Information processing apparatus, information processing method, and information processing program

Abstract

To induce a chemical reaction within a realistic computation time in a molecular-dynamics-based chemical reaction simulation.SOLUTION: An information processing apparatus according to an embodiment comprises at least one memory and at least one processor. The at least one processor identifies a plurality of target atoms that are subjects of chemical bonding from among a plurality of atoms, acquires information on a first acting force acting on the plurality of atoms, generated by inputting an atomic structure of the plurality of atoms into a neural network, acquires information on a first additional force to be applied to at least one of the plurality of target atoms, and executes a molecular-dynamics-based simulation for the plurality of atoms by using the information on the first acting force, the information on the first additional force, and positional information of the plurality of atoms.SELECTED DRAWING: Figure 3

Description

[Technical Field]

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

[0002] A neural network (hereafter referred to as NNP (Neural Network Potential)) is known that predicts the overall energy of an atomic state and the force acting on each atom. NNP can output energy and / or force in an extremely short time compared to simulations of electronic states such as density functional theory (DFT). NNP can perform general-purpose, highly accurate calculations for a variety of substances (a wide range of elements and structures). For example, because NNP is compatible with a variety of elements, it can handle chemical reactions.

[0003] Another similar technology to NNP is ReaxFF, which requires parameters to be determined for each type of chemical reaction and cannot simulate a wide range of substances.

[0004] On the other hand, even if molecular dynamics (MD) simulations are performed using NNP, chemical reactions take time, so chemical reactions are usually not observed. For this reason, it is difficult to dynamically simulate reaction processes within a realistic calculation time. For these reasons, it is difficult to directly handle chemical reactions such as polymerization and decomposition using MD simulations using NNP. Furthermore, simulations of chemical reactions using DFT take too long, so it is difficult to dynamically simulate reaction processes within a realistic calculation time. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Behrouz Arash, Barend J. Thijsse, Alessandro Pecenko, Angelo Simone, “Effect of water content on the thermal degradation of amorphous polyamide 6,6: A collective variable-driven hyperdynamics study” Polymer Degradation and Stability 146 (2017) 260-266 [Non-patent document 2] Aniruddh Vashisth, Chowdhury Ashraf, Weiwei Zhang, Charles E Bakis, Adri CT van Duin, “Accelerated ReaxFF simulations for Describing the Reactive Cross-Linking of Polymers” The Journal of Physical Chemistry A 2018 / 7 / 11, Vol. 122, No. 32, pp. 6633-6642, American Chemical Society,URL:https: / / scholar.google.co.jp / citations?view_op=view_citation&hl=ja&user=XAMQip0AAAAJ&citation_for_view=XAMQip0AAAAJ:4TOpqqG69KYC Summary of the Invention [Problem to be solved by the invention]

[0006] The problem to be solved by the present disclosure is to generate a chemical reaction in a molecular dynamics simulation within a realistic calculation time. [Means for solving the problem]

[0007] An information processing apparatus according to an embodiment includes at least one memory and at least one processor, which identifies target atoms among a plurality of atoms that are targets of chemical bonding, acquires information on first acting forces acting on the plurality of atoms generated by inputting the atomic structures of the plurality of atoms into a neural network potential, acquires information on a first additional force acting on at least one of the plurality of target atoms, and performs a molecular dynamics simulation on the plurality of atoms using the information on the first acting force, the information on the first additional force, and position information of the plurality of atoms. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing an example of a hardware configuration of an information processing apparatus according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of functional blocks in a processor according to the embodiment. [Figure 3] FIG. 3 is a flowchart showing an example of a procedure for the MD simulation execution process according to the embodiment. [Figure 4] FIG. 4 is a diagram showing an example of an initial structure arranged in a simulation space and equilibrated, a reference atom in a functional group in the initial structure, and a bondable atom according to the embodiment. [Figure 5] FIG. 5 is a diagram showing an example of a search range centered on the position of a specified reference atom and a plurality of atoms around the reference atom according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a plurality of different fixed boost potentials relative to interatomic distances according to the embodiment. [Figure 7] FIG. 7 is a diagram showing an example of a potential energy surface and a boost potential with respect to the interatomic distance between a reference atom and a bondable atom according to the embodiment. [Figure 8]FIG. 8 is a diagram showing an example of a plurality of boost potentials to be integrated with respect to interatomic distances, potential barriers, and boost application positions when a time-dependent boost potential is set as an acceleration condition according to the embodiment. [Figure 9] FIG. 9 is a diagram showing an example of a potential energy curved surface versus interatomic distance and an integrated boost potential when a time-dependent boost potential is set as an acceleration condition according to the embodiment. [Figure 10] FIG. 10 is a diagram showing an example of acceleration conditions set for a plurality of atom pairs related to one reference atom according to the embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of the added force corresponding to different fixed boost potentials in FIG. 6 according to the embodiment. [Figure 12] FIG. 12 is a schematic diagram illustrating an outline of the applied force in FIG. 5 according to the embodiment. [Figure 13] FIG. 13 is a diagram showing an example in which a chemical bond is formed between a reference atom and a bondable atom according to the embodiment. [Figure 14] FIG. 14 is a diagram showing an example of a neural network potential and a specific neural network according to an application example of the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, the embodiments will be described in detail with reference to the drawings.

[0010] (Embodiment) FIG. 1 is a block diagram showing an example of the hardware configuration of an information processing device 1 according to this embodiment. As shown in FIG. 1, the information processing device 1 may be connected to an external device 9A via a communication network 5. The information processing device 1 may also include an external device 9B connected via a device interface 39. The information processing device 1 may input a notation indicating the structure of a substance composed of multiple atoms input by a user. The substance is, for example, a molecule. Note that the substance is not limited to molecules and may be various crystals, etc.

[0011] The notation is, for example, SMILES (Simplified Molecular Input Line Entry System) notation related to the substance and input by the user. SMILES notation represents, for example, information about a specific molecule (information about atoms and how they are connected) according to certain rules. For example, SMILES notation represents granular information such as methane, where one C (carbon) is connected to four H (hydrogen). Note that the notation is not limited to SMILES notation, and other known notations may be used as long as they can uniquely identify the substance. Examples of other notations include SMARTS (SMiles ARbitrary Target Specification) notation. For the sake of concreteness, the information input by the user via an input device (described later) is assumed to be information corresponding to SMILES notation (hereinafter referred to as SMILES information).

[0012] The information processing device 1 includes a computer 30 and an external device 9B connected to the computer 30 via a device interface 39. The computer 30 includes, for example, a processor 31, a main storage device (memory) 33, an auxiliary storage device (memory) 35, a network interface 37, and a device interface 39. The information processing device 1 may be realized as the computer 30 in which the processor 31, the main storage device 33, the auxiliary storage device 35, the network interface 37, and the device interface 39 are connected via a bus 41.

[0013] Although the computer 30 shown in FIG. 1 includes one of each component, it may include multiple of the same component. Although FIG. 1 shows a single computer 30, the software may be installed on multiple computers, with each of the multiple computers executing the same or different parts of the software. In this case, a distributed computing configuration may be used in which the computers communicate with each other via a network interface 37 or the like to execute the processing. In other words, the information processing device 1 in this embodiment may be configured as a system in which one or more computers execute instructions stored in one or more storage devices to realize various functions described below. Furthermore, information transmitted from a terminal may be processed by one or more computers provided on the cloud, and the processing results may be transmitted to a terminal such as a display device (display unit) corresponding to the external device 9B.

[0014] Various computations of the information processing device 1 in this embodiment may be executed in parallel using one or more processors, or using multiple computers via a network. Furthermore, various computations may be distributed to multiple processor cores within a processor and executed in parallel. Furthermore, some or all of the processes, means, etc. disclosed herein may be executed by at least one of a processor and a storage device provided on a cloud that can communicate with the computer 30 via a network. Thus, various functions described below in this embodiment may be implemented in the form of parallel computing using one or more computers.

[0015] The processor 31 may be an electronic circuit (such as a processing circuit, processing circuitry, CPU (Central Processing Unit), GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), or ASIC (Application Specific Integrated Circuit)) including a control device and an arithmetic device of the computer 30. The processor 31 may also be a semiconductor device including a dedicated processing circuit. The processor 31 is not limited to an electronic circuit using electronic logic elements, but may also be realized by an optical circuit using optical logic elements. The processor 31 may also include an arithmetic function based on quantum computing.

[0016] The processor 31 performs arithmetic processing based on data and software (programs) input from each device, etc., configured internally of the computer 30, and can output the arithmetic results and control signals to each device, etc. The processor 31 may control each component constituting the computer 30 by executing the OS (Operating System) of the computer 30, applications, etc.

[0017] The information processing device 1 in this embodiment may be realized by one or more processors 31. Here, the processor 31 may refer to one or more electronic circuits arranged on one chip, or may refer to one or more electronic circuits arranged on two or more chips or two or more devices. When multiple electronic circuits are used, the electronic circuits may communicate with each other via wire or wirelessly.

[0018] The main memory device 33 is a memory device that stores instructions executed by the processor 31 and various data, and information stored in the main memory device 33 is read by the processor 31. The auxiliary memory device 35 is a memory device other than the main memory device 33. Note that these memory devices refer to any electronic component that can store electronic information, and may be semiconductor memory. The semiconductor memory may be either volatile memory or non-volatile memory. The memory device for saving various data used in the information processing device 1 according to this embodiment may be realized by the main memory device 33 or the auxiliary memory device 35, or may be realized by an internal memory built into the processor 31. For example, the memory unit in this embodiment may be realized by the main memory device 33 or the auxiliary memory device 35.

[0019] A plurality of processors may be connected (coupled) to one storage device (memory), or a single processor 31 may be connected. A plurality of storage devices (memories) may be connected (coupled) to one processor. When the information processing device 1 in this embodiment is configured with at least one storage device (memory) and a plurality of processors connected (coupled) to this at least one storage device (memory), it may include a configuration in which at least one of the plurality of processors is connected (coupled) to at least one storage device (memory). This configuration may also be realized by storage devices (memories) and processors 31 included in a plurality of computers. Furthermore, it may include a configuration in which a storage device (memory) is integrated with the processor 31 (for example, a cache memory including an L1 cache and an L2 cache).

