Reaction path analysis system, manufacturing device, analysis method, and program
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-12-17
- Publication Date
- 2026-08-13
Smart Images
Figure JP2025044084_13082026_PF_FP_ABST
Abstract
Description
Reaction Path Analysis System, Manufacturing Apparatus, Analysis Method, and Program
[0001] The present invention relates to a reaction path analysis system, a manufacturing apparatus, an analysis method, and a program. The present invention claims the priority of Japanese Patent Application No. 2025-017021 filed on February 4, 2025, and for designated countries where incorporation by reference is permitted, the contents described in that application are incorporated herein by reference.
[0002] In industrial products, it is important to understand chemical reactions such as combustion, deterioration, and decomposition of materials in the use environment. Understanding the reactions makes it possible to ensure the reliability and safety during product use. Also, even when unwanted reactions occur, it is possible to improve the efficiency of investigating the causes and designing materials to suppress the reactions. In recent years, due to the progress of computer simulation technology, methods for understanding the chemical changes of various materials have been developed.
[0003] Japanese Patent Application Laid-Open No. 2021-163422
[0004] For example, Patent Document 1 describes an analysis method in which a reaction network diagram showing the reaction path of a chemical reaction is prepared, the reaction rate in each reaction path is predicted using an artificial intelligence algorithm, and updated based on the actual implementation results to search for optimal reaction conditions. However, in reactions in the actual environment, there was a possibility that phenomena other than the predetermined reaction patterns would occur due to changes in reaction conditions that occur locally.
[0005] An object of the present invention is to provide a technique for analyzing a reaction path closer to actual phenomena, taking into account the changes that occur in reactions in the actual environment.
[0006] This application includes a plurality of means for solving at least part of the above problems. For example, they are as follows.
[0007] One aspect of the present invention is a reaction pathway analysis system having one or more processors and one or more memory resources. The memory resources store estimated values in the reaction pathway network of the reaction system. The processor receives an input of actual observed values in the reaction system and stores them in the memory resources, reads out and compares the estimated values and the observed values to extract differences, identifies chemical species existing only on the observed values from the differences, obtains the structural similarity between the chemical species existing only on the observed values and the chemical species existing on the estimated values, generates a reaction pathway from the chemical species existing on the estimated values as a starting point to the chemical species existing only on the observed values based on the structural similarity, updates the reaction pathway network, and outputs the updated reaction pathway network.
[0008] According to the present invention, a technique for analyzing a reaction pathway closer to an actual phenomenon can be provided.
[0009] Problems, configurations, and effects other than those described above will be clarified by the description of the following embodiments.
[0010] It is a configuration diagram of a reaction pathway analysis system according to an embodiment. It is a schematic diagram of reaction system information when using molecular dynamics calculation. It is a schematic diagram of an estimated reaction pathway network. It is a schematic diagram of observed value data. It is a schematic diagram of a difference table. It is a schematic diagram of a structural similarity table. It is a block diagram showing an example of the configuration of an analysis device. It is a flowchart showing an example of processing in an analysis device. It is a schematic diagram of a deletion permission selection screen. It is a schematic diagram of an update target selection screen. It is a schematic diagram of an additional reaction pathway. It is a schematic diagram of a reaction pathway selection screen.
[0011] Hereinafter, one embodiment of the present invention will be described with reference to the drawings. In all drawings used to describe the embodiment, the same reference numerals will be used for the same components as a general rule, and repeated descriptions will be omitted as appropriate. Furthermore, in the following embodiment, it goes without saying that the components (including element steps, etc.) are not necessarily essential unless specifically stated or considered to be clearly essential in principle. Furthermore, it goes without saying that other elements are not excluded unless specifically stated that only that element is included. Similarly, in the following embodiment, when referring to the shape, positional relationship, etc. of components, etc., it will include those that are substantially similar or similar to the shape, etc., unless specifically stated or considered to be clearly not the case in principle. In addition, the position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc., in order to facilitate understanding of the invention. For this reason, the present invention is not necessarily limited to the position, size, shape, range, etc., disclosed in the drawings.
