accepts input of constraint conditions indicating constraints that each element must satisfy and constraints when combining elements. The input unit 31 may accept input of the first objective function generated by the first objective function generation device 20, or may accept input of a first objective function generated by another device (not shown) or manually by an engineer or the like.

[0040] For example, constraints for producing new materials include specifications regarding the type of material to be selected (one from each material group, etc.), specifications regarding the distribution of material quantities (specifying the sum of the quantities of several materials, specifying the quantities of individual materials, etc.), specification of exclusive materials, etc. Other constraints for producing new materials include specifications regarding material processing (restrictions on processing temperature (such as upper limit temperature) and pressure depending on the material, etc.).

[0041] The objective function generator 32 generates an objective function (hereinafter referred to as a second objective function) in which stochastic fluctuations are set for the parameters of the input first objective function. Here, setting fluctuations for parameters means performing arithmetic operations such as addition, subtraction, multiplication, and division on the values ​​indicated by the fluctuations for the parameters. In addition, the targets for setting fluctuations include parameters that appear in the final first objective function (for example, Q in the above formula 3). ｉｊ Ya, L ｉ ), as well as parameters during formulation (for example, a ｉｊ Ya, Lmed ｉ ) is also included.

[0042] The parameters for which fluctuations are set are designated in advance. The designation method is arbitrary, and for example, the input unit 31 may receive an input specifying the parameters for which fluctuations are set from an engineer or the like. The fluctuations are set for the parameters of the objective function, and are not set for the constraint conditions.

[0043] Specifically, the objective function generator 32 sets fluctuations represented by random variables that follow a predetermined probability distribution for the parameters of the first objective function. Preferably, the objective function generator 32 sets fluctuations represented by random variables that follow a probability distribution whose average is zero for the parameters of the first objective function. Examples of the probability distribution whose average is zero include a normal distribution as exemplified by the following formula 4 and a uniform distribution as exemplified by the following formula 5.

[0044]

[0045] Here, to make the mean of the probability distribution zero, μ = 0 in the normal distribution of Equation 4, and a = -b (b > 0) in the uniform distribution of Equation 5. In the case of a normal distribution, the standard deviation σ is an index representing the magnitude of fluctuation. In the case of a uniform distribution, the interval width b - a is an index representing the magnitude of fluctuation. In other words, the larger this parameter is, the larger the fluctuation, and conversely, the smaller this parameter is, the smaller the fluctuation. It can also be said that the similarity to the original objective function (optimization problem) changes depending on the magnitude of the fluctuation applied.

[0046] Below, we will explain a method for setting, for the parameters of the first objective function, fluctuations represented by random variables that follow the probability distribution shown in the above-exemplified formula 4 or formula 5, using the above-exemplified formulas 1 to 3. The fluctuations set in this embodiment are expressed by an equation for a random variable x that follows the fluctuation probability distribution p(x).

[0047] For example, the probability distribution of fluctuations p(X ｉｊ ) is expressed by the normal distribution shown in the above formula 4. In this case, the fluctuation X ｉｊ The second objective function in which the above is set is expressed by the following formula 6. As shown in formula 6, the standard deviation σ is determined by, for example, the parameter a ｉｊ is set to a constant c times

[0048]

[0049] In Equation 6, the index representing the magnitude of fluctuation is p(X ｉｊ ) is the standard deviation σ of the X ｉｊ tends to become larger).

[0050] Similarly, for the above-mentioned formula 2, the fluctuation X ｉ The second objective function in which the above is set is expressed by the following formula 7. As shown in formula 7, the index representing the magnitude of fluctuation is, for example, the parameter Lmed ｉ is set to a constant c times

[0051]

[0052] Furthermore, for the above-mentioned formula 3, the fluctuation X ｉｊ and fluctuation X ｉ The second objective function in which the above is set is expressed by the following formula 8. As shown in formula 8, the index representing the magnitude of fluctuation is, for example, the parameter Q ｉｊ and L ｉ is set to a constant c times the standard deviation of the non-zero

[0053]

[0054] The above is an example of the fluctuation formula when the probability distribution is a normal distribution. The same applies when the probability distribution is a uniform distribution. For example, in the case of the above formula 1, the probability distribution is expressed by the following formula 9.

