Information processing method, program, information processing device, method for producing organic compounds, and method for constructing an estimation model.

Abstract

To provide an information processing method that can efficiently estimate a reaction performance at a predetermined scale.SOLUTION: An information processing method causes a computer to execute processing of, based on a reaction condition and a reaction performance related to reactions performed at a first scale and a second scale different from the first scale in a two or more phase liquid-liquid reaction system, by using an estimation model constructed to estimate the reaction performance for the reaction condition, estimating the reaction performance for the reaction condition related to a reaction performed at a predetermined scale in the two or more phase liquid-liquid reaction system.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing method, a program, an information processing device, a method for producing organic compounds, and a method for constructing an estimation model. [Background technology]

[0002] In manufacturing, when producing a product using new, large-scale equipment, it is common practice to first produce test samples using smaller equipment before transitioning to production using the larger equipment. The process of determining various parameters, such as reaction conditions, for the larger equipment based on results obtained from smaller equipment (lab-scale) in research laboratories, so that these results can be applied to the larger equipment, is called scaling up.

[0003] For example, Patent Document 1 discloses a multifluid processor that can be easily scaled up and in which the device does not easily become larger when scaled up. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2018-47393 [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] In scaling up, determining parameters to achieve equivalent reaction results in both small and large-scale equipment is not easy. Generally, scaling up involves a trial-and-error approach where data is acquired at multiple points with different reaction conditions at lab scale, then data is acquired using a large-scale or intermediate-sized instrument, and the process is repeated at lab scale based on the results to find the optimal parameters. This trial-and-error approach is time-consuming and costly.

[0006] To efficiently optimize reaction conditions during scale-up, a technique for efficiently estimating reaction performance at a given scale is desirable.

[0007] The main purpose of this disclosure is to provide an information processing method, etc., that can efficiently estimate reaction performance at a predetermined scale. [Means for solving the problem]

[0008] An information processing method according to one aspect of this disclosure involves a computer performing a process to estimate the reaction performance for a reaction carried out at a predetermined scale in a liquid-liquid reaction system of two or more phases, using an estimation model constructed to estimate the reaction performance for a reaction with reaction conditions based on the reaction conditions and reaction performance for a reaction carried out at a first scale and a second scale different from the first scale in a liquid-liquid reaction system of two or more phases.

[0009] A method for producing an organic compound according to one aspect of this disclosure involves obtaining the reaction performance required for a reaction carried out on a predetermined scale in a two-phase or more liquid-liquid reaction system, identifying reaction conditions that satisfy the obtained reaction performance using an estimation model constructed to estimate the reaction performance for a given reaction condition based on the reaction conditions and reaction performance for a first scale and a second scale different from the first scale in the two-phase or more liquid-liquid reaction system, and obtaining an organic compound based on the identified reaction conditions.

[0010] A method for constructing an estimation model according to one aspect of this disclosure involves obtaining reaction conditions and reaction results for a reaction carried out at a first scale and a second scale different from the first scale in a liquid-liquid reaction system with two or more phases, and constructing an estimation model that estimates the reaction results for a given reaction condition based on the obtained reaction conditions and reaction results. [Effects of the Invention]

[0011] According to this disclosure, reaction performance at a predetermined scale can be efficiently estimated. This makes it possible to efficiently optimize reaction conditions during scale-up. [BRIEF DESCRIPTION OF THE DRAWINGS]

[0012] [Figure 1] It is a diagram showing a configuration example of a manufacturing system. [Figure 2] It is an explanatory diagram relating to an estimation model. [Figure 3] It is a flow chart showing an example of a processing procedure for constructing an estimation model. [Figure 4] It is a flow chart showing an example of a processing procedure relating to scale-up using an estimation model. [MODE FOR CARRYING OUT THE INVENTION]

[0013] The present disclosure will be specifically described with reference to the drawings showing embodiments thereof.