[0012] Figure 1 is a configuration diagram of a reaction pathway analysis system 100 according to one embodiment of the present invention. The reaction pathway analysis system 100 includes an input / output device 10, an analysis device 20, a network 30, and a server 40. The input / output device 10 and the analysis device 20 are connected via an I / F unit (input / output interface unit) 23 to enable mutual transmission and reception of information. The analysis device 20 also communicates with an external server 40 via a network 30 such as a LAN (Local Area Network) or the Internet.
[0013] The input / output device 10 includes, for example, an input unit 11, which is implemented by a mouse, keyboard, touch panel, etc., and a display unit 12, which is implemented by a display, etc. The input unit 11 receives user input operations and transmits operation signals to the analysis device 20.
[0014] The display unit 12 displays the display screen output from the analysis device 20 and presents it to the user.
[0015] The input unit 11 and the display unit 12 are not limited to the input and display devices described above, but may also be implemented by a user-operated terminal device such as a PC or tablet connected to the analysis device 20. Such a terminal device may be connected to the analysis device 20 via the network 30, and it is possible to send and receive information with the analysis device 20 on the cloud to obtain the desired information. Alternatively, it may be a terminal device connected to the server 40.
[0016] The analysis device 20 constructs a reaction pathway network that more closely resembles real phenomena by using a predicted reaction pathway network derived from computational simulations and databases for the object of analysis, and observed values from actual chemical reaction experiments. The reaction system to be analyzed may include any compound or aggregate thereof, such as organic substances, inorganic substances, and polymers (monomers, oligomers).
[0017] Generally, the chemical reactions occurring in a reaction system differ depending on the type and composition ratio of the substances being analyzed, and the influence of localized temperature and pressure is also unique. Under diverse environmental conditions, reaction pathways may change, or additional reaction pathways that produce side reactions may occur, making high-precision analysis difficult with only computational simulations and database (hereinafter simply referred to as DB) predictions. For example, when creating reaction pathways based on reaction systems that have been pre-entered in the DB, the accuracy of the analysis decreases if the composition of the system being analyzed does not exist in the DB. In this respect, molecular dynamics calculations can track the time evolution of the motion of atoms and molecules by considering interaction parameters between atoms and external environments such as temperature and pressure, so it is possible to reveal reactions that are not in the DB. However, because the computable scale is small, on the order of nanometers, it is difficult to fully consider local environmental changes. Therefore, the analysis device 20 compares the predicted values in the reaction system with the actual observed values and updates the predicted reaction pathway network based on the difference to generate a new reaction pathway network that conforms to the actual environment.
[0018] Specifically, the analysis device 20 includes a calculation unit 21, a storage unit 22, and an I / F unit 23. The calculation unit 21 includes a reaction path generation unit 211 and a reaction path update unit 212. The storage unit 22 includes reaction system information 221, condition setting information 222, estimated reaction path network information 223, observed value data 224, and updated reaction path network information 225.
[0019] The reaction pathway generation unit 211 estimates the reaction pathway of the reaction system under analysis and generates an estimated reaction pathway network. Specifically, the reaction pathway generation unit 211 has the user register reaction system information 221, which is reaction system information necessary for deriving the estimated reaction pathway network, and generates the estimated reaction pathway network based on this information.
[0020] The reaction system information 221 is not defined in a general way, as it depends on the method for generating the predicted reaction pathway network. For generation, for example, the Python library RDKit, the quantum chemistry calculation programs Gausian and GAMESS, and the molecular dynamics calculation program LAMMPS (open-source software developed by Sandia National Laboratories of the U.S. Department of Energy) can be used. In this embodiment, the reaction pathway generation unit 211 will utilize a molecular dynamics calculation program. Molecular dynamics calculation is a method that calculates the forces acting on atoms using quantum chemistry calculations and tracks the time evolution of atomic coordinates and velocities from Newton's equations of motion. Any molecular dynamics calculation method can be used to perform molecular dynamics calculations, but more preferably, reaction force fields (ReaxFF, machine learning force fields) or first-principles molecular dynamics calculations, which can take chemical reactions into account, are used.