[0055]

[0056] In this way, the objective function generator 32 generates a second objective function in which stochastic fluctuations are set for the parameters of the first objective function. Furthermore, the objective function generator 32 may output the generated second objective function (i.e., the objective function in which fluctuations are set) and accept modifications by an engineer or the like.

[0057] For example, in the above-exemplified formula 1, the parameter a ｉｊ In this case, the objective function generator 32 sets fluctuations for the parameter a ｉｊ After setting fluctuations for the second objective function, the engineer outputs the generated second objective function. ｉ After determining W, the objective function generator 32 ｉ The input is accepted as the correction content, and the accepted W ｉ The objective function H Ｏ At this time, the objective function generator 32 generates the determined W ｉ Instead of W ｉ The objective function H Ｏ The input may be accepted as the correction content.

[0058] In addition, for example, in the above-mentioned formula 2, the parameter Lmed ｉIn this case, the objective function generator 32 sets fluctuations for the parameter Lmed ｉ After setting fluctuations for the second objective function, the engineer outputs the generated second objective function. ｉ After determining W, the objective function generator 32 ｉ The input is accepted as the correction content, and the accepted W ｉ The objective function H Ｏ Similarly to the above, the objective function generator 32 generates the determined W ｉ Instead of W ｉ The objective function H Ｏ The input may be accepted as the correction content.

[0059] In this way, by outputting the generated second objective function and accepting modifications by engineers, etc., the objective function after the fluctuations are set by engineers, etc. is verified, making it possible to generate a more preferable objective function (mathematical programming problem). In the following description, the objective function modified by engineers, etc. will also be referred to as the second objective function.

[0060] The optimization processing unit 33 optimizes a model including the second objective function and constraint conditions generated by the objective function generating unit 32. Specifically, the optimization processing unit 33 transmits the model to be optimized to the optimization processing device 40, causes the optimization processing to be performed, and receives the execution results.

[0061] Specifically, first, the optimization processing unit 33 generates a model of the optimization target from the second objective function and the constraint conditions in accordance with the optimization processing device 40. For example, as described above, it is assumed that the optimization processing device 40 is realized by a computer that executes a mathematical programming solver. In this case, the optimization processing unit 33 generates a mathematical optimization problem including the second objective function and the constraint conditions as a model of the optimization target, and causes the computer to execute the generated model.

[0062] Furthermore, for example, as described above, it is assumed that the optimization processing device 40 is realized by a device that performs annealing (annealing machine). In this case, the optimization processing unit 33 generates an Ising model to be optimized based on the second objective function and constraint conditions. Note that, since methods for generating an Ising model from an objective function and constraint conditions are widely known, detailed description thereof will be omitted here.

[0063] The output unit 34 outputs the variable values ​​of the second objective function obtained by optimization as a parameter set. The variable values ​​here are, specifically, information indicating the specific values ​​and settings of each element (e.g., type and amount of material, processing temperature, pressure, time, etc.). The output unit 34 may output the parameter set directly to the simulator 50, or may output it in a file format (e.g., CSV (Comma Separated Value) format). Furthermore, when the optimization processing device 40 is an annealing machine, the optimization results are obtained as binary variables, and therefore the output unit 34 may output a parameter set obtained by converting the optimization results.

[0064] The input unit 31, the objective function generation unit 32, the optimization processing unit 33, and the output unit 34 are realized by a processor (for example, a CPU (Central Processing Unit)) of a computer that operates according to a program (parameter generation program).

[0065] For example, the program may be stored in a storage unit (not shown) of the parameter generating device 30, and the processor may read the program and operate as the input unit 31, the objective function generating unit 32, the optimization processing unit 33, and the output unit 34 in accordance with the program. Furthermore, the functions of the parameter generating device 30 may be provided in a software as a service (SaaS) format.