[0014] FIG. 1 is a diagram showing a configuration example of a manufacturing system 100. In the present embodiment, a manufacturing system that identifies reaction conditions for a manufacturing apparatus of a predetermined scale and manufactures an organic compound that is a reaction product according to the reaction conditions by performing scale-up using an estimation model 1M that estimates reaction results with respect to reaction conditions will be described. The manufacturing system 100 includes an information processing apparatus 1 and a manufacturing apparatus 2.

[0015] The information processing apparatus 1 is an apparatus capable of performing various types of information processing and transmitting / receiving information, and is, for example, a server computer, a personal computer, a quantum computer, or the like. The information processing apparatus 1 includes a control unit 11, a storage unit 12, a communication unit 13, an input unit 14, and an output unit 15. The information processing apparatus 1 may be configured to be composed of a plurality of computers and perform distributed processing, may be realized by a plurality of virtual machines provided in one server, or may be realized by using a cloud server.

[0016] The control unit 11 includes a processor using one or more CPUs (Central Processing Units), GPUs (Graphics Processing Units), etc. The control unit 11 uses built-in memory such as ROM (Read Only Memory) or RAM (Random Access Memory), a clock, counters, etc., to control each component and execute processing.

[0017] The storage unit 12 includes, for example, a non-volatile memory such as a hard disk, flash memory, or SSD (Solid State Drive). The storage unit 12 may also be an external storage device connected to the information processing device 1. The storage unit 12 stores various programs and data referenced by the control unit 11. The storage unit 12 in this embodiment stores a program 1P that causes a computer to perform processing related to the estimation of reaction results and the identification of reaction conditions, and an estimation model 1M necessary for the execution of this program 1P. The estimation model 1M is a model that estimates reaction results according to reaction conditions for a reaction carried out on a predetermined scale.

[0018] The program (program product) including program 1P may be provided on a non-temporary recording medium 1A on which the program is recorded in a readable format. The storage unit 12 stores the program read from the recording medium 1A by a reading device (not shown). The recording medium 1A is, for example, a magnetic disk, an optical disk, or a semiconductor memory. Alternatively, the program may be downloaded from an external server connected to a communication network and stored in the storage unit 12. Program 1P may be a single computer program or composed of multiple computer programs, and may be executed on a single computer or on multiple computers interconnected by a communication network.

[0019] The communication unit 13 includes a communication module for communicating with an external device via the network N. The control unit 11 sends and receives data to and from the external device via the communication unit 13.

[0020] The input unit 14 receives various data necessary for carrying out the organic compound manufacturing method, such as experimental data for determining parameters in the estimated model 1M and manufacturing conditions related to the production of organic compounds (e.g., scale, reaction performance to be met, etc.). The input unit 14 sends the received input content to the control unit 11. The input unit 14 is equipped with, for example, a keyboard, mouse, touch panel device with a built-in display, and an interface for acquiring data from an external source.

[0021] The output unit 15 outputs various data related to the implementation of the organic compound manufacturing method, such as reaction conditions identified using the estimated model 1M. The output unit 15 outputs various information according to instructions from the control unit 11. The output unit 15 includes, for example, a display device.

[0022] The information processing device 1 may be configured to accept operations via an externally connected computer and output information to be notified to the external computer. In this case, the information processing device 1 does not need to have an input unit 14 and an output unit 15.

[0023] The manufacturing apparatus 2 is a device for producing organic compounds, and generates products as the reaction progresses in a two-phase or more liquid-liquid reaction system. As shown in Figure 1, the manufacturing apparatus 2 is equipped with a reaction vessel (stirred vessel) 21 of a predetermined scale. In this embodiment, scale refers to the capacity (volume) of the reaction vessel 21 or the volume of liquid in the reaction vessel 21.