[0021] Figure 2 shows an example of reaction system information 221 when molecular dynamics calculations are used. The reaction system information 221 includes an ID, which is a unique identifier for the data; a reaction system composition, which is information about the composition of atoms and molecules included in the reaction system to be analyzed; and setting conditions, which are information about the details of the reaction system. Setting conditions include, for example, information about the covalent bond radius of atoms, intermolecular interactions, cell size, cutoff distance, periodic boundary conditions, initial velocity, initial configuration, force field, ambient temperature, pressure, etc. When registering, the user may register their desired conditions themselves, or they can read a condition setting pattern from the pre-stored condition setting information 222 and select it. The condition setting information 222 may also be stored on an external server 40 and retrieved via the network 30.
[0022] The reaction pathway generation unit 211 performs molecular dynamics calculations based on the reaction system information 221 and determines the presence or absence of bonds from the distance between atoms, etc., thereby obtaining the temporal change in structure (reaction dynamics) as physical property information. From such physical property information, it is possible to know the structure, pathway, and amount of product (yield) from the starting material to the intermediate and final product. The reaction pathway generation unit 211 also maps the obtained physical property information onto a network to generate a predicted reaction pathway network. An example of this is shown in Figure 3. As shown in the figure, the reaction pathway network is a one-dimensional trajectory map that connects the elementary reaction processes from a stable structure through an energetically unstable transition state to another stable structure. Although omitted in the figure, it is desirable that information such as the amount of product (yield) and radicals obtained as physical property information is also displayed on the reaction pathway network. For example, notation that can represent the molecular structure, such as SMILES, can also be used. In this embodiment, yield is used as an indicator of the amount of product, but reaction efficiency may also be used.
[0023] The reaction pathway update unit 212 updates the above-mentioned predicted reaction pathway network based on measured values. Specifically, the reaction pathway update unit 212 accepts input of actual observed values in the reaction system under analysis. The timing of the observation is not particularly limited, but it will be described below assuming that it is after the reaction (i.e., when the reaction system has reached equilibrium). Examples of observed values include analytical data from GC-MS, etc. Data input may be performed by the user, or the measured values may be transmitted directly from the analytical instrument. From such observed values, the reaction pathway update unit 212 generates observed value data 224 that includes at least the structure of the chemical species contained in the reaction system and the amount (yield) of each chemical species produced. An example of such observed value data is shown in Figure 4. The user may be allowed to directly input the structure of the chemical species and the amount (yield) of each chemical species.
[0024] Furthermore, the reaction pathway update unit 212 compares the estimated value with the observed value for each chemical species and updates the estimated reaction pathway network information 223 based on the difference. In this example, the yield, which is easy to grasp intuitively, is used to compare the estimated value with the observed value. Specifically, the reaction pathway update unit 212 calculates the difference between the estimated yield and the observed yield for each chemical species. Figure 5 shows the difference table obtained as a result of this calculation process. The difference can be calculated in any way, but in this example, it is calculated as estimated yield (%) - observed yield (%). A value of zero means that the observed yield and the estimated yield match, and the larger the absolute value of the difference, the greater the discrepancy between the observed value and the estimated value. Also, a positive difference indicates that it appears more frequently in the real environment (observed value), and a negative difference indicates that it appears more frequently in the virtual environment (on the estimated reaction pathway network). In particular, if either the observed yield or the estimated yield is zero and the difference is not zero, there is a great need to modify the reaction pathway leading to that chemical species.
[0025] Therefore, the reaction pathway update unit 212 performs a deletion process to remove chemical species that exist only in the predicted reaction pathway network from the reaction pathway network. This makes it possible to remove chemical species and reaction pathways that do not actually occur from the predicted reaction pathway network. Alternatively, a selection screen may be generated to allow the user to choose whether or not to delete such reaction pathways and output to the display unit 12. Figure 9 shows an example of a deletion permission selection screen. Through such a selection screen, the user can choose whether or not to delete the reaction pathways. This prevents unintended deletions by the user and allows the user to perform the desired simulation.
