FORECAST SYSTEM, FORECAST PROGRAM AND FORECAST METHOD

Abstract

A prediction system according to the present invention comprises a detector configured to detect information concerning molecules contained in a base oil stored at a normal temperature of 20 °C and a normal pressure of 1.013 × 10 5Pa is liquid, as molecules of interest, a single-molecule information detector configured to detect a 3D structure of the molecule of interest as single-molecule information, a liquid structure generator configured to generate a liquid structure by defining a simulation cell with a freely selectable parallelepiped shape and arranging a freely selectable number of units of the single-molecule information within the simulation cell, an optimized structure detector configured to detect a liquid-optimized structure optimized by relaxing the liquid structure, a feature calculator configured to calculate a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from the liquid-optimized structure and calculating a radial distribution function,representing an abundance ratio of one molecule to another molecule at each atomic distance, and a predictor configured to predict a density level of an oil film formed by the molecules of interest based on the feature.

Description

TECHNICAL FIELD

[0001] The present invention relates to a prediction system, a prediction program and a prediction method. BACKGROUND

[0002] In an industrial product equipped with industrial machinery such as an internal combustion engine, hydraulic machine, compression machine, turbine, transmission element, bearing, refrigeration device, etc., various lubricants such as engine oil, hydraulic oil, compressor oil, turbine oil, gear oil and refrigeration machine oil are used to ensure smooth operation of these machines.

[0003] A lubricant mainly comprises one or more base oils and one or more additives. As such a lubricant, for example, a lubricant composition comprising an ester-based base oil and a lubricant additive, and the like, are disclosed (see, for example, Patent Document 1). STATE OF THE ART DOCUMENTS

[0004] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2022-158124 SUMMARY OF THE INVENTION PROBLEMS TO BE SOLVED BY THE INVENTION

[0005] For a lubricant to deliver the desired lubricating performance, it is important to select a base oil that does not inhibit the effect of an additive or further enhances the effect of the additive. One method for selecting a base oil is, for example, a method in which the structure of an oil film formed from the base oil is analyzed by molecular simulation for each molecule that makes up the base oil.

[0006] Specifically, for example, as described in “Understanding the effect of the base oil on the physical adsorption process of organic additives using molecular dynamics” (M. Konishi and H. Washizu, Tribology International, 2020, Vol. 149, Article 105568), there is a molecular dynamics simulation method that investigates the diffusion behavior of an additive in a liquid phase of a base oil and the adsorption time of the additive on the surface of a metal plate using an atomic / molecular model in which an additive is arranged in a liquid phase of a base oil and sandwiched between metal plates.

[0007] However, because this method requires complex calculations such as modeling and post-analysis, it takes a long time to perform calculations such as determining the analysis conditions, and it is not realistic to perform calculations for all the molecules that make up a base oil. For example, analyzing the structure of an oil film formed by a base oil using molecular simulation for a specific base oil type can take several days to several months to complete the calculation. Therefore, a method for efficiently selecting a base oil that maximizes the effect of additives was desired.

[0008] An object of one aspect of the present invention is to shorten the time required for selecting a base oil that brings out or emphasizes the effect of an additive. MEANS TO SOLVE THE PROBLEM

[0009] One aspect of the present invention is a prediction system comprising: a detector configured to detect information regarding molecules contained in a base oil stored at normal temperature of 20°C and normal pressure of 1.013 × 10 5Pa is liquid, as molecules of interest, a single-molecule information detector that is configured to detect a three-dimensional (3D) structure of the molecule of interest as single-molecule information, a liquid structure generator that is configured to generate a liquid structure by defining a simulation cell that has a freely selectable parallelepiped shape and arranging a freely selectable number of units of the single-molecule information within the simulation cell, an optimized structure detector that is configured to detect a liquid-optimized structure that can be created by relaxing orRelaxing the liquid structure, a feature calculator configured to calculate a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from the liquid-optimized structure and calculating a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance, and a predictor configured to predict a density degree of an oil film formed from the molecules of interest based on the feature.