[0024] The reaction vessel 21 is a vessel for mixing a first solution and a second solution, which serve as the raw material solutions, and carrying out the reaction. There may be three or more raw material solutions. The raw material solutions may be solutions in which solid raw materials are dissolved in liquid. The reaction vessel 21 may be a batch system or a continuous flow system. The first solution and the second solution are, for example, an oil-water system, where one is oil-based and the other is water-based. A jacket 22, for example, through which a heat transfer medium flows, is wound around the outer circumference of the reaction vessel 21, and the reaction vessel 21 can be set to a predetermined temperature by supplying the heat transfer medium into the jacket 22.

[0025] The reaction vessel 21 is equipped with a stirring blade 23 for stirring and mixing the solution supplied into the vessel. Droplets are formed in the liquid-liquid (oil-water) mixture system in response to the shear caused by the rotation of the stirring blade 23. The reaction proceeds through the mixing of the first solution and the second solution within the reaction vessel 21. The reaction vessel 21 is equipped with a control device (not shown), which can be used to control various operating conditions, including the rotation speed of the stirring blade 23.

[0026] In the manufacturing process using manufacturing apparatus 2, it is necessary to pre-set the reaction conditions during manufacturing, such as the scale in reaction vessel 21 and the stirring conditions in reaction vessel 21. The reaction conditions are determined so that the reaction performance required for manufacturing can be achieved and the desired organic compound can be produced.

[0027] In the pre-production stage using manufacturing apparatus 2, test samples are generated using a laboratory-scale test apparatus smaller than the actual machine in order to determine the reaction conditions for manufacturing apparatus 2, i.e., the actual machine of a predetermined scale. Based on the results of generating the test samples, scaling up is performed. In this embodiment, efficient scaling up is achieved by utilizing the estimated model 1M.

[0028] Before describing the details of the scale-up process in this embodiment, we will explain the general trial-and-error approach to scale-up. As an example, we will describe a case where the scale-up to the third scale, corresponding to the second actual machine, is performed based on experimental results from the first scale, corresponding to the test apparatus, and the second scale, corresponding to the first actual machine. The first, second, and third scales are all different scales.

[0029] First, the operator acquires a first dataset consisting of multiple data points related to a first-scale reaction obtained through laboratory experiments. The acquired data includes the reaction conditions and the reaction results for those conditions.

[0030] Based on the data obtained from multiple points on the first scale, arbitrary reaction conditions are selected for the third scale that are predicted to yield reaction results equivalent to those on the first scale. The reaction results are obtained by actually performing the reaction on the third scale using the selected reaction conditions. If the error between the actual reaction results on the third scale and the expected reaction results is less than a predetermined value, the selected reaction conditions are judged to be appropriate, and the selected reaction conditions are decided as the reaction conditions for the third scale.

[0031] If the error between the actual reaction results and the expected reaction results exceeds a predetermined value, a second dataset is obtained. Using the obtained second dataset, a new reaction condition is selected, and the cycle of verifying the reaction results is repeated. For example, the second dataset may include one additional data point for the reaction at the first scale and one data point for the reaction at the second scale.

[0032] If the reaction conditions cannot be determined based on the second dataset, a third dataset is acquired, and the above process is repeated. For example, the third dataset may include one additional data point for the reaction at the second scale and one data point for the reaction at the third scale. The cycle of acquiring and validating new experimental data is repeated until a third-scale reaction condition that satisfies the predetermined conditions is selected.

[0033] The above-described process may be carried out not only for the third scale but also for the second scale. That is, the reaction conditions for both the second and third scales may be determined. When determining the reaction conditions for both the second and third scales, it is expected that even more search iterations will be required than when only the third scale is considered.

[0034] As mentioned above, trial and error requires repeating the exploration process multiple times until the optimal reaction conditions are obtained. The amount of experimental data increases with the number of cycles. Parameter adjustment relies heavily on human experience and intuition, and less experienced operators are more likely to require an even longer number of cycles.

[0035] Next, the scale-up method of this embodiment will be described. First, a dataset is obtained, for example, consisting of multiple data points related to the reaction at the first scale obtained from a laboratory experiment, and a single data point related to the reaction at the second scale. As an example, the scale ratio of the second scale to the first scale (second scale / first scale) can be 100 or more. The number of data points is illustrative and is not limited to the example above.