[0020] The network interface 37 is an interface for connecting to the communication network 5 wirelessly or via a wire. The network interface 37 may be an appropriate interface, such as one that conforms to an existing communication standard. Information may be exchanged with an external device 9A connected via the communication network 5 through the network interface 37. The communication network 5 may be any one of a WAN (Wide Area Network), a LAN (Local Area Network), a PAN (Personal Area Network), etc., or a combination thereof, as long as information is exchanged between the computer 30 and the external device 9A. An example of a WAN is the Internet, an example of a LAN is IEEE802.11 or Ethernet (registered trademark), and an example of a PAN is Bluetooth (registered trademark) or NFC (Near Field Communication), etc.

[0021] The device interface 39 is an interface such as a USB (Universal Serial Bus) that directly connects to an output device such as a display device, an input device, and an external device 9 B. The output device may also have a speaker that outputs sound and the like.

[0022] The external device 9A is a device connected to the computer 30 via a network. The external device 9B is a device connected directly to the computer 30.

[0023] The external device 9A or the external device 9B may be, for example, an input device (input unit). The input device is, for example, a device such as a camera, a microphone, a motion capture device, various sensors, a keyboard, a mouse, or a touch panel, and provides acquired information to the computer 30. The external device 9A or the external device 9B may also be a device equipped with an input unit, a memory, and a processor, such as a personal computer, a tablet terminal, or a smartphone.

[0024] Furthermore, the external device 9A or the external device 9B may be, for example, an output device (output unit). The output device may be, for example, a display device (display unit) such as an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube), a PDP (Plasma Display Panel), or an organic EL (Electro Luminescence) panel, or may be a speaker that outputs sound or the like. Furthermore, the external device 9A or the external device 9B may be a device that includes an output device, a memory, and a processor, such as a personal computer, a tablet terminal, or a smartphone.

[0025] Furthermore, the external device 9A or the external device 9B may be a storage device (memory). For example, the external device 9A may be a network storage or the like, and the external device 9B may be a storage such as an HDD.

[0026] Furthermore, the external device 9A or the external device 9B may be a device having some of the functions of the components of the information processing device 1 in this embodiment. In other words, the computer 30 may transmit or receive some or all of the processing results of the external device 9A or the external device 9B.

[0027] 2 is a diagram showing an example of functional blocks realized by one or more processors 31. The processor 31 has, as functions realized by the processor 31, for example, a setting unit 311, an identification unit 313, an acting force determination unit 315, an applied force determination unit 317, a molecular dynamics (MD) simulation unit 319, a determination unit 321, and an evaluation unit 323. The functions realized by the setting unit 311, the identification unit 313, the acting force determination unit 315, the applied force determination unit 317, the MD simulation unit 319, the determination unit 321, and the evaluation unit 323 are each stored as a program in, for example, the main storage device 33 or the auxiliary storage device 35. The processor 31 can realize the functions related to the setting unit 311, the identification unit 313, the acting force determination unit 315, the added force determination unit 317, the MD simulation unit 319, the judgment unit 321, and the evaluation unit 323 by reading and executing each program stored in the main memory unit 33 or the auxiliary memory unit 35, etc.

[0028] The setting unit 311 sets various conditions related to the MD simulation (hereinafter referred to as simulation conditions). For example, the setting unit 311 sets the simulation conditions based on a user instruction input via an input device or by reading simulation conditions from the main storage device 33 and / or the auxiliary storage device 35. The setting unit 311 stores the set simulation conditions in the main storage device 33 and / or the auxiliary storage device 35.

[0029] The simulation conditions include, for example, search conditions for searching for multiple target atoms that are the target of chemical bonding, acceleration conditions for accelerating chemical bonding, the temperature and pressure in the virtual space (simulation space) where the MD simulation is performed, and the composition in the virtual space where the MD simulation is performed (e.g., type of monomer, type of solvent, type of starting material, composition ratio, etc.).

[0030] The search conditions are, for example, a search range for searching for multiple target atoms. The search range may be set by a user or may be set in advance. The search range corresponds to, for example, a region between the minimum and maximum distances from a reference atom that serves as a reference for chemical bonding among the multiple target atoms. In this case, the search conditions correspond to the two distances, the minimum and maximum distances.

[0031] For the sake of specificity, the following description will be given assuming that the multiple target atoms are a pair (atom pair) of a reference atom and one atom (hereinafter referred to as a bondable atom) that can form a chemical bond with the reference atom. The bondable atom corresponds to an atom that has the potential to react with the reference atom. In other words, the bondable atom corresponds to an atom that is likely to react with the reference atom. Note that the multiple target atoms will be described as appropriate when they have a reference atom and multiple bondable atoms with the reference atom.

[0032] The acceleration condition is, for example, a condition that specifies a bulk energy (hereinafter referred to as a boost potential) that promotes chemical bonding between multiple target atoms. For example, when a time-fixed boost potential (hereinafter referred to as a fixed boost potential) is applied to a bondable atom, the acceleration condition corresponds to a parameter that specifies the fixed boost potential. The parameter that specifies the fixed boost potential corresponds to the strength of the energy (potential) and the range of influence of the energy (potential). Specifically, when the fixed boost potential is expressed by a Gaussian function, the strength of the boost potential corresponds to the maximum value of the Gaussian function, and the range of influence of the boost potential corresponds to the half-width (full width at half maximum or half width at half maximum) of the Gaussian function. In other words, the fixed boost potential corresponds to a boost potential that is independent of time. Note that the shape of the boost potential is not limited to a Gaussian function that specifies a Gaussian potential, and any function such as a Morse potential or a Lennard-Jones potential may be used, or the boost potential may be specified by any function expressed by a neural network, etc.

[0033] Furthermore, when the boost potential applied to the atom pair is a time-dependent boost potential (hereinafter referred to as a time-dependent boost potential), the acceleration condition corresponds to a parameter that defines the time-dependent boost potential applied per unit time. The parameters that define the time-dependent boost potential include the strength of the energy (potential), the range of influence of the energy (potential) applied per unit time, and the frequency of applying the energy (potential). Furthermore, the shape of the boost potential may be defined by a shape obtained by inverting a probability distribution that indicates the existence probability according to the coordinates of bondable atoms according to a Boltzmann distribution or the like.

[0034] When the time-dependent boost potential is for multiple target atoms (atom population), i.e., when the multiple target atoms have a reference atom and multiple bondable atoms, and the boost potential applied to the atomic population composed of multiple bondable atoms is a time-dependent boost potential, the acceleration conditions correspond to the strength of the boost potential, the range of influence of the boost potential, the shape of the activation function, the degree of weighting for each of the multiple population variables, etc. The shape of the activation function and the degree of weighting will be explained later.

[0035] Prior to the execution of an MD simulation, the setting unit 311 arranges a plurality of atoms (molecules) in a virtual space in which the MD simulation is to be executed. By arranging a plurality of atoms in the virtual space, the setting unit 311 sets an initial structure of the simulation target. The setting unit 311 stores the set initial structure in the main storage device 33 and / or the auxiliary storage device 35.

[0036] The setting unit 311 performs equilibration on the initial structure. Specifically, the setting unit 311 moves the positions of multiple molecules and / or multiple atoms in the virtual space so that the states of multiple molecules and / or multiple atoms included in the initial structure become mechanically and thermally stable (hereinafter referred to as metastable states) according to the temperature, pressure, and other simulation conditions set for the virtual space having the initial structure. Since known processes can be applied for the equilibration process, a description thereof will be omitted. The setting unit 311 stores the positions of the multiple molecules and / or multiple atoms after equilibration in the simulation space (hereinafter referred to as initial positions) in the main storage device 33 and / or the auxiliary storage device 35.

[0037] The identification unit 313 identifies multiple target atoms that are targets of chemical bonding among multiple atoms. Specifically, the identification unit 313 identifies multiple target atoms that are targets of chemical bonding among multiple atoms arranged in a virtual space related to the MD simulation. The multiple target atoms correspond to atoms that are susceptible to chemical reactions. Identifying the multiple target atoms corresponds to listing atoms that are susceptible to chemical reactions. For example, in response to a user instruction via the input interface, the identification unit 313 identifies multiple atoms that are susceptible to chemical bonding (chemical reaction) among multiple atoms arranged in the virtual space. An atom that is susceptible to chemical bonding (chemical reaction) is, for example, an atom contained in a functional group such as a radical or vinyl carbon, and corresponds to a reference atom that serves as a reference for chemical bonding. For example, the identification unit 313 identifies a site that accelerates bond formation / decomposition by functional group in accordance with a user instruction via the input interface.

[0038] The identification unit 313 searches for atoms included in the search range in a virtual space related to the MD simulation, with each of the identified atoms (reference atoms) as the center. For each of the multiple reference atoms, the identification unit 313 identifies atoms included in the search range as bondable atoms. Note that for each of the multiple reference atoms, atoms located closer than the minimum distance in the search range are likely to already be chemically bonded to the reference atom, and therefore, these atoms may be excluded from the search for bondable atoms. The identification unit 313 associates the bondable atoms identified in the search with the reference atom that serves as the reference for the search. In this way, the identification unit 313 identifies atom pairs consisting of the reference atom and the bondable atom. The identification unit 313 stores the identified multiple atom pairs, i.e., multiple target atoms, in the main memory device 33 and / or the auxiliary memory device 35. At this time, the reference atom and the bondable atoms are associated using an index that distinguishes the atoms, and stored in the main memory device 33 and / or the auxiliary memory device 35.