[0066] Furthermore, the input unit 31, the objective function generation unit 32, the optimization processing unit 33, and the output unit 34 may each be realized by dedicated hardware. Furthermore, some or all of the components of each device may be realized by general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits, etc., and a program.

[0067] Furthermore, when some or all of the components of the parameter generating device 30 are realized by a plurality of information processing devices, circuits, etc., the plurality of information processing devices, circuits, etc. may be centrally or decentralized. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, or the like, in a form in which each is connected via a communication network.

[0068] Next, the operation of the parameter generating device 30 of this embodiment will be described.

[0069] The input unit 31 accepts input of a first objective function and constraint conditions (step S11). As described above, the first objective function is a function that defines a combination of elements related to the production of a material. The objective function generation unit 32 generates a second objective function in which stochastic fluctuations are set for the parameters of the first objective function (step S12). The optimization processing unit 33 optimizes a model including the second objective function and constraint conditions (step S13). More specifically, the optimization processing unit 33 causes the optimization processing device 40 to execute the optimization process. Then, the output unit 34 outputs the values ​​of the variables of the second objective function obtained by the optimization as a parameter set (step S14).

[0070] As described above, in this embodiment, the input unit 31 accepts input of a first objective function and constraint conditions, the objective function generator 32 generates a second objective function in which stochastic fluctuations are set for the parameters of the first objective function, the optimization processor 33 optimizes a model including the second objective function and the constraint conditions, and the output unit 34 outputs the values ​​of the variables of the second objective function obtained by the optimization as a parameter set.

[0071] With the above configuration, it is possible to obtain a combination of elements (i.e., a parameter set) that indicates multiple methods for producing a desired material. Then, by performing a simulation based on this parameter set, it is possible to determine whether the desired material has been obtained. As a result, it becomes possible to discover multiple methods for producing the desired material.

[0072] Furthermore, in this embodiment, the optimization processing unit 33 causes a computer that executes a mathematical programming solver to execute the optimization process, so that it becomes possible to quickly find a parameter set that realizes the desired properties.

[0073] Furthermore, in this embodiment, the objective function generator 32 sets fluctuations only for the objective function without changing the constraints, so that a mathematical programming solver can also obtain a variety of parameter sets that satisfy the constraints. In this case, the objective function generator 32 sets the fluctuations to be set for the objective function using a probability distribution that centers on the original model (objective function), so that a parameter set close to the optimal solution for the original model can be obtained.

[0074] Furthermore, in this embodiment, the objective function generation unit 32 sets fluctuations based on a probability distribution, which allows the degree of fluctuation to be continuously changed, thereby making it possible to obtain a variety of parameter sets, from parameter sets that are close to the optimal solution to parameter sets that are relatively far from the optimal solution.

[0075] Next, an overview of the present invention will be described. Figure 3 is a block diagram showing an overview of a parameter generation device according to the present invention. A parameter generation device 80 (e.g., parameter generation device 30) according to the present invention includes input means 81 (e.g., input unit 31) that accepts input of a first objective function that defines a combination of elements related to material production (e.g., type and amount of material, processing temperature, pressure, time, etc.) and constraints (e.g., selection of material type, material quantity allocation, designation of exclusive material, material processing method, etc.), objective function generation means 82 (e.g., objective function generation unit 32) that generates a second objective function in which stochastic fluctuations are set for the parameters of the first objective function, optimization processing means 83 (e.g., optimization processing unit 33) that optimizes a model including the second objective function and the constraints, and output means 84 (e.g., output unit 34) that outputs the values ​​of variables of the second objective function obtained by optimization as a parameter set.

[0076] Such a configuration makes it possible to discover multiple methods for producing a desired material. That is, the above configuration makes it possible to obtain a combination of elements (parameter set) that indicates multiple methods for producing a desired material, and by performing a simulation based on this parameter set, it is possible to determine whether the desired material has been obtained. As a result, it becomes possible to discover multiple methods for producing a desired material.

[0077] Furthermore, the objective function generating means 82 may generate a second objective function in which a fluctuation represented by a random variable following a predetermined probability distribution is set as a parameter.