[0036] Using the obtained dataset, we construct an estimation model 1M that estimates reaction performance according to reaction conditions. Specifically, we identify the parameters of estimation model 1M that estimates reaction performance according to reaction conditions. Using the constructed estimation model 1M, we derive the third scale reaction conditions so that the reaction performance estimated by estimation model 1M satisfies the predetermined reaction conditions. This determines the three scale reaction conditions.

[0037] According to the method of this embodiment, reaction performance can be accurately estimated by using an estimation model 1M constructed with a minimum number of data points. The selection of reaction conditions and verification of reaction performance can be achieved in a single cycle, and the man-hours required for scaling up can be reduced.

[0038] Details of the estimation model 1M and its construction method in this embodiment will be described. Figure 2 is an explanatory diagram of the estimation model 1M. The estimation model 1M is a model that represents the relationship between reaction conditions and reaction results for liquid-liquid reaction systems with two or more phases, and is a model that takes reaction conditions as input and outputs reaction results.

[0039] In this embodiment, as an example, estimation model 1M takes stirring conditions and equipment information as explanatory variables as input and estimates reaction characteristic values ​​as the objective variable.

[0040] The stirring conditions included in the reaction conditions include, for example, the droplet size d. pExamples of equipment information include the scale of the reaction vessel 21. The equipment information may further include information regarding the shape of the reaction vessel 21 and the type of agitator blade 23. In the example shown in Figure 2, the equipment information includes the scale of the reaction vessel 21 and the blade diameter of the agitator blade 23.

[0041] Reaction characteristic values ​​included in reaction results include, for example, the reaction rate constant k. The reaction rate constant k is an indicator of the reaction rate.

[0042] The droplet diameter d indicates the stirring conditions of this embodiment. p For [m], the following equation (1) holds true.

[0043]

number

[0044] In equation (1), C is a constant, d is the blade diameter [m] of the impeller 23, We is the Weber number, and α is a parameter.

[0045] We is a dimensionless number that represents the ratio of the inertial force of the flow to the surface tension, and can be expressed by the following equation (2).

[0046]

number

[0047] In equation (2), ρ is the continuous phase (fluid) density [kg / m³]. 3 ], where n is the stirring speed [1 / s], d is the blade diameter [m], and σ is the surface tension [N / m].

[0048] As described above, the droplet size d indicates the stirring conditions. p This includes a parameter α related to the Weber number. This parameter α is one of the parameters that are subject to scale-up. The value of parameter α is known to change depending on the scale of the reaction vessel 21. Droplet diameter d p By adjusting (controlling) this, the reaction rate can be adjusted.

[0049] The information processing apparatus 1 specifies parameters α1 and α2 in an estimation model 1M by performing fitting calculation of the estimation model 1M based on actually measured data of reaction conditions and reaction results in a first scale and a second scale.

[0050] An example of a method for specifying parameters α1 and α2 will be described. The information processing apparatus 1 acquires the scale of a reaction tank 21, the stirring rotation speed, the impeller diameter, density, surface tension, and the reaction rate constant k for a plurality of reaction conditions (experimental conditions) based on actually measured data in the first scale and the second scale. In the example shown in Figure 2, three pieces of data related to the first scale and one piece of data related to the second scale are acquired. The method for measuring and calculating each of the above data through experiments is well known, so a detailed description thereof is omitted.

[0051] The information processing apparatus 1 calculates We for each reaction condition based on the acquired values. Furthermore, using the calculated We, the droplet diameter d p is calculated. Here, as the parameter α used for calculating the droplet diameter d p different parameters α1 and α2 are selected for each scale of the reaction tank 21. Based on We in each reaction condition of the first scale, the impeller diameter of the first scale, the parameter α1, and the like, the droplet diameter d p is obtained for each reaction condition. Similarly, based on We in each reaction condition of the second scale, the impeller diameter of the second scale, the parameter α2, and the like, the droplet diameter d p is obtained for the reaction condition of the second scale. Any values may be set as initial values for the parameters α1 and α2.