[0010] Another aspect of the present invention is a prediction program that causes a computer to perform: an acquisition step of acquiring information regarding molecules contained in a base oil stored at normal temperature of 20°C and normal pressure of 1.013 × 10 5Pa is liquid, as molecules of interest, a single-molecule information acquisition step for acquiring a 3D structure of molecules of interest as single-molecule information, a liquid structure generation step for generating a liquid structure by defining a simulation cell with an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information within the simulation cell, an optimized structure acquisition step for acquiring a liquid-optimized structure which can be created by relaxing orRelaxing the liquid structure is optimized, a feature calculation step of calculating a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure and calculating a radial distribution function representing an abundance ratio of one molecule to another molecule at each atomic distance, and a prediction step of predicting a density degree of an oil film formed of the molecules of interest based on the feature.

[0011] Another aspect of the present invention is a prediction method in which a computer is configured to perform: an acquisition step of acquiring information regarding molecules contained in a base oil stored at normal temperature of 20°C and normal pressure of 1.013 × 10 5Pa is liquid, as molecules of interest, a single-molecule information acquisition step for acquiring a 3D structure of molecules of interest as single-molecule information, a liquid structure generation step for generating a liquid structure by defining a simulation cell with an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information within the simulation cell, an optimized structure acquisition step for acquiring a liquid-optimized structure which can be created by relaxing orRelaxing the liquid structure is optimized, a feature calculation step of calculating a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure and calculating a radial distribution function representing an abundance ratio of one molecule to another molecule at each atomic distance, and a prediction step of predicting a density degree of an oil film formed from the molecules of interest based on the feature. EFFECTS OF THE INVENTION

[0012] One aspect of the present invention shortens the time required to select a base oil that will enhance the effect of an additive. BRIEF DESCRIPTION OF THE DRAWINGS [ Fig. 1] Fig. Figure 1 is an explanatory diagram illustrating the prediction of a density level of an oil film from a molecule of interest. [ Fig. 2] Fig. Figure 2 is an explanatory diagram illustrating a difference in adsorption states of an additive depending on a difference in the density degree of an oil film. [ Fig. 3] Fig. 3 is a diagram illustrating a configuration of a prediction system according to an embodiment of the present invention. [ Fig. 4] Fig. 4 is a functional block diagram illustrating a configuration of a prediction device. [ Fig. 5] Fig. Figure 5 is a diagram illustrating an example of a table describing structural formulas (SMILES). [ Fig. 6] Fig. Figure 6 is a diagram illustrating an example of generating a 3D structure from a single-molecule structure of molecules of interest. [ Fig. 7] Fig. Figure 7 is a diagram illustrating an example of generating a liquid structure from the 3D structure of the molecule of interest. [ Fig. 8] Fig. Figure 8 is a graph illustrating an example of a radial distribution function. [ Fig. 9] Fig. Figure 9 is a diagram illustrating an example of a distance between two molecules. [ Fig. 10] Fig. Figure 10 is a graph illustrating an example of a radial distribution function of an ordinary normal alkane liquid. [ Fig. 11] Fig. Figure 11 is a diagram showing an example of a radial distribution function that only relates to an atomic distance between molecules in Fig. 10 concentrated. [ Fig. 12] Fig. 12 is a block diagram showing a hardware configuration of the prediction device. [ Fig. 13] Fig. 13 is a flowchart illustrating a prediction method according to the present embodiment. DESCRIPTION OF EMBODIMENTS

[0013] Embodiments of the present invention are described in detail below. For clarity, the description will be clearer, and redundant descriptions will be omitted throughout the drawings. In this description, "up to" in the context of a numerical range means that the numerical values ​​described before and after "up to" are shown as the lower limit and upper limit values, unless otherwise specified. <vorhersagesystem>

[0014] A prediction system according to one embodiment of the present invention is described. The prediction system according to the present embodiment predicts a density level of an oil film formed from a base oil consisting of molecules of interest.

[0015] In the present embodiment, molecules of interest are molecules that form a base oil that is volatile at a normal temperature of 20°C and a normal pressure of 1.013 × 10 5 Pa is liquid.