[0026] Furthermore, the reaction pathway update unit 212 performs additional processing to add new reaction pathways to the predicted reaction pathway network for "chemical species that exist only in the observed values." This adds chemical species and reaction pathways that do not appear in calculations but actually occur, updating the reaction pathway network to a form closer to the real environment.
[0027] In this process, the reaction pathway update unit 212 extracts alternative chemical species that are thought to have been generated in the virtual environment (on the predicted reaction pathway network) in order to create additional reaction pathways leading to chemical species that exist only in the actual environment (on the observed values). Specifically, the reaction pathway update unit 212 extracts alternative chemical species from the predicted reaction pathway network that share at least some constituent elements with the "chemical species that exist only in the observed values" and whose "predicted yield is greater than the observed yield." Chemical species that exist only in the actual environment (on the observed values) are thought to have been replaced by the extracted alternative chemical species (or one of the candidates if there are multiple). By selecting alternative chemical species under these conditions, more accurate extraction becomes possible.
[0028] Next, the reaction pathway update unit 212 identifies the chemical species to be updated from the candidates and generates the reaction pathway. Structural similarity is used to identify the chemical species to be updated. Specifically, for each candidate, the structural similarity to the "chemical species that exists only in the observed values" is calculated, and the one with the highest similarity is determined to be the chemical species to be updated. An example of a structural similarity table, which is the result of calculating structural similarity, is shown in Figure 6. Any method can be used to calculate structural similarity, for example, the Tanimoto coefficient, cosine similarity, Euclidean distance, and Dice coefficient. In this embodiment, the Euclidean distance is used. In Figure 6, the reciprocal of the structural similarity obtained by the Euclidean distance is shown, and a larger value indicates a higher structural similarity. Since species with high structural similarity are more likely to be added to the reaction system by replacing chemical species that exist only in the estimated values, more accurate analysis is possible.
[0029] Furthermore, reaction pathways may be generated and added for other candidates as well. For example, one method is to prioritize selecting chemical species that have a large difference in yield and also have high structural similarity, rather than just selecting the chemical species with the highest structural similarity. Chemical species with a large difference in yield have a large deviation from actual observed values, and updating them may significantly correct the reaction pathway to better reflect reality. It is also possible to have the user select the chemical species to be updated from among the candidates. For example, a selection screen displaying an overview of the reaction pathway and information on each candidate chemical species can be presented, allowing the user to select a candidate. An example of such an update target selection screen is shown in Figure 10. This allows the user's intentions and inferences to be reflected in the reaction pathway network.
[0030] Next, the reaction pathway update unit 212 generates an additional reaction pathway network connecting all chemical species other than the raw material, or the chemical species located immediately before the target chemical species, that are included in the reaction pathway from the raw material to the target chemical species, and leads to "chemical species that exist only in the observed values." This network is then added to the predicted reaction pathway network to update it. This allows for the generation of reaction pathways and their branching positions based on measured values with high accuracy. If multiple reaction pathways exist, the change in the predicted yield when each reaction pathway is added is calculated, and the reaction pathway that minimizes the difference from the observed yield can be selected. This makes it possible to bring the reaction pathway closer to the measured values with high accuracy. One or more reaction pathways may be added.
[0031] Alternatively, the user may be presented with a reaction pathway selection screen to allow them to decide whether to add pathways and to select the desired pathways. An example of a reaction pathway selection screen is shown in Figure 12. The reaction pathway selection screen displays the selectable reaction pathways, the estimated yield when the selected reaction pathway is added, and a difference change table showing the change in the difference between the estimated yield and the observed yield. By referring to the difference changes before and after the update and selecting the reaction pathway with the smallest difference, the user can obtain a reaction pathway network that is closer to the actual measured value and intuitively understand what kind of reactions are occurring. Of course, the user can also choose not to add any reaction pathways.
[0032] The reaction pathway update unit 212 then generates a difference table again from the updated reaction pathway network information 225, which stores the updated reaction pathway network, and determines whether or not there are any chemical species that exist only in the observed values. If there are any, the process of adding the reaction pathway described above is repeated for those chemical species.