[0052] The information processing apparatus 1 specifies, based on the correlation between the calculated droplet diameter d p and the reaction rate constant k, the parameters α1 and α2 with which the correlation between the droplet diameter d p and the reaction rate constant k satisfies a predetermined condition. Specifically, the droplet diameter d pThe information processing device 1 identifies parameters α1 and α2 that maximize the absolute value of the correlation coefficient calculated based on the reaction rate constant k. The information processing device 1 may determine the optimal values ​​of parameters α1 and α2, for example, by the least squares method. The information processing device 1 may also identify parameters α1 and α2 that have an absolute value of the correlation coefficient greater than or equal to a predetermined value.

[0053] The correlation coefficient can be calculated, for example, using the formula (correlation coefficient = covariance between the value of the first data point included in multiple data sets and the value of the second data point included in multiple data sets other than the first data point / (standard deviation of the first data point value × standard deviation of the second data point value)).

[0054] Parameter α may be subject to predetermined constraints. A first example of a constraint is an upper and lower limit based on the characteristics of We (e.g., 0.6 ≤ α ≤ 1.0). A second example of a constraint is a limit on the number of digits based on the ease of handling of parameter α (e.g., one decimal place). The information processing device 1 identifies parameters α1 and α2, taking into account the set constraints.

[0055] Through the above process, the droplet diameter d including parameter α is obtained. p An estimation model 1M is constructed that outputs a reaction rate constant k corresponding to the given value.

[0056] Next, the information processing device 1 uses the estimation model 1M constructed for the second scale, i.e., the estimation model 1M including parameter α2, to derive reaction conditions for the third scale that satisfy the predetermined reaction performance. Specifically, the reaction conditions are identified such that the reaction performance estimated using the estimation model 1M satisfies the predetermined reaction performance.

[0057] For example, based on the estimated model 1M, the droplet size d corresponds to the required reaction rate constant k. p Identify the droplet diameter d. p Based on this, further conditions are identified. Specifically, the droplet diameter d is determined based on the third-scale airfoil diameter and parameter α2. pThe corresponding We is calculated. Based on the calculated We, the density ρ according to the liquid properties, and the surface tension σ, the stirring speed for the third scale is determined. This determines the droplet diameter d for the third scale to satisfy the predetermined reaction performance. p The reaction conditions, including the stirring speed, are then identified.

[0058] In this embodiment, an estimation model 1M including parameter α2 corresponding to the second scale is applied to identify the reaction conditions for the actual apparatus at a third scale, which is a different scale from the second scale. As described above, the value of parameter α in estimation model 1M changes according to the scale of the reaction vessel 21. In other words, for the third scale, which is within the scale range to which a specific parameter αn can be applied and whose scale ratio to the second scale is within a predetermined range, it is possible to estimate the reaction performance using estimation model 1M including parameter αn. Therefore, it is preferable to set the scale of the actual apparatus to be considered within the predetermined scale range to which parameter αn can be applied, and to identify other reaction conditions.

[0059] According to the inventors' investigation, it was confirmed that when the scale ratio (third scale / second scale) of the third scale of the second prototype being studied to the second scale of the first prototype from which parameter α2 was obtained is in the range of 1 to approximately 2.5, the predicted value by the estimation model 1M for reaction performance related to the third scale is in close agreement with the measured value (the error between the predicted value and the measured value is below a predetermined standard).

[0060] If the scale of the actual machine under consideration falls outside the applicable range of parameter αn, it is advisable to identify a new parameter αn+1 using measurement data from another actual machine with a scale close to that of the machine under consideration. Using the identified parameter αn+1, the reaction performance can be estimated using estimation model 1M for actual machines within the applicable range of parameter αn+1. In this way, by adjusting parameter α in estimation model 1M for each scale range, the reaction performance can be estimated accurately and efficiently using estimation model 1M.