[0039] When the determination unit 321 (described later) determines that a chemical bond has been formed and subsequently initializes the sum of the boost potentials, the identification unit 313 updates the attribute information of the atom with which the chemical bond has been formed based on the functional groups of the multiple atoms with which the chemical bond has been formed. The attribute information of the atom with which the chemical bond has been formed is attribute information that indicates whether the multiple atoms included in the molecule determined to have formed a chemical bond are susceptible to chemical reaction. This updates the reference atom for the formed chemical bond. Next, the identification unit 313 applies the center of the search range to the position of the updated reference atom to identify at least one atom that can bond with the reference atom and that can form a chemical bond with the reference atom. In this way, the identification unit 313 updates the multiple target atoms. The updating of the reference atom in the multiple atoms with which the chemical bond has been formed is performed by a known program or the like that is set in advance.

[0040] Furthermore, when the determination unit 321 (described later) determines that a chemical bond has not been formed and also determines that the MD simulation unit 319 (described later) has executed the MD simulation for a predetermined time or a predetermined number of times, the identification unit 313 identifies a reference atom (or an index of the reference atom) that is not related to the application of the boost potential. At this time, the applied force determination unit 317 (described later) executes processing for the identified reference atom. The predetermined time and the predetermined number of times are set in advance by the setting unit 311 according to instructions from a user or the like, and are stored in the main storage device 33 and / or the auxiliary storage device 35.

[0041] The applied force determination unit 315 acquires information on first applied forces (hereinafter referred to as applied forces or first applied forces) acting on multiple atoms, which is generated by inputting the atomic structures of the multiple atoms into a trained neural network potential (hereinafter referred to as NNP). For example, the applied force determination unit 315 inputs the atomic structures of the multiple atoms into a trained neural network to generate information on the first applied forces acting on the multiple atoms. The information on the first applied force conceptually includes, for example, the value of the first applied force itself and / or information necessary for determining the first applied force. Specifically, the applied force determination unit 315 inputs the positions of the multiple atoms arranged in a virtual space and information on the types of the multiple atoms (atomic structures) into the trained NNP to determine the first applied forces for each of the multiple atoms. The atomic structure includes, for example, information on the types of the multiple atoms and position information of the multiple atoms. The position information of the atoms includes, for example, the coordinates of the atoms. Note that the position information of the atoms may be in any format as long as it is information on the positions of the atoms. The NNP is realized, for example, by a highly versatile trained graph neural network that can generate accurate energy values ​​and forces acting on each of multiple atoms for various atomic structures. Since any known neural network can be used as the trained graph neural network, a detailed description thereof will be omitted. The acting force determination unit 315 stores multiple acting forces acting on each of multiple atoms arranged in a virtual space in the main memory device 33 and / or the auxiliary memory device 35.

[0042] Furthermore, when the determination unit 321 (described later) determines that a chemical bond has not been formed and also determines that the MD simulation has not been performed for a predetermined time or a predetermined number of times, the force determination unit 315 acquires information about the second force acting on the plurality of atoms, which is generated by inputting the atomic structures of the plurality of atoms into the NNP. The information about the second force conceptually includes, for example, the value of the second force itself and / or information necessary for determining the second force. For example, the force determination unit 315 determines the second force acting on each of the plurality of atoms by inputting the positions of the plurality of target atoms and the types of the plurality of atoms after (immediately after) the MD simulation has been performed into the NNP. In other words, the force determination unit 315 sequentially calculates the force acting on each of the plurality of atoms moved by the MD simulation each time the MD simulation is performed. The force determination unit 315 may also be referred to as a force acquisition unit.

[0043] The additional force determination unit 317 acquires information about a first additional force to be applied to at least one of the multiple target atoms. Specifically, the additional force determination unit 317 applies a boost potential to each of the multiple target atoms using the coordinates of each of the multiple target atoms. For example, the additional force determination unit 317 applies a boost potential based on the position of a reference atom having a selected index from the multiple reference atoms and the position of a bondable atom corresponding to the selected reference atom. For example, if the boost potential is expressed as a Gaussian function and the multiple target atoms are an atom pair, the boost potential is applied to the position of the bondable atom so that the maximum value of the Gaussian function is located at the position of the reference atom and the position of the bondable atom. Note that the additional force determination unit 317 may apply an additional force to only one of the multiple target atoms, rather than to both of the multiple target atoms. The information about the first additional force may conceptually include, for example, the value of the first additional force itself and / or information necessary for determining the first additional force.

[0044] The additional force determination unit 317 determines a force (hereinafter referred to as an additional force or a first additional force) to be applied to each of the target atoms based on the coordinates of the target atoms and the boost potential. For example, the additional force determination unit 317 determines information about the first additional force based on the position information of the target atoms. Specifically, the additional force determination unit 317 determines information about the first additional force based on the position information of the target atoms and the boost potential. The information about the first additional force is not limited to the value of the first additional force itself, but may also be information necessary for determining the first additional force. The information about the first additional force may also be information about a value by which the first additional force is multiplied. Specifically, when the first additional force generated by NNP is in the desired direction and / or in the direction in which the reaction proceeds, the information about the first additional force may be expressed by, for example, multiplying the first additional force calculated by NNP by 1.5, rather than adding the additional forces as a vector. Alternatively, the information on the first additional force may be information that the value of the first additional force is zero (no force is applied). For example, when the distance between the plurality of target atoms is smaller than a predetermined threshold, the additional force determination unit 317 determines, as the information on the first additional force, that the first additional force is not to be applied to the plurality of target atoms.

[0045] The calculation of the additional force applied to the plurality of target atoms (e.g., atom pairs) from the boost potential can be realized by a known method, such as analytical calculation based on the boost potential and the positions of the plurality of target atoms (e.g., atom pairs), and therefore a description thereof will be omitted. The additional force determination unit 317 stores the additional force applied to each of the plurality of target atoms (e.g., atom pairs) in the main storage device 33 and / or the auxiliary storage device 35.

[0046] The additional force determination unit 317 may determine information about the first additional force without using the boost potential. For example, the first additional force may be a fixed value. In this case, the additional force determination unit 317 may determine the fixed value based on position information between atoms and distance information between atoms. The additional force determination unit 317 may also be configured to add the fixed value each time the MD simulation is repeated.

[0047] Furthermore, if the judgment unit 321 determines that a chemical bond has not been formed and also determines that the MD simulation has not been performed for a predetermined time or a predetermined number of times, the additional force determination unit 317 may further impart a boost potential to the position of the atom pair before movement (e.g., the position of the bondable atom). At this time, the additional force determination unit 317 again imparts a boost potential to the position of the bondable atom. That is, if it continues to be determined that a chemical bond has not been formed in multiple MD simulations performed for a predetermined time or a predetermined number of times, the additional force determination unit 317 accumulates the boost potentials that have been applied a predetermined number of times according to the past positions of the paired atoms. Note that the accumulation of the boost potential may be performed by the setting unit 311. That is, if the judgment unit 321 determines that a chemical bond has not been formed between multiple target atoms, the MD simulation unit 319 executes another MD simulation using further information on the first additional force. The further use of information on the first additional force corresponds to, for example, accumulating the boost potential. Note that the meaning of "performing an MD simulation using XX and YY" is not limited to directly using XX and YY to perform a simulation, but also includes performing a simulation using ZZ generated based on XX and YY.

[0048] For example, if it is determined that no chemical bond has been formed, the additional force determination unit 317 calculates the sum of the boost potential for the first additional force determined before the MD simulation was performed and the boost potential to be applied to the coordinates of multiple target atoms (e.g., atom pairs) after the MD simulation was performed. The additional force determination unit 317 then determines an additional force (second additional force) to be applied to each of the multiple target atoms based on the coordinates of the multiple target atoms after the MD simulation was performed and the sum of the boost potentials. That is, the additional force determination unit 317 acquires information about the second additional force to be applied to at least one of the multiple target atoms. The information about the second additional force conceptually includes, for example, the value of the second additional force itself and / or information necessary for determining the second additional force.

[0049] Note that if the distance between the atom pair, i.e., the distance between the reference atom and the bondable atom, is less than a predetermined distance (which may be referred to as, for example, the minimum distance of the search range or a predetermined threshold), the additional force determination unit 317 may not impart a boost potential to the position of the bondable atom. Furthermore, the additional force determination unit 317 may set a potential wall (potential barrier) greater than the maximum value of the boost potential at a position the maximum distance away from the reference atom. The additional force determination unit 317 may be referred to as an additional force acquisition unit.

[0050] The MD simulation unit 319 performs a molecular dynamics simulation on a plurality of atoms using information on the first action force, information on the first added force, and position information on the plurality of atoms. Specifically, the MD simulation unit 319 first calculates the sum of the action force determined by the action force determination unit 315 and the added force determined by the added force determination unit 317 before the MD simulation is performed. For example, the MD simulation unit 319 calculates the sum of the first action force and the first added force. Next, the MD simulation unit 319 performs an MD simulation on the plurality of atoms for a predetermined short period of time using the sum of the first action force and the first added force and the positions of the plurality of atoms. That is, the MD simulation unit 319 performs an MD simulation on the plurality of atoms using the sum of the first action force and the first added force and the position information on the plurality of atoms.

[0051] Furthermore, after the initialization of the boost potential, the MD simulation unit 319 calculates the sum of the second acting force and the second added force. Next, the MD simulation unit 319 executes an MD simulation on the plurality of atoms for a preset short time using the sum of the second acting force and the second added force and the positions of the plurality of atoms. As a result, the MD simulation unit 319 executes an MD simulation on the plurality of atoms again using information on the second acting force, information on the second added force, and position information of the plurality of atoms after the MD simulation. Since known methods can be applied to the processing procedure of the MD simulation, a description thereof will be omitted. The MD simulation unit 319 stores the positions of the plurality of atoms after movement as a result of the MD simulation in the main storage device 33 and / or the auxiliary storage device 35 together with the measurement time of the MD simulation.