[0078] Specifically, the objective function generating means 82 may generate a second objective function in which a fluctuation represented by a random variable that follows a probability distribution with an average of zero is set as a parameter. With such a configuration, a parameter set close to the optimal solution for the original model can be obtained.

[0079] Furthermore, the objective function generating means 82 may generate a second objective function in which a fluctuation represented by a random variable following a normal distribution or a uniform distribution is set as a parameter.

[0080] Specifically, the objective function generating means 82 may generate a second objective function in which a fluctuation represented by a random variable following a normal distribution whose standard deviation is a constant multiple of the parameter that gives the fluctuation is set as a parameter.

[0081] Alternatively, the optimization processing means 83 may generate a mathematical optimization problem including the second objective function and the constraints as a model to be optimized, and cause a computer that executes a mathematical programming solver (e.g., the optimization processing device 40) to execute the generated model. Such a configuration makes it possible to quickly find a parameter set that realizes desired properties.

[0082] On the other hand, the optimization processing means 83 may generate an Ising model to be optimized based on the second objective function and the constraints, and cause an annealing machine (for example, the optimization processing device 40) to execute the generated Ising model. Such a configuration makes it possible to obtain parameter sets with different properties from similar objective functions.

[0083] Furthermore, the objective function generating means 82 may output the generated second objective function and accept modifications made to the second objective function by a user. Then, the optimization processing means 83 may optimize a model to be optimized, including the second objective function reflecting the modifications and the constraints. With such a configuration, the objective function after the fluctuations set by an engineer or the like is verified, making it possible to generate a more preferable objective function (mathematical programming problem).

[0084] 4 is a block diagram showing an overview of a simulation system according to the present invention. The simulation system 200 (e.g., simulation system 100) according to the present invention includes a prediction model generation device 60 (e.g., prediction model generation device 10) that uses past experimental data as training data to train a prediction model in which materials are used as explanatory variables and characteristic values ​​indicating the properties of the material are used as objective variables, a first objective function generation device 70 (e.g., first objective function generation device 20) that uses the prediction model to generate a first objective function that defines a combination of elements related to the production of the material, and a parameter generation device 80 (e.g., parameter generation device 30) that generates a parameter set using the first objective function.

[0085] The first objective function generating device 70 generates a first objective function including a linear sum of characteristic values ​​indicated by objective variables as a combination of elements, and inputs the first objective function to the parameter generating device 80 .

[0086] The configuration of the parameter generating device 80 is the same as the parameter generating device 80 illustrated in FIG.

[0087] Even with such a configuration, it is possible to discover multiple ways to produce the desired material.

[0088] 5 is a schematic block diagram showing the configuration of a computer according to at least one embodiment. The computer 1000 includes a processor 1001, a main memory device 1002, an auxiliary memory device 1003, and an interface 1004. The computer 1000 may also be connected to a computer that executes a mathematical programming solver, an annealing machine, a simulator, or the like.

[0089] The parameter generating device 80 described above is implemented in a computer 1000. The operations of the above-described processing units are stored in the form of a program (parameter generating program) in an auxiliary storage device 1003. The processor 1001 reads the program from the auxiliary storage device 1003, loads it into the main storage device 1002, and executes the above-described processing in accordance with the program.

[0090] In at least one embodiment, the auxiliary storage device 1003 is an example of a non-transitory tangible medium. Other examples of non-transitory tangible media include a magnetic disk, a magneto-optical disk, a CD-ROM (Compact Disc Read-only memory), a DVD-ROM (Read-only memory), and a semiconductor memory connected via the interface 1004. In addition, when this program is distributed to the computer 1000 via a communication line, the computer 1000 that receives the program may load the program into the main storage device 1002 and execute the above-described processing.

[0091] The program may be a program for realizing part of the above-described functions. Furthermore, the program may be a so-called differential file (differential program) that realizes the above-described functions in combination with another program already stored in the auxiliary storage device 1003.

[0092] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes.