[0033] Furthermore, AI (Artificial Intelligence) or predictions from existing databases may be used to determine the validity of selecting chemical species to be updated, selecting additional reaction pathways, and determining whether or not the update is feasible. Any method of determination is acceptable, but for example, a new predicted reaction pathway network can be generated under the same conditions using existing generation AI or databases, and the update content that corresponds to the reaction pathway closest to these can be selected. In addition, for additional processing, the activation energy of the reaction can be determined by first-principles calculations, and screening can be performed by selecting the reaction pathway with the lowest energy. The AI and database may be implemented in the analysis device 20, but external ones can also be used, for example, those provided on the server 40 can be used. For example, the reaction pathway update unit 212 may instruct an existing external generation AI service to determine the validity of the reaction pathway update using at least one of the following: a database of chemical reactions or first-principles calculations. This allows for more accurate updates.
[0034] Figure 7 shows an example of the hardware configuration of the analysis device 20. The analysis device 20 has a hardware configuration that is realized by a so-called server device, workstation, personal computer, smartphone, or tablet terminal housing. The analysis device 20 includes a processor 101, memory 102, storage device 103, communication device 104, and a bus 105 that connects each device.
[0035] The processor 101 is an arithmetic unit that reads various programs stored in the memory 102 and executes processing corresponding to each program. The processor 101 is, for example, a microprocessor, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), a quantum processor, or other arithmetic semiconductor device.
[0036] Memory 102 is a memory device such as RAM (Random Access Memory).
[0037] The storage device 103 is a non-volatile storage device capable of storing digital information, such as a hard disk drive, solid state drive (SSD), or flash memory.
[0038] The communication device 104 is a network interface card (NIC) or the like that communicates with other devices via the data communication network 50.
[0039] The reaction path generation unit 211 and the reaction path update unit 212 of the analysis device 20 described above are implemented by a program that causes the processor 101 to perform processing. This program is stored in memory 102, storage device 103, or a ROM device (not shown), and is loaded into memory 102 for execution and executed by the processor 101.
[0040] Furthermore, the storage unit 22 of the analysis device 20 is realized by the memory 102 and the storage device 103. In addition, the communication unit 150 is realized by the communication device 104. The above is an example of the hardware configuration of the analysis device 20.
[0041] The configuration of the analysis device 20 can be further classified into many components depending on the processing content. Alternatively, it can be classified so that each component performs even more processing.
[0042] Furthermore, each processing unit (reaction path generation unit 211 and reaction path updating unit 212) may be constructed using dedicated hardware (such as an ASIC or GPU) to realize its respective function. Also, the processing of each processing unit may be executed on a single piece of hardware or on multiple pieces of hardware.
[0043] Next, the operation of the analysis device 20 in this embodiment will be described. Figure 8 is a diagram showing an example of the processing flow of the analysis device 20. In this processing flow, as an example of a reaction system, the analysis process for the combustion of methane in air will be performed and described below. When the analysis device 20 receives a processing request from the user via the input / output unit, it starts this process.
[0044] When the reaction pathway generation unit 211 receives a processing request from the user, it accepts the registration of reaction system information (step S1).
[0045] Specifically, the reaction pathway generation unit 211 transmits a predetermined reaction system information registration screen to the display unit 12 of the input / output device 10 via the I / F unit 23. Reaction system information is information required for molecular dynamics calculations, and the user inputs, for example, the reaction system composition, which is information about the composition of atoms and molecules included in the reaction system to be analyzed, and the setting conditions, which are information about the details of the reaction system. From the contents of the input screen, the reaction pathway generation unit 211 creates reaction system information 221 as shown in Figure 2 and stores it in the storage unit 22.
[0046] As shown in Figure 2, the user provides the following reaction system information: • Reaction system composition CH ４ :100 molecules O ２ :100 molecules N ２ Quantity: 100 molecules; Setting conditions: Ambient temperature: 1800K; Cell size: 20.0 nm ３ It will be assumed that this has been registered. Other setting conditions will be omitted.
[0047] The reaction pathway generation unit 211 performs molecular dynamics calculations based on the reaction system information 221 received in this manner and creates a predicted reaction pathway network (step S2).