[0052] The determination unit 321 determines whether or not a chemical bond is formed based on the positional information of multiple target atoms after the MD simulation is performed. For example, the determination unit 321 reads, for example, determination conditions for determining whether or not a chemical bond is formed (hereinafter referred to as bond determination conditions) from the main storage device 33 and / or the auxiliary storage device 35. The bond determination conditions are, for example, a correspondence table indicating the distance between atoms (hereinafter referred to as bond determination distance) for each type of bond, such as a covalent bond. Note that the bond determination distance may be a bond distance (e.g., a covalent bond distance for a covalent bond) set according to the type of bond, plus a predetermined margin. The margin may be set in advance by a user's instruction via an input interface and / or by the setting unit 311, etc. The correspondence table is set in advance using a chemical database or the like and stored in the main storage device 33 and / or the auxiliary storage device 35.

[0053] The determination unit 321 calculates the distance between the reference atom and the bondable atom (hereinafter referred to as the interatomic distance). The determination unit 321 also identifies the type of bond between the reference atom and the bondable atom. The determination of the type of bond is based on a known method for identifying the type of functional group formed by the reference atom and the bondable atom. Therefore, a description thereof will be omitted. The determination unit 321 identifies the bond judgment distance from the correspondence table by comparing the identified bond type with the correspondence table. Next, the determination unit 321 compares the interatomic distance with the bond judgment distance to determine whether or not a chemical bond exists between the reference atom and the bondable atom. Specifically, if the interatomic distance is equal to or less than the bond judgment distance, the determination unit 321 determines that the reference atom and the bondable atom are chemically bonded. On the other hand, if the interatomic distance exceeds the bond judgment distance, the determination unit 321 determines that the reference atom and the bondable atom are not chemically bonded.

[0054] Furthermore, if it is determined that a chemical bond has not been formed, the determination unit 321 determines whether the MD simulation has been executed for a predetermined time or a predetermined number of times. For example, if the sum of the infinitesimal times related to the execution of the MD simulation is less than a predetermined time, a boost potential is again applied to the bondable atom to which a boost potential was applied before the MD simulation was executed. Furthermore, if it is determined that a chemical bond has not been formed and the sum of the infinitesimal times (the sum of the execution times of the MD simulation related to the reference atom) is equal to or greater than a predetermined time, the determination unit 321 initializes (sets to 0) the sum of the boost potentials applied to the multiple target atoms. Note that the initialization of the sum of the boost potentials may be executed by the identification unit 313 or the applied force determination unit 317.

[0055] After the identifying unit 313 updates the attribute information, the determining unit 321 determines whether the number of atoms included in the attribute of the specified atom (i.e., the updated attribute) is equal to or less than a certain number. The certain number may be set in advance or may be set by the setting unit 311. For example, if the attribute of the specified atom is a vinyl group and there are initially 100 carbon atoms in the vinyl group (before the MD simulation is performed), the number of vinyl groups gradually decreases as chemical bonds are formed by the MD simulation. When the number of carbon atoms in the vinyl group reaches zero, the reaction no longer progresses, and the MD simulation proceeds to the next step, for example, a process of identifying multiple target atoms to be chemically bonded by the identifying unit 313. The certain number may be, for example, the above-mentioned zero, but is not limited thereto. For example, the MD simulation may be terminated when the number of atoms to be attributed reaches 50% of the initial value (in the above example, 50 carbon atoms in the vinyl group). In other words, the certain number may be the number of atoms involved in a chemical reaction in a functional group involved in a chemical reaction, or the ratio of the number of reacted atoms to the total number of atoms involved in the chemical reaction. If the number of atoms included in the specified atom attribute is less than a certain number, evaluation is performed on the molecules produced by the chemical reaction.

[0056] The evaluation unit 323 evaluates the physical properties of molecules (e.g., macromolecules such as polymers) generated by the MD simulation. The physical properties include, for example, the thermal properties (e.g., heat resistance) and mechanical properties (e.g., tensile properties) of the generated molecules. The evaluation unit 323 may also evaluate the length distribution of the polymer. Since known methods can be applied to evaluate the physical properties of molecules, a description thereof will be omitted. The evaluation unit 323 stores the evaluated physical properties of the molecules in the main storage device 33 and / or the auxiliary storage device 35. The evaluation unit 323 may also display the evaluated molecular properties on a display or the like.

[0057] The above has described the configuration of the information processing device 1. Below, the procedure of the MD simulation processing executed by the information processing device 1 (hereinafter referred to as MD simulation execution processing) will be described with reference to FIG.

[0058] FIG. 3 is a flowchart showing an example of a procedure for the MD simulation execution process.

[0059] (MD simulation execution process) (Step S301) The setting unit 311 sets the simulation conditions. The setting of the simulation conditions is not limited to reading the simulation conditions from the main storage device 33 and / or the auxiliary storage device 35, but may also be set by a user's instruction via an input interface. In this case, the simulation conditions may be adjusted as appropriate by the user.

[0060] (Step S302) The setting unit 311 arranges a plurality of atoms in a simulation space to set an initial structure. For example, the setting unit 311 sets the initial structure by randomly arranging a plurality of atoms in the simulation space. The simulation space can be set to any shape, such as a rectangular parallelepiped, a cube, or a sphere.

[0061] (Step S303) The setting unit 311 performs equilibration on the plurality of atoms. For example, the setting unit 311 achieves equilibration by moving the plurality of atoms arranged in the simulation space so that the plurality of atoms are in a metastable state.

[0062] (Step S304) The identification unit 313 identifies a plurality of target atoms that are targets for chemical bonding among a plurality of atoms arranged in the simulation space. That is, the identification unit 313 identifies an atom that is likely to undergo a chemical reaction among the plurality of atoms as a reference atom. For example, a user's instruction via the input interface specifies, in functional group units, a site (a plurality of atoms) that accelerates a chemical reaction such as bond formation or decomposition. At this time, the identification unit 313 identifies, in the specified functional group, a reference atom that serves as a reference for the chemical reaction.

[0063] The identifying unit 313 may include a process of excluding pairs of atoms that already have chemical bonds when identifying multiple target atoms that are targets of chemical bonding. The most typical process performed by the identifying unit 313 in this step is to prepare a list of atoms in advance and update the list as bonds are formed or eliminated. For example, if a reaction occurs between two atoms in different atom groups A={a1, a2, a3,...} and atom group B={b1, b2, b3,...} (e.g., a1 + b2 → a1 - b2), when the reaction occurs, the corresponding atoms (a1, a2) involved in the reaction are deleted from atom groups A and B. In this case, the identifying unit 313 may determine whether a chemical bond has occurred between atoms based on the distance between the atoms.

[0064] The method for identifying the target atom is not limited to the above. For example, the identification unit 313 may identify the atom's attribute information based on bond information generated from the distances between each atom, and identify the target atom using a general reaction rule. For example, by inputting SMILES notation, the identification unit 313 can identify a carbon radical having only three bonds and one unpaired electron. Similarly, by inputting SMILES notation, the identification unit 313 can identify a vinyl carbon having a carbon atom connected by a double bond. A general reaction rule corresponds to, for example, the knowledge that a carbon radical reacts with a vinyl carbon. By storing such major reaction rules in the main memory device 33 and / or the auxiliary memory device 35 in advance, the identification unit 313 can identify the target atom by comparing the rules with the atom's attribute information.

[0065] The identifying unit 313 identifies, as a bondable atom, an atom that is closest to the reference atom among the atoms included in a search range centered on the position of the identified reference atom. Note that the identifying unit 313 may identify, as a bondable atom, a predetermined number of atoms included in the search range in order of proximity to the reference atom.

[0066] 4 is a diagram showing an example of an initial structure INS placed in a simulation space SS and equilibrated, and a reference atom SA and a bondable atom JA in a functional group FG in the initial structure INS. As shown in FIG. 4, the identification unit 313 identifies an atom pair of a reference atom SA and a bondable atom JA that are targets for chemical bonding from among the multiple atoms included in the initial structure INS placed in the simulation space SS.

[0067] 5 is a diagram showing an example of a search range SR centered on the position of the identified reference atom SA, and a plurality of atoms around the reference atom JA. For example, the identification unit 313 identifies the atom that is included in the search range SR and is closest to the reference atom SA as the bondable atom JA, as shown in FIG. 5. The identification unit 313 performs the search shown in FIG. 5 for each of the identified reference atoms, thereby identifying a plurality of atom pairs corresponding to a plurality of target atoms.

[0068] (Step S305) The applied force determination unit 315 inputs information on the types of atoms and the positions of the atoms (atomic structure) to the NNP. Based on the output from the NNP, the applied force determination unit 315 determines an applied force (first applied force) by position differentiation using backward processing. The applied force may be determined using an NNP that outputs a force directly from the output layer without using backward processing. The applied force determination unit 315 stores the determined applied force in the main storage device 33 and / or the auxiliary storage device 35. For example, if no chemical bond is formed with the identified reference atom and the MD simulation has not been performed for a predetermined period of time, the applied force determination unit 315 updates the applied force by inputting the types of atoms and the positions of the atoms moved by the MD simulation to the NNP each time the MD simulation is performed.

[0069] Furthermore, for example, if a chemical bond is formed with the identified reference atom and the number of atoms included in the attribute of the specified atom is not equal to or less than a certain number, the applied force determination unit 315 inputs the types and positions of the multiple atoms into the NNP, and determines an applied force (second applied force) by position differentiation using backward processing. The applied force determination unit 315 stores the determined applied force in the main memory device 33 and / or the auxiliary memory device 35.

[0070] (Step S306) The setting unit 311 sets the boost potential. For example, when setting a fixed boost potential expressed by a Gaussian function as an acceleration condition, the setting unit 311 sets the maximum value of the Gaussian function corresponding to the maximum energy of the boost potential and the half-width of the Gaussian function corresponding to the range affected by the energy. The setting unit 311 sets the position of the fixed boost potential based on the positions of the reference atom and the bondable atom in the atom pair. The setting unit 311 stores the set fixed boost potential and the position to which the fixed boost potential is applied in the main storage device 33 and / or the auxiliary storage device 35. When a fixed boost potential is set, the boost potential remains constant in the repetition of the MD simulation for the identified reference atom. In this case, if the reference atom is changed, the fixed boost potential is set again for the changed reference atom.