[0093] (Supplementary Note 1) A parameter generation device comprising: an input means for receiving an input of a first objective function and constraint conditions that define a combination of elements related to the production of a material; an objective function generation means for generating a second objective function in which stochastic fluctuations are set for parameters of the first objective function; an optimization processing means for optimizing a model including the second objective function and the constraint conditions; and an output means for outputting values ​​of variables of the second objective function obtained by the optimization as a parameter set.

[0094] (Supplementary Note 2) The parameter generating device according to Supplementary Note 1, wherein the objective function generating means generates a second objective function having a parameter set to a fluctuation represented by a random variable that follows a predetermined probability distribution.

[0095] (Supplementary Note 3) The parameter generating device according to Supplementary Note 1 or Supplementary Note 2, wherein the objective function generating means generates a second objective function whose parameter is a fluctuation represented by a random variable that follows a probability distribution whose mean is zero.

[0096] (Supplementary Note 4) The parameter generation device according to any one of Supplementary Note 1 to Supplementary Note 3, wherein the objective function generation means generates a second objective function having a parameter set to fluctuations represented by random variables that follow a normal distribution or a uniform distribution.

[0097] (Supplementary Note 5) The parameter generation device according to any one of Supplementary Note 1 to Supplementary Note 4, wherein the objective function generation means generates a second objective function in which a fluctuation represented by a random variable following a normal distribution whose standard deviation is a constant multiple of the parameter that provides the fluctuation is set as a parameter.

[0098] (Supplementary Note 6) The parameter generation device according to any one of Supplementary Note 1 to Supplementary Note 5, wherein the optimization processing means generates a mathematical optimization problem including a second objective function and constraint conditions as a model to be optimized, and causes a computer that executes a mathematical programming solver to execute the generated model.

[0099] (Supplementary Note 7) The parameter generation device according to any one of Supplementary Note 1 to Supplementary Note 5, wherein the optimization processing means generates an Ising model to be optimized based on the second objective function and the constraint conditions, and causes an annealing machine to execute the generated Ising model.

[0100] (Supplementary Note 8) The parameter generation device described in any one of Supplementary Note 1 to Supplementary Note 7, wherein the objective function generation means outputs the generated second objective function and accepts modifications made by a user to the second objective function, and the optimization processing means optimizes a model to be optimized that includes the second objective function reflecting the modifications and the constraint conditions.

[0101] (Supplementary Note 9) A parameter generation system comprising: a prediction model generation device that uses past experimental data as training data to learn a prediction model in which materials are used as explanatory variables and characteristic values ​​indicating material properties are used as objective variables; a first objective function generation device that uses the prediction model to generate a first objective function that specifies a combination of elements related to material production; and a parameter generation device that generates a parameter set using the first objective function, wherein the first objective function generation device generates a first objective function that includes a linear sum of the characteristic values ​​indicated by the objective variables as the combination of the elements, and the parameter generation device comprises: input means that accepts input of the first objective function and constraint conditions; objective function generation means that generates a second objective function in which stochastic fluctuations are set for parameters of the first objective function; optimization processing means that optimizes a model including the second objective function and the constraint conditions; and output means that output values ​​of the variables of the second objective function obtained by the optimization as a parameter set.

[0102] (Supplementary Note 10) A parameter generation method comprising: a computer receiving input of a first objective function and constraint conditions that define a combination of elements related to material production; the computer generating a second objective function in which stochastic fluctuations are set for parameters of the first objective function; the computer optimizing a model including the second objective function and the constraint conditions; and the computer outputting values ​​of variables of the second objective function obtained by optimization as a parameter set.

[0103] (Supplementary Note 11) A program storage medium storing a parameter generation program for causing a computer to execute: an input process for accepting input of a first objective function and constraint conditions that define a combination of elements related to material production; an objective function generation process for generating a second objective function in which stochastic fluctuations are set for parameters of the first objective function; an optimization process for optimizing a model including the second objective function and the constraint conditions; and an output process for outputting values ​​of variables of the second objective function obtained by the optimization as a parameter set.