[0048] Specifically, the reaction pathway generation unit 211 performs molecular dynamics calculations based on the reaction system information 221, maps the obtained physical property information onto a network, and generates a predicted reaction pathway network. In this example, as shown in Figure 3, the starting material is CH ４ The structure of the chemical species from the (raw material) through the intermediate to the final product is displayed as a reaction pathway. The amount of each product produced (yield %) may also be displayed on the pathway. The reaction pathway generation unit 211 stores such a predicted reaction pathway network as predicted reaction pathway network information 223 in the storage unit 22.
[0049] Next, the reaction pathway update unit 212 receives the registration of observed values and generates observed value data (step S3).
[0050] Specifically, the reaction pathway update unit 212 accepts input of actual observed values in the reaction system to be analyzed. The observed values may be entered by the user, for example, via a registration screen displayed on the display unit 12, or the observed values themselves or the file data after analysis may be uploaded. Based on such observed values, the reaction pathway update unit 212 generates observed value data 224, as shown in Figure 4, which includes at least the structure of the chemical species contained in the reaction system and the amount (yield) of each chemical species, and stores it in the storage unit 22.
[0051] The reaction pathway updating unit 212 calculates the difference in yield between the observed value and the estimated value for each chemical species present in the reaction system (step S4).
[0052] Specifically, the reaction pathway update unit 212 extracts each chemical species present in the reaction system from the estimated reaction pathway network information 223 and the observed value data 224, calculates the difference between the observed yield (observed yield) and the estimated yield (estimated yield), and generates a difference table as shown in Figure 5.
[0053] Furthermore, the reaction pathway updating unit 212 removes chemical species that exist only in the virtual environment (on the estimated values) from the estimated reaction pathway network (step S5).
[0054] Specifically, the reaction pathway update unit 212 first extracts chemical species that exist only on the predicted reaction pathway network. In other words, it only needs to identify chemical species on the difference table where "observed yield = 0% and predicted yield > 0%". In this example, (4) CH ３ O ２ H This applies. The reaction pathway update unit 212 then deletes the identified chemical species and the reaction pathway leading only to them, updates the predicted reaction pathway network information 223, and stores it in the storage unit 22 as updated reaction pathway network information 225.
[0055] The reaction pathway update unit 212 may also allow the user to decide whether or not to delete each reaction pathway. For example, a deletion selection screen, as shown in Figure 9, may be displayed on the display unit 12, and the deletion process is performed if the user selects to delete.
[0056] Subsequently, the reaction path update unit 212 extracts chemical species that exist only in the actual environment (observed values) and alternative chemical species candidates generated in the virtual environment (on the inferred reaction path network) instead of such chemical species (step S6).
[0057] Specifically, the reaction path update unit 212 first extracts "chemical species that exist only in the observed values". That is, it is only necessary to identify chemical species on the difference table where "inferred yield = 0%, observed yield > 0%". In this example, from the difference table shown in FIG. 5, (8) NO, (9) H ２ , (10) C ２ H ２ correspond to this.
[0058] Then, in order to create additional reaction paths leading to (8) NO, (9) H ２ , (10) C ２ H ２ , the reaction path update unit 212 extracts alternative chemical species that are considered to be generated instead of these. Here, as an example, the case of (10) C ２ H ２ will be described. The reaction path update unit 212 identifies chemical species whose constituent elements are at least partially common with (10) C ２ H ２ (including C or H), and "chemical species with an inferred yield greater than the observed yield" as candidates. According to the difference table in FIG. 5, those corresponding to these are (2) C ２ H ６ , (4) CH ３ O ２ H, (5) CH ２ O, (7) CO ２ . Therefore, the reaction path leading to (10) C ２ H ２ is considered to be replaced by any of these reaction paths.
[0059] Next, the reaction path update unit 212 identifies the chemical species to be updated based on the structural similarity from the candidates and adds a reaction path (step S7).