[0071] FIG. 6 is a diagram showing an example of a plurality of different fixed boost potentials with respect to interatomic distances. Although three fixed boost potentials are shown in FIG. 6, the present invention is not limited to this. The shape of the fixed boost potential can be set arbitrarily depending on the setting of the parameters that define the fixed boost potential. Fixed boost potential E rest is defined by, for example, the following equation (1).

[0072]

number

[0073] F1 on the right side of formula (1) indicates the strength of the boost potential. F2 on the right side of formula (1) indicates the distance that the boost potential affects. R on the right side of formula (1) 12 indicates the equilibrium position of the reference atom and the bondable atom. ij indicates the distance between the reference atom and the bondable atom in the equilibrated initial structure INS.

[0074] FIG. 7 is a diagram showing an example of a potential energy surface (PES) for the interatomic distance between a reference atom and a bondable atom (referred to as a reaction coordinate or collective variable) and a boost potential (BP). The potential energy surface (PES) represents the distribution of potential energy for the reaction coordinate or collective variable (e.g., an energy function for the interatomic distance). As shown in FIG. 7, the potential energy surface (PES) corresponding to the interatomic distance after the chemical reaction (hereinafter referred to as the post-reaction position) (ARD) corresponds to the potential energy (ARP) after the chemical reaction and has a minimum value. On the other hand, as shown in FIG. 7, the potential energy surface (PES) corresponding to the interatomic distance before the chemical reaction (hereinafter referred to as the pre-reaction position) (BRD) corresponds to the potential energy (BRP) before the chemical reaction and has a minimum value.

[0075] As shown in FIG. 7, the boost potential BP corresponds to, for example, potential energy added to the minimum value of the potential energy BRP. For example, the setting unit 311 sets the boost potential BP so that the maximum value of the boost potential BP is located at the pre-reaction position BRD corresponding to the minimum value BRP of the potential energy surface PES. Note that the setting position of the maximum value of the boost potential BP is not limited to the pre-reaction position BRD. For example, the boost potential BP may be set so that the maximum value of the boost potential BP is located at a position a predetermined distance away from the pre-reaction position BRD.

[0076] Furthermore, the setting unit 311 may set a potential barrier at the maximum distance position of the search range SR for the reference atom SA in addition to the boost potential BP. Furthermore, the setting unit 311 may set in advance so that the boost potential BP is not applied to a position less than the minimum distance of the search range SR for the reference atom SA (hereinafter referred to as a non-boosting range).

[0077] Furthermore, when setting a time-dependent boost potential as an acceleration condition, the setting unit 311 sets the maximum value of the Gaussian function, the half-width of the Gaussian function, the frequency of applying the time-dependent boost potential, and the like. The frequency can be set arbitrarily, such as per unit time or per execution count of the MD simulation. The setting unit 311 sets the boost potential to be applied to the bondable atom corresponding to the identified reference atom, for example, by accumulating it for each execution of the MD simulation (for example, the sum of Gaussian potentials according to the number of execution counts of the MD simulation). The setting unit 311 stores the accumulated time-dependent boost potential in the main storage device 33 and / or the auxiliary storage device 35. The accumulation of the time-dependent boost potential may be performed in step S307, which will be described later. The accumulation of the time-dependent boost potential may also be performed by the applied force determination unit 317 in this step.

[0078] 8 is a diagram showing an example of multiple boost potentials PBP integrated with respect to interatomic distances, potential barriers PB, and boost application positions when a time-dependent boost potential is set as an acceleration condition. As shown in FIG. 8, multiple time-dependent boost potentials are set, for example, for each MD simulation according to a set frequency. Also, as shown in FIG. 8, the potential barrier PB is set, for example, at the end of the maximum distance in the search range SR so that the bondable atom does not move away from the reference atom. Also, as shown in FIG. 8, the boost-disabled range BNPB is set at the end of the minimum distance in the search range SR. The maximum interatomic distance in the boost-disabled range BNPB (the minimum distance in FIG. 8) corresponds to the cutoff of the boost potential.

[0079] As shown in FIG. 7, when the interatomic distance is located in a predetermined range (hereinafter referred to as the out-barrier range) BOR away from the potential barrier, the setting unit 311 may add a boost potential BP to the potential energy surface PES. That is, in the boost-unavailable range BNPB, the setting unit 311 does not need to add a boost potential BP to the potential energy surface PES. The out-barrier range BOR may be set to an interatomic distance greater than a predetermined position PP away from the pre-reaction position BRD. The predetermined position PP can be set, for example, using an empirical value based on the target atom. Furthermore, the setting unit 311 may add multiple boost potentials PBP shown in FIG. 8 to bring the time-dependent boost potential closer to the boost potential BP shown in FIG. 7.

[0080] FIG. 9 shows an example of a potential energy surface for interatomic distances and an integrated boost potential SBP when a time-dependent boost potential is set as an acceleration condition. The potential energy surface PES represents the distribution of potential energy with respect to a reaction coordinate or collective variable. As shown in FIG. 9, the potential energy surface PES corresponding to the interatomic distance ARD after the chemical reaction corresponds to the potential energy ARP after the chemical reaction and has a minimum value. On the other hand, as shown in FIG. 9, the potential energy surface PES corresponding to the interatomic distance BRD before the chemical reaction corresponds to the potential energy BRP before the chemical reaction and has a local minimum value. As shown in FIG. 9, the integrated boost potential SBP is shown in an approximately trapezoidal shape by superimposing multiple Gaussian functions corresponding to multiple boost potentials.

[0081] Furthermore, when the boost potential applied to an atomic population composed of multiple bondable atoms is a time-dependent boost potential, the setting unit 311 sets, as acceleration conditions, the shape of the activation function applied to the multiple bondable atoms, the degree of weighting for each of multiple population variables corresponding to the multiple bondable atoms, etc. Hereinafter, with reference to Fig. 10, an example of acceleration conditions in the case where multiple bondable atoms can bond to one reference atom (multiple atom pairs) will be described.

[0082] FIG. 10 is a diagram showing an example of acceleration conditions set for a plurality of atom pairs related to one reference atom. In the acceleration conditions, the interatomic distances between the reference atom and each of a plurality of bondable atoms are calculated based on the positions of the plurality of atom pairs and the plurality of atom pairs. The activation function AF calculates each of a plurality of interatomic distances corresponding to each of the plurality of bondable atoms as a function of local distortions X i This is a function that converts the local distortion X i The subscript i in is a natural number greater than or equal to 2 that defines the atom pair.

[0083] The activation function AF shown in FIG. 10 is a function that converts the interatomic distance to zero when the reference atom and each of the multiple bondable atoms are not reacting, and to 1 when any of the bondable atoms is bonded to the reference atom. That is, the local distortion X i corresponds to an index that indicates the degree of chemical bonding between each of a plurality of bondable atoms and the reference atom, with a value of 0 to 1, depending on the interatomic distance. Note that the activation function AF is not limited to being expressed by a logistic function as shown in FIG. 10, and may be expressed by a trigonometric function.

[0084] Global distortions X shown in Figure 10 t As shown in Figure 10, i It is calculated using the following formula (2).

[0085]

number

[0086] As shown in Figure 10, the global distortion X calculated by Eq. (2) t is the local distortion)X i The power average corresponds to an index showing whether any one of the i bondable atoms is reacted. Global distortion X t is zero when all bondable atoms have not reacted with the reference atom, and is 1 when one bondable atom has reacted with the reference atom.

[0087] Then, as shown in FIG. 10, the global distortion X t is transformed into the reaction coordinate (also called collective variable) CV(=η) by the transformation function TF. This gives the global distortion X t is converted into an index of whether the entire system with respect to the reference atom to which the boost potential is applied is close to or far from being in a reacting state. The setting unit 311 stores the converted reaction coordinate η together with the current time t in the main memory device 33 and / or the auxiliary memory device 35.

[0088] As shown in Figure 10, using the reaction coordinate η(t) at the current time t and the reaction coordinate η(t') at the time t', which is an infinitesimal time before the current time t, the boost potential ΔV(η) at the reaction coordinate η is calculated by the following equation (3). The reaction coordinate η(t') corresponds to the history CVh of the reaction coordinate.

[0089]

number

[0090] 2δ in Eq. (3) 2corresponds to the range affected by the boost potential and corresponds to the half-width (full-width at half-maximum or half-height at half-maximum) of the Gaussian function. Also, w in Equation (3) indicates the degree of weighting for each of the plurality of collective variables η. Further, the index (t = τG, τG, 2τG, ···, t’ < t) indicating the range of summation in the summation symbol Σ indicates the time at the execution time point of the MD simulation in the past from the current time t for the reference atom. Here, τG indicates the infinitesimal time in the execution of the MD simulation. The setting unit 311 sets the range 2δ 2 affected by the boost potential and the weight w for each of the plurality of collective variables η.

[0091] Prior to the calculation by Equation (3), the setting unit 311 reads out the reaction coordinate η(t’) stored in the main storage device 33 and / or the auxiliary storage device 35 from the main storage device 33 and / or the auxiliary storage device 35. Next, from these, the setting unit 311 determines that the boost potential ΔV(η) in the reaction coordinate η in Equation (3) is calculated by the procedure shown in FIG. 10 for the reaction coordinate η at the current time t, the reaction coordinate η(t’) in the past from the current time t, the range 2δ 2 affected by the boost potential, and the weight w, and executes the calculation of Equation (3). Thereby, the setting unit 311 sets, for example, by integrating (summating) for each execution of the MD simulation, the boost potential ΔV(η) applied to the plurality of atom pairs corresponding to the specified reference atom. The setting unit 311 stores the set fixed boost potential and the position (reaction coordinate η) to which the fixed boost potential is applied in the main storage device 33 and / or the auxiliary storage device 35.

[0092] (Step S307) The additional force determination unit 317 determines the additional force based on the set boost potential and the coordinates of multiple target atoms. Specifically, the additional force determination unit 317 determines the additional force by differentiating the boost potential with respect to the positions of the bondable electrons. For example, when the boost potential is set in a shape as shown in FIG. 6, the additional force determination unit 317 determines the acting force by an analytical method for the boost potential.