[0104] (Supplementary Note 12) A parameter generation program for causing a computer to execute: an input process for accepting input of a first objective function and constraint conditions that define a combination of elements related to material production; an objective function generation process for generating a second objective function in which stochastic fluctuations are set for parameters of the first objective function; an optimization process for optimizing a model including the second objective function and the constraint conditions; and an output process for outputting values ​​of variables of the second objective function obtained by the optimization as a parameter set.

[0105] The present invention is preferably applied to a parameter generation device that generates desired parameters. Specifically, the present invention is preferably applied to fields where prototypes and simulations are repeatedly performed in research sites for new material discovery.

[0106] REFERENCE SIGNS LIST 10 Prediction model generation device 11 Storage unit 12 Learning unit 13 Model output unit 20 First objective function generation device 30 Parameter generation device 31 Input unit 32 Objective function generation unit 33 Optimization processing unit 34 Output unit 40 Optimization processing device 50 Simulator

Claims

1. an input means for receiving an input of a first objective function and constraint conditions that define a combination of elements related to the production of a material; an objective function generating means for generating a second objective function in which a stochastic fluctuation is set for a parameter of the first objective function; an optimization processing means for optimizing a model including the second objective function and the constraint condition; and an output means for outputting values ​​of variables of the second objective function obtained by the optimization as a parameter set. A parameter generating device comprising:

2. The objective function generating means generates a second objective function having a parameter set to a fluctuation represented by a random variable following a predetermined probability distribution. The parameter generating device according to claim 1.

3. The objective function generating means generates a second objective function with a parameter set to fluctuations represented by random variables that follow a probability distribution with a mean of zero. The parameter generating device according to claim 1.

4. The objective function generating means generates a second objective function with a parameter set to the fluctuation represented by a random variable following a normal distribution or a uniform distribution. The parameter generating device according to claim 1.

5. The objective function generating means generates a second objective function with a parameter set to the fluctuation represented by a random variable following a normal distribution whose standard deviation is a constant multiple of the parameter that gives the fluctuation. The parameter generating device according to claim 1.

6. The optimization processing means generates a mathematical optimization problem including the second objective function and the constraint conditions as a model to be optimized, and causes a computer that executes a mathematical program solver to execute the generated model. The parameter generating device according to claim 1.

7. The optimization processing means generates an Ising model to be optimized based on the second objective function and the constraint conditions, and causes the annealing machine to execute the generated Ising model. The parameter generating device according to claim 1.

8. a prediction model generating device that uses past experimental data as training data to learn a prediction model in which materials are used as explanatory variables and characteristic values ​​indicating the characteristics of the materials are used as objective variables; a first objective function generating device that generates a first objective function that specifies a combination of factors related to the production of a material using the prediction model; a parameter generating device that generates a parameter set by utilizing the first objective function; the first objective function generating device generates a first objective function including a linear sum of the characteristic values ​​indicated by the objective variables as a combination of the elements, and inputs the first objective function to the parameter generating device; The parameter generating device comprises: an input means for receiving an input of the first objective function and a constraint condition; an objective function generating means for generating a second objective function in which a stochastic fluctuation is set for a parameter of the first objective function; an optimization processing means for optimizing a model including the second objective function and the constraint condition; and output means for outputting values ​​of variables of the second objective function obtained by the optimization as a parameter set. A parameter generation system comprising:

9. The computer receives an input of a first objective function and constraint conditions that define a combination of factors related to the production of the material; the computer generates a second objective function by setting a stochastic fluctuation in a parameter of the first objective function; the computer optimizes a model including the second objective function and the constraints; The computer outputs the values ​​of the variables of the second objective function obtained by optimization as a parameter set. A parameter generation method comprising:

10. On the computer, an input process for receiving an input of a first objective function and constraint conditions that define a combination of factors related to the production of materials; an objective function generation process for generating a second objective function in which a stochastic fluctuation is set for a parameter of the first objective function; an optimization process for optimizing a model including the second objective function and the constraints; and An output process of outputting the values ​​of the variables of the second objective function obtained by the optimization as a parameter set. A parameter generation program for executing the above.