[0060] The reaction path update unit 212 compares (10) C ２ H ２ with (2) C ２ H ６(4) CH ３ O ２ H (5)CH ２ O (7)CO ２ Then, by comparing each of them based on structural similarity, for example, the one with the highest structural similarity is determined to be the substitute chemical species. In this example, from the structural similarity table shown in Figure 6, C ２ H ２ The one with the highest structural similarity is C ２ H ６ Therefore, the reaction pathway updating unit 212 is C ２ H ６ This chemical species is determined to be subject to renewal.
[0061] The reaction pathway update unit 212 may allow the user to select the chemical species to be updated. Specifically, the reaction pathway update unit 212 displays an update target selection screen, as shown in Figure 10, on the display unit 12, and generates a reaction pathway for the chemical species selected by the user.
[0062] The reaction pathway updating unit 212 is CH ４ (Raw material) from C ２ H ６ All chemical species other than the starting material, or the chemical species to be replaced and the chemical species located immediately before it (in this example, CH ３ and C ２ H ６ Starting from C ２ H ２ An additional reaction pathway network is generated to connect the two, leading to [the first step]. An example of a reaction pathway is shown in Figure 11, and a part of it is shown below. Figure 11 is a schematic diagram showing an example of an additional reaction pathway network. (a) C ２ H ６ → C ２ H ４ +H ２ → C ２ H ２ +2H ２ (b) C ２ H ６ +O → C ２ H ４ +H ２ +O → C ２ H ２ +H ２ O+H２ (c) CH ３ +CH ３ +2O → 2CH ２ +2OH → C ２ H ４ +2OH → C ２ H ３ +H ２ O + OH → C ２ H ２ +2H ２ O
[0063] The reaction path update unit 212 adds these additional reaction paths to the predicted reaction path network, updates the predicted reaction path network information 223, and stores it in the storage unit 22 as updated reaction path network information 225.
[0064] The user may decide whether or not to add each reaction pathway. For example, the reaction pathway update unit 212 displays a reaction pathway selection screen on the display unit 12, as shown in Figure 12. The user selects a pathway by referring to the changes in the difference change table when each reaction pathway is selected. In this example, since the difference is minimized when reaction pathway (b) is selected, (b) is assumed to be selected.
[0065] The reaction pathway update unit 212 generates a difference table again from the updated reaction pathway network included in the updated reaction pathway network information 225 (step S8). In this example, it is assumed that reaction pathway (b) is selected, and a table similar to the difference change table shown in Figure 12 is generated.
[0066] Next, the reaction pathway update unit 212 determines whether or not there are any chemical species that exist only in the observed values (step S9). If there are any (Y in step S9), the process returns to step S6 and repeats the reaction pathway addition process. If there are no such species (N in step S9), the process ends.
[0067] In this example, since NO has an estimated yield of 0% and an observed yield of 0%, the reaction pathway update unit 212 performs additional processing on NO. The reaction pathway for NO is N ２ +O ２ → N ２ O ２ → ON２ By adding O → 2NO, the presence of "chemical species that exist only in the observed values" is eliminated, and the process is terminated. The reaction pathway update unit 212 may output the updated reaction pathway network to the display unit 12 to present the final reaction pathway network to the user.
[0068] Embodiments of the present invention have been described in detail above. According to this embodiment, it is possible to obtain a reaction pathway network that is closer to the actual phenomenon by taking into account the changes that occur in the reaction in the actual environment in the calculated reaction pathway network. That is, by deleting reaction pathways that do not occur under observation and adding reaction pathways that occur only under observation, a more accurate reaction pathway network can be obtained. Furthermore, since the user can choose whether or not to update the network, the user can perform the simulation as intended.
[0069] Understanding reaction pathways with such high precision allows for the investigation of the causes of chemical changes such as combustion, degradation, and decomposition in products, as well as the control of these reactions, thereby ensuring safety and reliability. In particular, it provides useful insights for suppressing or promoting reactions during their progression. Furthermore, it can improve the efficiency of material design, contributing to the selection of more suitable materials. Applying such a system to manufacturing equipment enables efficient control of the manufacturing process.
[0070] The present invention is not limited to the embodiments described above, and various modifications are possible. For example, the embodiments described above are described in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace or add to the configurations of one embodiment with those of another embodiment.