[0093] Fig. 11 is a diagram showing an example of the additional forces corresponding to the different fixed boost potentials in Fig. 6. Fig. 11 shows three additional forces corresponding to the three fixed boost potentials, respectively, by analytically calculating the three fixed boost potentials in Fig. 6. As shown in Fig. 11, the additional forces differ depending on the shape of the boost potential.

[0094] Fig. 12 is a schematic diagram showing an outline of the additional force in Fig. 5. As shown in Fig. 12, an additional force BF is set between the identified reference atom SA and the bondable atom JA. The additional force BF acts, for example, as an attractive force between the reference atom SA and the bondable atom JA. This accelerates the reaction (bonding) between the reference atom SA and the bondable atom JA in the MD simulation described below.

[0095] (Step S308) The MD simulation unit 319 calculates the sum of the acting force and the applied force. The MD simulation unit 319 performs an MD simulation on the plurality of atoms over a very short period of time using the sum of the acting force and the applied force and the positions of the plurality of atoms. At this time, the MD simulation unit 319 may perform the MD simulation further using potential barriers that prevent the plurality of target atoms from separating from each other. By performing the MD simulation, each of the plurality of atoms is moved.

[0096] (Step S309) The determination unit 321 determines whether or not a chemical bond has been formed between multiple target atoms. For example, the determination unit 321 determines whether or not a chemical reaction has been formed for at least one pair (a reference atom and one atom that can form a chemical bond with the reference atom). Specifically, the determination unit 321 compares the interatomic distance between the reference atom and the bond determination distance with the bond determination distance. If a chemical bond between the reference atom and the bondable atom has not been formed (No in step S309), the process of step S310 is executed. If a chemical bond between the reference atom and the bondable atom has been formed (Yes in step S309), the process of step S311 is executed. FIG. 13 is a diagram showing an example in which a chemical bond has been formed between a reference atom SA and a bondable atom JA. As shown in FIG. 13, when a chemical bond is formed, a molecule including the reference atom SA is generated. The method for determining whether a chemical reaction has occurred is not limited to the above, and may also use the angles of multiple atoms (at least one pair of atoms), dihedral angles, or root-mean-square deviation (RMSD) from a template prepared in advance.

[0097] (Step S310) The determination unit 321 determines whether the MD simulation executed in step S308 has been executed for a predetermined time. For example, the determination unit 321 compares the total execution time of the MD simulation with the predetermined time. If the total execution time of the MD simulation is less than the predetermined time, the processes from step S305 onwards are repeated. At this time, if a fixed boost potential has been set, the process of step S306 is omitted. On the other hand, if a time-dependent boost potential has been set, the boost potential is reset in step S306 using the positions of the bondable electrons immediately before the execution of the MD simulation. If the total execution time of the MD simulation is equal to or greater than the predetermined time, the process of step S311 is executed. Note that the determination in this step is not limited to the above, and may be performed, for example, by determining whether the average value and / or maximum value of the boost potential BP exceeds a pre-specified value.

[0098] The determination unit 321 may reset the measurement of the "execution time" in step S310 to 0 when step S310 becomes Yes, that is, when the execution time reaches a predetermined time. That is, the determination unit 321 may count the execution time again in the loop processing from step S305 to step S309. There may be two types of execution time: the total execution time in the MD simulation execution processing, and the execution time used for the determination in step S310. In this case, the latter execution time may be used for the determination in step S310.

[0099] (Step S311) The determination unit 321 initializes the sum of the boost potentials applied to multiple target atoms. For example, the determination unit 321 initializes the sum of the fixed boost potential and the time-dependent boost potential applied for each execution of the MD simulation.

[0100] (Step S312) The identifying unit 313 updates the attribute information of the atom with which the chemical bond is formed based on the functional groups related to the atoms with which the chemical bond is formed. This allows the identifying unit 313 to update the multiple target atoms. For example, in FIG. 13 , when it is determined that a chemical bond is formed between the reference atom SA and the bondable atom JA, the identifying unit 313 identifies a new bondable atom NJA by updating the attribute information of the bondable atom using a program based on predetermined rules for chemically bonded functional groups. After this step, the processing from step S304 onwards is repeated. That is, when the determining unit 321 determines that a chemical bond is formed, the identifying unit 313 updates the attribute information of the multiple target atoms with which the chemical bond is formed, and then identifies the other multiple target atoms.

[0101] (Step S313) The determination unit 321 determines whether the attribute of the specified atom (i.e., the updated attribute) is equal to or less than a certain number. Specifically, if the total number of reference atoms (e.g., atom pairs) identified by the identification unit 313 is 100 and the certain number is 10, the determination unit 321 determines whether a boost potential is set for each of the 90 reference atoms. If the attribute of the specified atom (i.e., the updated attribute) is equal to or more than the certain number (No in step S313), steps S304 and subsequent steps are executed. If it is determined that a chemical bond has been formed, the identification unit 313 identifies other target atoms among the multiple atoms that are targets of the chemical bond (step S304). Next, the applied force determination unit 315 inputs the atomic structures of the multiple atoms after the MD simulation into the trained NNP to determine a second applied force to be applied to the multiple atoms (step S305). In addition, the applied force determination unit 317 determines information on the second applied force to be applied to at least one of the other multiple target atoms (steps 306 and S307). Next, the MD simulation unit 319 executes a molecular dynamics simulation on the plurality of atoms again using information on the second acting force and the second applied force and position information on the plurality of atoms after execution of the MD simulation.

[0102] That is, the processes of steps S304 to S312, for example, identifying multiple target atoms, acquiring information on acting forces, acquiring information on added forces, and executing molecular dynamics simulations, are repeatedly executed until a predetermined criterion is satisfied. If the number of attributes of the specified atoms (i.e., updated attributes) exceeds a certain number (Yes in step S313), the process of step S314 is executed.

[0103] In the above process, there may be a series of cases where the determination in step S313 is No. For example, there is a possibility that the determination in step S313 is No in cases where the number of atoms involved in the reaction is extremely small or where the atoms and molecules themselves do not move much in the simulation space (when the viscosity of the atoms and molecules is high). For this reason, a maximum number of times (a predetermined number of times) for looping the processes from step S304 to step S313 may be set (specified) in advance. In this case, in the determination in step S313, if the number of times for looping the processes from step S304 to step S313 reaches the predetermined number of times or is performed more than the predetermined number of times, the process of step S314 is executed.

[0104] (Step S314) The evaluation unit 323 evaluates the physical properties of molecules generated by chemical reactions accompanying the MD simulation. The evaluation unit 323 outputs the evaluation of the physical properties of the molecules to the main storage device 33, the auxiliary storage device 35, and / or a display. In addition to displaying the evaluation of the physical properties on the display, a moving image (animation) of the movement of multiple atoms by the MD simulation may be displayed on the display in chronological order.

[0105] Based on the above, the information processing device 1 according to this embodiment identifies multiple target atoms among multiple atoms that are targets of chemical bonding, acquires information on first forces acting on the multiple atoms generated by inputting the atomic structures of the multiple atoms into a trained neural network potential, acquires information on a first additional force to be acted on at least one of the multiple target atoms, and performs an MD simulation on the multiple atoms using the information on the first force, the information on the first additional force, and position information on the multiple atoms. Furthermore, the information processing device 1 according to this embodiment determines whether or not the multiple target atoms have formed chemical bonds based on the position information of the multiple target atoms after performing the MD simulation.

[0106] As a result, the information processing device 1 according to the present embodiment can apply a boost potential between atoms where a reaction is desired to be induced, and can execute an MD simulation using an additional force based on the boost potential and an acting force based on the NNP. As a result, the information processing device 1 according to the present embodiment can generate chemical reactions within a realistic calculation time for MD simulation, and can handle chemical reactions such as polymerization and decomposition for a wide range of substances.

[0107] Furthermore, when it is determined that chemical bonds between the multiple target atoms have not been formed, the information processing device 1 according to the embodiment acquires information on second forces acting on the multiple atoms, which is generated by inputting the multiple atomic structures after the MD simulation into a neural network, acquires information on a second additional force acting on at least one of the multiple target atoms, and re-executes the MD simulation on the multiple atoms using the information on the second force, the information on the second additional force, and position information on the multiple atoms after the MD simulation. Furthermore, when it is determined that chemical bonds between the multiple target atoms have not been formed, the information processing device 1 according to the embodiment further executes the MD simulation again using information on the first additional force.

[0108] As a result, the information processing device 1 according to the embodiment can execute an MD simulation using an additional force determined by accumulating the boost potential as the MD simulation is executed. Therefore, the information processing device 1 according to the embodiment can further improve the acceleration (boost) of a chemical reaction and further reduce the calculation time for generating a chemical reaction in an MD simulation.

[0109] Furthermore, when it is determined that a chemical bond has not been formed between the target atoms, the information processing device 1 according to the embodiment identifies the other target atoms among the target atoms that are the target of the chemical bond, acquires information on a second acting force acting on the atoms, which is generated by inputting the atomic structures of the target atoms after the MD simulation into a trained neural network, acquires information on a second additional force acting on at least one of the other target atoms, and performs an MD simulation on the multiple atoms again using the information on the second acting force, the information on the second additional force, and the positional information of the target atoms after the MD simulation. Thus, according to the information processing device 1 according to the embodiment, by dynamically applying a boost potential according to the movement history of the target atoms (metadynamics), the additional force can be efficiently applied to the target atoms, thereby further reducing the calculation time required to generate a chemical reaction in the MD simulation.

[0110] Furthermore, when the distance between multiple target atoms is smaller than a predetermined threshold, the information processing device 1 according to the embodiment determines that the first additional force is not to be applied to the multiple target atoms as information on the first additional force. As a result, the information processing device 1 according to the embodiment can execute an MD simulation without applying a boost potential and without applying an additional force when the bondable electron approaches the reference atom up to the covalent bond radius. Therefore, the information processing device 1 according to the embodiment can execute an MD simulation in a natural situation, thereby generating a chemical reaction in a realistic calculation time.