[0061] Furthermore, the identification of reaction conditions using the estimation model 1M including parameter α2 can also be performed for the second scale. The information processing device 1 may use the estimation model 1M including parameter α2 to identify reaction conditions that satisfy the predetermined reaction performance required for the reaction at the second scale, in addition to or instead of the third scale. That is, the estimation model 1M including parameter α2 may be used to identify reaction conditions other than those for which measured data was obtained, with respect to the second scale.

[0062] The reaction conditions are not limited to the examples described above. The reaction conditions may include, for example, information regarding the stirring power. Furthermore, parameter α may also relate to the stirring power.

[0063] The stirring power is the energy used to stir the fluid in the tank, and can be expressed by the following equation (3).

[0064]

number

[0065] In formula (3), P V This is the stirring power per unit volume [kg / m³] 3 ], N p ρ is the power number, d is the blade diameter [m], n is the stirring rotation speed [1 / s], and ρ is the density [kg / m³]. 3 ], V is the liquid volume [m 3 The liquid volume corresponds to the scale of reaction vessel 21.

[0066] The reaction conditions are not limited to those related to stirring operations, but may include other factors that change with scale-up. The reaction conditions may include information such as the temperature of the reaction vessel 21, the rate of heat removal or heating in the reaction vessel 21, the flow rate of the heat transfer medium in the jacket 22 (heat source), the temperature of the heat transfer medium in the jacket 22 (heat source), the substrate concentration, the catalyst concentration, etc. Furthermore, parameter α may relate to the heat transfer coefficient, for example.

[0067] Similarly, the reaction results are not limited to the examples described above. The reaction results may include any characteristic value that correlates with the reaction conditions, such as the weight-average molecular weight Mw at a predetermined time after the start of the reaction.

[0068] To obtain products of equivalent quality before and after scale-up, it is crucial to match the reaction rates between different scales. By constructing an estimated model 1M, the reaction conditions that satisfy the desired reaction rate can be efficiently identified, thereby improving the efficiency of scale-up.

[0069] Figure 3 is a flowchart showing an example of the procedure for constructing the estimated model 1M. The processes in each flowchart below may be executed by the control unit 11 according to the program 1P stored in the memory unit 12 of the information processing device 1, or they may be implemented by a dedicated hardware circuit (e.g., FPGA or ASIC) provided in the control unit 11, or they may be implemented by a combination of the above.

[0070] The control unit 11 of the information processing device 1 acquires multiple measured data regarding the reaction at the first scale and the second scale (step S11).

[0071] The control unit 11 acquires reaction conditions and reaction results based on the acquired measured data (step S12). In step S12, the control unit 11 randomly sets provisional values ​​for parameters α1 and α2, for example, within the range of constraints for parameter α. Based on the set parameters α1 and α2 and the measured data, the control unit 11 determines the droplet diameter d for each data point according to parameters α1 and α2. p The scale, airfoil diameter, and reaction rate constant are obtained.

[0072] The control unit 11 determines the droplet size d as a reaction condition. pBased on the reaction rate constant k as the reaction result, parameters α1 and α2 in the estimation model 1M are identified so that the correlation between the reaction conditions and the reaction result satisfies predetermined conditions (step S13). Specifically, the control unit 11 determines the droplet diameter d for all data points. p A correlation coefficient is calculated based on the reaction rate constant k, and parameters α1 and α2 are optimized, for example, by the least squares method, so that the absolute value of the calculated correlation coefficient is maximized.

[0073] The above process completes the identification of parameter α1 for the first scale and parameter α2 for the second scale, and the construction of estimation model 1M for the first and second scales.

[0074] Figure 4 is a flowchart showing an example of the processing procedure for scaling up using the estimated model 1M.

[0075] The control unit 11 of the information processing device 1 acquires the reaction performance required for the reaction carried out in the third-scale manufacturing device 2, for example, by receiving input from a manufacturing worker (step S21).