[0071] 100...Reaction pathway analysis system, 10...Input / output device, 11...Input unit, 12...Display unit, 20...Analysis device, 21...Calculation unit, 22...Storage unit, 23...I / F unit, 211...Reaction pathway generation unit, 212...Reaction pathway update unit, 221...Reaction system information, 222...Condition setting information, 223...Predicted reaction pathway network information, 224...Observed value data, 225...Updated reaction pathway network information, 30...Network, 40...Server.
Claims
1. A reaction pathway analysis system comprising one or more processors and one or more memory resources, wherein the memory resources store estimated values in the reaction pathway network of a reaction system, and the processor performs the following processes: receiving input of actual observed values in the reaction system and storing them in the memory resources; reading out the estimated values and the observed values, comparing them and extracting the difference; identifying chemical species that exist only on the observed values from the difference; determining the structural similarity between the chemical species that exist only on the observed values and the chemical species that exist on the estimated values; generating a reaction pathway starting from the chemical species that exist on the estimated values and leading to the chemical species that exist only on the observed values based on the structural similarity, and updating the reaction pathway network; and outputting the updated reaction pathway network.
2. A reaction pathway analysis system according to claim 1, wherein the processor further performs the following: a process of identifying chemical species that exist only on the estimated values; and a process of deleting reaction pathways leading to chemical species that exist only on the estimated values and updating the reaction pathway network.
3. A reaction pathway analysis system according to claim 1, wherein the processor further performs a process of generating the reaction pathway network by molecular dynamics calculation and storing it in the memory resource.
4. A reaction pathway analysis system according to claim 1, characterized in that the chemical species present on the estimated starting point is the chemical species with the greatest structural similarity, or the chemical species located one step before the chemical species in the reaction pathway from the raw material to the chemical species with the greatest structural similarity.
5. A reaction pathway analysis system according to claim 1, characterized in that the chemical species present on the estimated starting value are selected preferentially from chemical species whose difference from the observed value of a chemical species present only on the observed value is larger.
6. A reaction pathway analysis system according to claim 1, wherein the processor causes the user to select the starting chemical species.
7. A reaction pathway analysis system according to claim 1, characterized in that the comparison between the observed value and the estimated value is based on the yield of the chemical species contained in each.
8. A reaction pathway analysis system according to claim 1, wherein the processor calculates the change in the difference between the estimated value and the observed value before and after the update, and performs the update so as to minimize the difference.
9. A reaction pathway analysis system according to claim 1, wherein the processor allows the user to select whether or not to perform the update.
10. A reaction pathway analysis system according to claim 1, wherein the processor uses at least one of AI, a database, or first-principles calculations to determine the validity of the update.
11. Apparatus for producing a substance, comprising the reaction pathway analysis system described in claim 1.
12. A reaction pathway analysis method using a reaction pathway analysis system having one or more processors and one or more memory resources, comprising: storing estimated values in the reaction pathway network of a reaction system; receiving and storing actual observed values in the reaction system; reading out and comparing the estimated values and the observed values and extracting the difference; identifying chemical species that exist only on the observed values from the difference; determining the structural similarity between the chemical species that exist only on the observed values and the chemical species that exist on the estimated values; generating a reaction pathway starting from the chemical species that exist on the estimated values and leading to the chemical species that exist only on the observed values based on the structural similarity, and updating the reaction pathway network; and outputting the updated reaction pathway network.
13. A program that causes a computer to function as a reaction pathway analysis device, the program causing the computer to function as a reaction pathway analysis device, which includes: a process of storing estimated values in the reaction pathway network of a reaction system; a process of receiving and storing actual observed values in the reaction system; a process of reading out and comparing the estimated values and the observed values and extracting the difference; a process of identifying chemical species that exist only in the observed values from the difference; a process of determining the structural similarity between the chemical species that exist only in the observed values and the chemical species that exist in the estimated values; a process of generating a reaction pathway starting from the chemical species that exist in the estimated values and leading to the chemical species that exist only in the observed values, based on the structural similarity, and updating the reaction pathway network; and a process of outputting the updated reaction pathway network.