[0111] Furthermore, the information processing device 1 according to the embodiment executes the MD simulation by further using a potential barrier that prevents the coordinates of multiple target atoms from becoming separated from each other. As a result, the information processing device 1 according to the embodiment can prevent the reference atom and the bondable atom from becoming separated more than necessary by applying a boost potential, and can generate a chemical reaction within a realistic calculation time.

[0112] Furthermore, the information processing device 1 according to the embodiment determines information on the first additional force based on position information of multiple target atoms. Specifically, the information on the first additional force is determined based on the position information of multiple target atoms and a boost potential. That is, the information processing device 1 according to the embodiment can apply a boost potential to multiple bondable atoms with respect to a reference atom. As a result, the information processing device 1 according to the embodiment can generate chemical reactions more efficiently within a realistic calculation time.

[0113] Furthermore, the information processing device 1 according to the embodiment executes an MD simulation for a plurality of atoms using the sum of the first acting force and the first additional force and position information of the plurality of atoms. For example, when it is determined that a chemical bond has been formed, the information processing device 1 according to the embodiment initializes a boost potential, identifies other target atoms among the plurality of atoms that are targets of the chemical bond, determines a second acting force acting on each of the plurality of atoms by inputting the types of the plurality of atoms and the positions of the plurality of atoms into a trained neural network potential, determines a second additional force acting on each of the other target atoms based on the coordinates of the other target atoms, executes a molecular dynamics simulation for the plurality of atoms using the sum of the second acting force and the second additional force and the positions of the plurality of atoms, and determines whether or not a chemical bond has been formed based on the positions of the plurality of target atoms after the MD simulation has been executed.

[0114] As a result, the information processing device 1 according to the embodiment can suppress the reverse reaction (decomposition reaction) of the chemical bond by initializing the boost potential every time a chemical bond is formed. Therefore, according to the information processing device 1 according to the embodiment, the influence of the application of the boost potential can be efficiently imparted to a plurality of other target atoms, and the chemical reaction can be generated in a more realistic calculation time.

[0115] (Application example) This application example involves identifying multiple target atoms using a trained neural network (hereinafter referred to as a specific neural network (second neural network)) that uses a structure up to the final hidden layer (hereinafter referred to as the final hidden layer) of a trained neural network potential (first neural network) and an output layer connected after the final hidden layer. The input to the specific neural network, like NNP, is the positions of multiple atoms arranged in virtual space and the types (atomic structures) of the multiple atoms. Note that the hidden layer connected to the output layer of the specific neural network is not limited to the final hidden layer and may be another hidden layer. The specific neural network is generated by learning (e.g., transfer learning) using the features of the hidden layer calculated by NNP. The output (output information) from the specific neural network is, for example, a label (label information) indicating the type (type of atom) of a functional group containing multiple atoms that are susceptible to chemical bonding (chemical reaction). The label may be, for example, "carbon radical" or "carbon of vinyl group."

[0116] The identification unit 313 outputs multiple labels corresponding to multiple atoms by inputting the positions of multiple atoms arranged in virtual space and the types of the multiple atoms into the identification neural network. When a label (functional group) related to a chemical bond is specified by a user via an input interface, the identification unit 313 identifies a label corresponding to the specified functional group from the multiple labels output from the identification neural network. The label corresponds to the identifier of the target atom. Next, the identification unit 313 identifies multiple target atoms using a search range, with the atom corresponding to the identified label as a reference atom. As a result, the identification unit 313 identifies multiple target atoms using the atomic structures of the multiple atoms and the trained second neural network. That is, the identification unit 313 identifies multiple target atoms based on output information from the trained second neural network. At this time, the identification unit 313 obtains the output information by inputting the atomic structures of the multiple atoms into the trained second neural network. Since the identification of multiple target atoms using a search range is similar to the embodiment, a description thereof will be omitted.

[0117] FIG. 14 illustrates an example of an NNP (first neural network) and a specific neural network (second neural network) SNN. As shown in FIG. 14, the NNP receives inputs of the positions and types of atoms, and outputs the energy of each atom. Meanwhile, as shown in FIG. 14, the input layer and hidden layer are common between the NNP and the specific neural network SNN. The difference between the specific neural network SNN and the NNP is that the output destination of the final hidden layer FML of the NNP is a different output layer from that of the NNP. As shown in FIG. 14, in the specific neural network SNN, labels are generated using the hidden layer of the NNP, which outputs the local positional relationships between each atom as features, and the input of the NNP. Therefore, the features generated by the NNP potentially contain information about functional groups. The features of the NNP are generated, for example, using a TeaNet or a reaction neural network.

[0118] The specific neural network SNN is generated by learning using the positions and types of atoms as training data and data of preset labels corresponding to the training data as ground truth data. This learning is based on, for example, known transfer learning, and therefore a detailed description thereof will be omitted.

[0119] As a modified example of the application example, the identification unit 313 may acquire output information by inputting information (features) obtained in the intermediate layer of the trained neural network (first neural network) when the atomic structures of multiple atoms are input to the trained neural network (first neural network). The other neural network and the specific neural network SNN may be collectively referred to as a second neural network. In this case, if the input information to the second neural network is an atomic structure, the second neural network corresponds to the specific neural network SNN. Furthermore, if the input information to the second neural network is information from the intermediate layer of the NNP (first neural network), the second neural network corresponds to the other neural network. Furthermore, in the above description, the output information output from the second neural network is described as label information for each input atom, but this is not limited to this. The output information output from the second neural network may be information on whether or not each input atom (a plurality of atoms input to the input layer of the first neural network) is capable of reacting.

[0120] As described above, the information processing device 1 according to an application example of this embodiment identifies multiple target atoms using the atomic structures of multiple atoms and a trained second neural network. For example, the information processing device 1 according to an application example of this embodiment identifies multiple target atoms based on output information from the trained second neural network. More specifically, the information processing device 1 according to an application example of this embodiment acquires output information by inputting the atomic structures of multiple atoms into the trained second neural network. Note that the information processing device 1 according to an application example of this embodiment may acquire output information by inputting information obtained in an intermediate layer of the trained neural network when the atomic structures of multiple atoms are input into the trained second neural network. The information processing device 1 according to an application example of this embodiment can automatically identify multiple target atoms by a user selecting the type of functional group.

[0121] As a result, the information processing device 1 according to the application example of this embodiment can reduce the burden on the user in preparations before carrying out an MD simulation, without using substructure matching using SMARTS or the like, and without the user having to specify multiple reference atoms in the equilibrated initial structure INS arranged in the simulation space SS as shown in Fig. 4. Other effects are the same as those of the embodiment, and therefore will not be described here.

[0122] When the technical idea of ​​the embodiments is realized as an information processing method, in the information processing method, at least one computer identifies multiple target atoms among multiple atoms that are targets for chemical bonding, acquires information on first forces acting on the multiple atoms generated by inputting the atomic structures of the multiple atoms into a trained neural network potential, acquires information on a first additional force to be acted on at least one of the multiple target atoms, and performs a molecular dynamics simulation on the multiple atoms using the information on the first force, the information on the first additional force, and position information of the multiple atoms. The procedure and effects of the MD simulation execution process related to the information processing method are similar to those described in the embodiments, so a description thereof will be omitted.

[0123] When the technical idea in the embodiment is realized by an information processing program, the information processing program causes at least one computer to identify multiple target atoms among multiple atoms that are targets for chemical bonding, acquire information on a first action force acting on the multiple atoms generated by inputting the atomic structures of the multiple atoms into a trained neural network potential, acquire information on a first additional force to be acted on at least one of the multiple target atoms, and perform a molecular dynamics simulation on the multiple atoms using the information on the first action force, the information on the first additional force, and position information of the multiple atoms.

[0124] For example, the information processing program can be realized by installing the information processing program in a computer such as a simulation device or simulation server that executes MD simulations for multiple atoms and expanding the program in memory. In this case, the program that can cause a computer to execute MD simulation execution processing can also be stored in a storage medium such as a magnetic disk (hard disk, etc.), an optical disk (CD-ROM, DVD, etc.), or a semiconductor memory and distributed. The procedure and effects of the MD simulation execution processing using the information processing program are the same as those in the embodiment, so a description thereof will be omitted.

[0125] Some or all of the devices in the above-described embodiments may be configured as hardware, or may be configured as information processing software (programs) executed by a CPU, a GPU, or the like. In the case of software information processing, software that realizes at least some of the functions of each device in the above-described embodiments may be stored on a non-transitory storage medium (non-transitory computer-readable medium) such as a flexible disk, a CD-ROM (Compact Disc-Read Only Memory), or a USB memory, and the software information processing may be executed by loading the software into the computer 30. The software may also be downloaded via the communication network 5. Furthermore, the software may be implemented in a circuit such as an ASIC or FPGA, so that the information processing is executed by hardware.

[0126] The type of storage medium that stores the software is not limited. The storage medium is not limited to removable media such as magnetic disks or optical disks, but may be fixed storage media such as hard disks or memory. The storage medium may be provided inside the computer or outside the computer.

[0127] In this specification (including the claims), when the expression "at least one of a, b, and c" or "at least one of a, b, or c" (including similar expressions) is used, it includes any of a, b, c, ab, ac, bc, or abc. It may also include multiple instances of any element, such as aa, abb, aabbcc, etc. Furthermore, it also includes the addition of elements other than the enumerated elements (a, b, and c), such as having d, as in abcd.

[0128] In this specification (including the claims), when expressions such as "using data as input / based on / according to / in response to" (including similar expressions) are used, unless otherwise specified, this includes cases where various data itself is used as input, or where various data that has been processed in some way (e.g., noise-added, normalized, intermediate representation of various data, etc.) is used as input. Furthermore, when a statement is made that a result is obtained "based on / according to / in response to data," this includes cases where the result is obtained based solely on the data in question, as well as cases where the result is obtained as a result of being influenced by other data, factors, conditions, and / or states other than the data in question. Furthermore, when a statement is made that "data is output," this includes cases where various data itself is used as output, or where various data that has been processed in some way (e.g., noise-added, normalized, intermediate representation of various data, etc.) is output, unless otherwise specified.