[0076] The control unit 11 acquires the reaction conditions for the reaction carried out in the third-scale manufacturing apparatus 2 based on the acquired reaction results (step S22). In step S22, the control unit 11 may set the reaction conditions based on the parameter α2 in the estimation model 1M that includes the third scale in its scope, i.e., the estimation model 1M for the second scale.

[0077] The control unit 11 estimates the reaction performance for the acquired reaction conditions based on the estimation model 1M which includes the parameter α2 (step S23). The control unit 11 determines whether the estimated reaction performance meets the reaction performance required for the third scale obtained in step S21 (step S24). If it is determined that the estimated reaction performance does not meet the required reaction performance (step S24: NO), the control unit 11 returns to step S22.

[0078] If the control unit 11 determines that the estimated reaction performance meets the required reaction performance (step S24: YES), it identifies the estimated reaction conditions as reaction conditions for the third scale and outputs the estimated reaction conditions to the output unit 15 (step S25). The control unit 11 may also output the estimated reaction conditions to an external computer. For example, the control unit 11 identifies the droplet diameter d p Based on the estimated reaction conditions, further conditions may be identified, such as determining the stirring speed. The control unit 11 then terminates the example process.

[0079] In the process described above, the reaction conditions set using the estimated model 1M are predicted to satisfy the desired reaction results with a high probability. Therefore, scaling up to a predetermined scale can be performed efficiently.

[0080] The above example described scaling up to the third scale, but scaling up to the second scale can be done using the same procedure.

[0081] Next, a method for producing an organic compound according to this embodiment, to which the above-described information processing method is applied, will be explained.

[0082] The method for producing the organic compound according to this embodiment involves producing the organic compound using the production apparatus 2 under reaction conditions related to the third scale identified by the information processing method according to this embodiment.

[0083] A method for producing an organic compound comprises the steps of: obtaining the reaction performance required for a reaction carried out on a predetermined scale in a two-phase or more liquid-liquid reaction system; identifying reaction conditions that satisfy the obtained reaction performance using an estimation model constructed to estimate the reaction performance for reaction conditions based on the reaction conditions and reaction performance for a first scale and a second scale different from the first scale in a two-phase or more liquid-liquid reaction system; and obtaining an organic compound based on the identified reaction conditions.

[0084] In the process of obtaining reaction results, the reaction results required for the manufacturing apparatus 2 of a predetermined scale are obtained. The predetermined scale corresponding to the manufacturing apparatus 2 can be the second scale or the third scale described above.

[0085] The step of identifying reaction conditions that satisfy the acquired reaction results can be performed in the same manner as the information processing method described above. That is, based on the estimated model 1M equipped with the constructed parameter α2, the reaction conditions that satisfy the reaction results required for the manufacturing apparatus 2 are identified.

[0086] In the process of obtaining an organic compound based on specified reaction conditions, the organic compound is produced by controlling the operation of a predetermined scale manufacturing apparatus 2 according to, for example, specified stirring speed, blade diameter, etc. The organic compound can be obtained by appropriately combining known synthesis reactions and mixing the raw material solutions.

[0087] According to this embodiment, reaction performance can be efficiently estimated by constructing an estimation model that estimates reaction performance for given reaction conditions. By setting the parameter α in the estimation model for each scale, reaction performance can be estimated with high accuracy using the estimation model within the applicable scale range. By constructing an estimation model with a small number of data points and then scaling it up, the trial and error involved in scaling up can be significantly reduced. Since the number of experimental data required for scaling up can be reduced, it leads to cost reduction.

[0088] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The technical features described in each embodiment can be combined with each other, and the scope of the present invention is intended to include all modifications within the claims and equivalents thereof. The sequences shown in each embodiment are not limiting, and within the bounds of consistency, the order of each processing step may be changed, and multiple processes may be executed in parallel. The processing entity for each process is not limiting, and within the bounds of consistency, the processing of each device may be executed by other devices.