[0129] When the terms "connected" and "coupled" are used in this specification (including the claims), they are intended as open-ended terms that encompass any of direct connection / coupling, indirect connection / coupling, electrically connection / coupling, communicatively connection / coupling, functionally connection / coupling, and physically connection / coupling. These terms should be interpreted appropriately according to the context in which they are used, but any form of connection / coupling that is not intentionally or naturally excluded should be interpreted as being included in these terms without limitation.

[0130] In this specification (including the claims), the expression "A configured to B" may include the physical structure of element A having a configuration capable of performing operation B, and the permanent or temporary setting / configuration of element A being configured / set to actually perform operation B. For example, if element A is a general-purpose processor, it is sufficient that the processor has a hardware configuration capable of performing operation B, and is configured to actually perform operation B by setting a permanent or temporary program (instruction). Also, if element A is a dedicated processor or dedicated arithmetic circuit, it is sufficient that the circuit structure of the processor is implemented to actually perform operation B, regardless of whether control instructions and data are actually attached.

[0131] When used in this specification (including the claims), terms implying containing or possessing (e.g., "comprising / including" and "having") are intended to be open-ended terms that include containing or possessing things other than the object designated by the object of the term. When the object of such a term implies no quantity or a singular number (e.g., an article such as "a" or "an"), the expression should be construed as not being limited to a specific number.

[0132] In this specification (including the claims), even if expressions such as "one or more" or "at least one" are used in some places and expressions that do not specify a quantity or that imply a singular number (expressions using the articles "a" or "an") are used in other places, the latter expressions are not intended to mean "one." In general, expressions that do not specify a quantity or that imply a singular number (expressions using the articles "a" or "an") should be interpreted as not necessarily being limited to a specific number.

[0133] In this specification, when a particular advantage / result is described as being obtained from a particular configuration of an embodiment, it should be understood that the same advantage / result can also be obtained from one or more other embodiments having the same configuration, unless otherwise stated. However, it should be understood that the presence or absence of the effect generally depends on various factors, conditions, and / or states, etc., and that the effect is not necessarily obtained by the configuration. The effect is merely obtained by the configuration described in the embodiment when various factors, conditions, and / or states, etc. are satisfied, and the effect does not necessarily occur in a claimed invention that defines the same or a similar configuration.

[0134] When used in this specification (including the claims), terms such as "maximize" include finding a global maximum, finding an approximation of a global maximum, finding a local maximum, and finding an approximation of a local maximum, and should be interpreted appropriately according to the context in which the term is used. It also includes finding approximations of these maxima probabilistically or heuristically. Similarly, when used in this specification (including the claims), terms such as "minimize" include finding a global minimum, finding an approximation of a global minimum, finding a local minimum, and finding an approximation of a local minimum, and should be interpreted appropriately according to the context in which the term is used. It also includes finding approximations of these minima probabilistically or heuristically. Similarly, when used in this specification (including the claims), terms such as "optimize" include finding a global optimum, finding an approximation of a global optimum, finding a local optimum, and finding an approximation of a local optimum, and should be interpreted appropriately according to the context in which the term is used. It also includes finding approximations of these optima probabilistically or heuristically.

[0135] In this specification (including claims), when multiple pieces of hardware perform a predetermined process, the pieces of hardware may cooperate to perform the predetermined process, or some of the hardware may perform all of the predetermined process. Furthermore, some of the hardware may perform part of the predetermined process, and other hardware may perform the rest of the predetermined process. In this specification (including claims), when an expression such as "one or more pieces of hardware perform a first process, and the one or more pieces of hardware perform a second process" is used, the hardware performing the first process and the hardware performing the second process may be the same or different. In other words, it is sufficient that the hardware performing the first process and the hardware performing the second process are included in the one or more pieces of hardware. Note that hardware may include an electronic circuit or a device including an electronic circuit.

[0136] In this specification (including the claims), when multiple storage devices (memories) store data, each of the multiple storage devices (memories) may store only a portion of the data, or may store the entire data.

[0137] Although the embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to the individual embodiments described above. Various additions, modifications, substitutions, partial deletions, etc. are possible within the scope of the conceptual idea and spirit of the present invention derived from the content defined in the claims and their equivalents. For example, in all of the above-described embodiments, when numerical values ​​or formulas are used in the explanation, they are shown as examples and are not limited to these. Furthermore, the order of each operation in the embodiments is shown as an example and is not limited to these. [Explanation of symbols]

[0138] 1. Information processing equipment 5. Communication Network 9A external device 9B External device 30 Computer 31 processors 33 Main memory 35 Auxiliary storage device 37 Network Interface 39 Device Interfaces 41 Bus 311 Settings 313 Specific part 315 Acting force determination section 317 Additional force determination unit 319 MD Simulation Department 321 Judgment section 323 Evaluation Department

Claims

1. at least one memory; at least one processor; The at least one processor Identifying a plurality of target atoms that are targets of chemical bonding among the plurality of atoms; acquiring information on a first acting force acting on the plurality of atoms, the first acting force being generated by inputting the atomic structures of the plurality of atoms into a neural network; acquiring information on a first applied force to be applied to at least one of the plurality of target atoms; performing a molecular dynamics simulation on the plurality of atoms using information on the first acting force, information on the first applied force, and position information on the plurality of atoms; Information processing device.

2. The at least one processor determining whether or not the chemical bonds are formed between the plurality of target atoms based on position information of the plurality of target atoms after the execution of the molecular dynamics simulation; The information processing device according to claim 1 .

3. If it is determined that the chemical bonds of the plurality of target atoms are not formed, the at least one processor: acquiring information on a second acting force acting on the plurality of atoms, the second acting force being generated by inputting the atomic structures of the plurality of atoms after the execution of the molecular dynamics simulation into the neural network; acquiring information on a second additional force to be applied to at least one of the plurality of target atoms; performing a molecular dynamics simulation on the plurality of atoms again using information on the second acting force, information on the second applied force, and position information on the plurality of atoms after performing the molecular dynamics simulation; The information processing device according to claim 2 .

4. When it is determined that the chemical bonds between the plurality of target atoms are not formed, the at least one processor performs the molecular dynamics simulation again, further using information on the first applied force. The information processing device according to claim 3 .

5. and wherein the at least one processor performs the molecular dynamics simulation further using a potential barrier that prevents the plurality of target atoms from separating from each other. The information processing device according to claim 1 .

6. If it is determined that the chemical bond is formed, the at least one processor: Identifying other multiple target atoms that are targets for chemical bonding among the multiple atoms; acquiring information on a second acting force acting on the plurality of atoms, the second acting force being generated by inputting the atomic structures of the plurality of atoms after the execution of the molecular dynamics simulation into the neural network; acquiring information on a second additional force to be applied to at least one of the other plurality of target atoms; performing a molecular dynamics simulation on the plurality of atoms again using information on the second acting force, information on the second applied force, and position information on the plurality of atoms after performing the molecular dynamics simulation; The information processing device according to claim 1 .

7. when it is determined that the chemical bond is formed, the at least one processor updates attribute information of the plurality of target atoms with which the chemical bond is formed, and then identifies the other plurality of target atoms; The information processing device according to claim 6 .

8. the at least one processor determines, when a distance between the plurality of target atoms is smaller than a predetermined threshold, that the first applied force is not to be applied to the plurality of target atoms as information of the first applied force; The information processing device according to claim 1 .

9. The at least one processor Identifying a plurality of target atoms, obtaining information on the acting forces, obtaining information on the added forces, and performing a molecular dynamics simulation; Repeatedly execute until a predetermined criterion is met. The information processing device according to claim 1 .

10. the at least one processor determines information about the first applied force based on position information of the plurality of target atoms. The information processing device according to claim 1 .

11. the at least one processor determines information about the first applied force based on position information of the plurality of target atoms and a boost potential. The information processing device according to claim 1 .

12. The at least one processor performing the molecular dynamics simulation on the plurality of atoms using the sum of the first acting force and the first applied force and position information of the plurality of atoms; The information processing device according to claim 1 .

13. The at least one processor identifying the plurality of target atoms using the atomic structures of the plurality of atoms and a second neural network; The information processing device according to claim 1 .

14. The at least one processor identifying the plurality of target atoms based on output information from the second neural network; The information processing device according to claim 13.

15. The at least one processor obtaining the output information by inputting the atomic structures of the plurality of atoms into the second neural network; The information processing device according to claim 14.

16. The at least one processor and acquiring the output information by inputting information obtained in an intermediate layer of the neural network when the atomic structures of the plurality of atoms are input to the neural network into the second neural network. The information processing device according to claim 14.

17. the at least one processor inputs atomic structures of the plurality of atoms into the neural network to generate information of the first force acting on the plurality of atoms; The information processing device according to claim 1 .

18. At least one computer Identifying a plurality of target atoms that are targets of chemical bonding among the plurality of atoms; acquiring information on a first acting force acting on the plurality of atoms, the first acting force being generated by inputting the atomic structures of the plurality of atoms into a neural network; acquiring information on a first applied force to be applied to at least one of the plurality of target atoms; performing a molecular dynamics simulation on the plurality of atoms using information on the first acting force, information on the first applied force, and position information on the plurality of atoms; Information processing methods.

19. On at least one computer, Identifying a plurality of target atoms that are targets of chemical bonding among the plurality of atoms; acquiring information on a first acting force acting on the plurality of atoms, the first acting force being generated by inputting the atomic structures of the plurality of atoms into a neural network; acquiring information on a first applied force to be applied to at least one of the plurality of target atoms; performing a molecular dynamics simulation on the plurality of atoms using information on the first acting force, information on the first applied force, and position information on the plurality of atoms; An information processing program that makes this possible.

Images