[0089] The matters described in each embodiment can be combined with each other. Furthermore, the independent and dependent claims described in the claims can be combined with each other in any combination, regardless of the form of reference. In addition, the claims use a form in which claims referencing two or more other claims (multi-claim form), but are not limited to this. A form in which multi-claims referencing at least one multi-claim (multi-multi-claim) may also be used. [Explanation of Symbols]

[0090] 100 Manufacturing Systems 1. Information Processing Device 11 Control Unit 12 Storage section 13 Communications Department 1P Program 1M Estimated Model 1A Recording medium 2 Manufacturing equipment 21 Reaction vessel 22 Jackets 23 Stirring blade

Claims

1. Based on the reaction conditions and results for reactions carried out at a first scale and a second scale different from the first scale in a two-phase or more liquid-liquid reaction system, an estimation model constructed to estimate the reaction performance for a given reaction condition is used to estimate the reaction performance for a reaction carried out at a predetermined scale in a two-phase or more liquid-liquid reaction system. An information processing method in which a computer performs the processing.

2. Based on the reaction results, identify reaction conditions that satisfy the reaction results required for a reaction carried out on a predetermined scale. The information processing method according to claim 1.

3. The estimation model includes scale-specific parameters identified such that the correlation between the reaction conditions and the reaction results satisfies predetermined conditions. The information processing method according to claim 1 or claim 2.

4. The reaction conditions include stirring conditions. The information processing method according to claim 1 or claim 2.

5. The aforementioned stirring conditions include a parameter relating to the We number, which represents the ratio of the flow's inertial force to its surface tension. The information processing method according to claim 4.

6. The first scale or the second scale corresponds to the scale of the test apparatus, and the predetermined scale corresponds to the scale of the actual machine. The information processing method according to claim 1 or claim 2.

7. The predetermined scale is equal to or greater than the larger of the first scale and the second scale. The information processing method according to claim 1 or claim 2.

8. Using an estimation model constructed to estimate reaction performance for a given scale of reaction in a liquid-liquid reaction system with two or more phases, based on the reaction conditions and reaction performance for a first scale and a second scale different from the first scale, the reaction performance for a given scale of reaction in a liquid-liquid reaction system with two or more phases is estimated. A program that causes a computer to perform a process.

9. Based on the reaction conditions and results for reactions carried out at a first scale and a second scale different from the first scale in a two-phase or more liquid-liquid reaction system, an estimation model constructed to estimate the reaction performance for a given reaction condition is used to estimate the reaction performance for a reaction carried out at a predetermined scale in a two-phase or more liquid-liquid reaction system. It includes a control unit that performs processing. Information processing device.

10. To obtain the reaction performance required for reactions carried out on a predetermined scale in liquid-liquid reaction systems with two or more phases, Based on the reaction conditions and results for reactions carried out on a first scale and a second scale different from the first scale in a two-phase or more liquid-liquid reaction system, an estimation model constructed to estimate the reaction results for a given reaction condition is used to identify reaction conditions that satisfy the acquired reaction results. Organic compounds are obtained based on the identified reaction conditions. A method for producing organic compounds.

11. Using the estimation model described above, the reaction performance for reaction conditions related to the reaction carried out on the predetermined scale is estimated. Identify reaction conditions that satisfy the obtained reaction results based on the estimated reaction results. A method for producing an organic compound according to claim 10.

12. In a liquid-liquid reaction system with two or more phases, reaction conditions and reaction results are obtained for reactions carried out on a first scale and a second scale different from the first scale. Based on the acquired reaction conditions and reaction results, we construct an estimation model to estimate the reaction results for given reaction conditions. Method for constructing estimation models.

13. Based on multiple sets of reaction conditions and reaction results in the first scale and at least one set of reaction conditions and reaction results in the second scale, parameters in the estimation model are identified for each scale such that the correlation between reaction conditions and reaction results satisfies predetermined conditions. A method for constructing an estimation model according to claim 12.

Images