[0016] Base oil is an oil used as a base material for a lubricant, grease, or the like. It has functions such as dissolving an additive evenly and stably and delivering the additive to a location where it is needed for lubrication. Base oil is mainly divided into mineral oils obtained by petroleum refining, synthetic oils obtained by chemical synthesis, blended oils thereof, and the like. Lubricant or grease is a mixture containing one or more base oils and one or more additives. Examples of lubricants include engine oils, hydraulic oils, compressor oils, turbine oils, gear oils, and refrigeration oils, which are used in industrial machinery such as internal combustion engines, hydraulic machines, compression machines, turbines, transmission elements, bearings, and refrigeration machines.Examples of the additives include plasticizers, stabilizers, oiliness agents, friction modifiers, wear inhibitors, antioxidants, UV absorbers, lubricants, mold release agents, antistatic agents, rust inhibitors, defoamers, viscosity index improvers and the like.

[0017] As in Fig. As shown in Figure 1, the prediction system according to the present embodiment predicts a density level of an oil film containing molecules of interest (or molecules of interest) by performing a molecular simulation, such as a molecular dynamics method, using an optimal liquid structure of a single-molecule structure of molecules of interest contained in a base oil and calculating a feature indicating an intermolecular space of the molecules of interest. As shown in Fig. 2, in a sliding portion of an industrial machine or the like, with respect to a sliding surface that rubs against a surface of another sliding member (hereinafter simply referred to as "sliding surface of a sliding portion"), as an oil film formed from a base oil consisting of molecules of interest becomes denser, an additive contained in the base oil is less likely to penetrate through the inside of the oil film and less likely to reach the sliding surface of the sliding portion; therefore, the effect of the additive is less likely to be exhibited (see Fig. 2(a)). As the oil film becomes thinner, gaps are more likely to appear in the oil film, and the additive is more likely to penetrate through the gaps in the oil film, so that the additive migrates into the oil film and easily reaches the sliding surface of the sliding section (see Fig. 2(b)). Therefore, the effect of the additive, such as friction reduction, is more likely to occur. The prediction system according to the present embodiment predicts a density level of an oil film formed of molecules of interest by using, as a feature of the molecules of interest contained in a base oil, a feature that correlates with an atomic distance between atoms of different molecules (for example, a slope, a center point, or the like of an approximate representation of a radial distribution function described later), instead of an atomic distance in a molecule. The prediction system according to the present embodiment uses a prediction result of a density level of an oil film to select a base oil that exhibits an effect of an additive, thereby shortening the time required for selecting a base oil.

[0018] Fig. 3 is a diagram illustrating a configuration of the prediction system according to the present embodiment. As shown in Fig. As shown in Figure 3, the prediction system 1 comprises a prediction device 10, a memory 20, and a machine learning resource 30. In the prediction system 1, the prediction device 10, the memory 20, and the machine learning resource 30 are connected via a communication network 40, and input values ​​for the prediction device 10, the memory 20, and the machine learning resource 30, as well as output values ​​of the prediction device 10, the memory 20, and the machine learning resource 30, can be transmitted via the communication network 40. At least one of the memories 20 or the machine learning resource 30 can be stored in a cloud.

[0019] In the present embodiment, the prediction device 10, the memory 20, and the machine learning capability 30 are connected via the communication network 40, but may be connected in a wired or wireless manner. The prediction device 10, the memory 20, and the machine learning capability 30 may transmit and receive data via a communication network including a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or the Internet. The prediction device 10 may be a single device, such as a personal computer (PC) equipped with components of the prediction device 10.

[0020] The prediction device 10 predicts a density level of an oil film formed from a base oil. Details of the prediction device 10 will be described later.

[0021] The memory 20 stores a data table containing information about a lubricant and other information.

[0022] The information about a lubricant includes information about a type of lubricant, a base oil and an additive contained in the lubricant, and the like.

[0023] The type of lubricant is engine oil, hydraulic oil, gear oil, refrigeration oil or the like.

[0024] Examples of information about a base oil and an additive include the types of base oil and additive, components and structural formulas that make up the base oil and additive, and ratios of the base oil and additive.

[0025] Examples of a base oil include mineral oils, synthetic oils, animal and vegetable oils and blends thereof.

[0026] Examples of an additive include plasticizers, stabilizers, oiliness agents, friction modifiers, wear inhibitors, metallic detergents, antioxidants, friction and wear inhibitors, extreme pressure agents, UV absorbers, lubricants, mold release agents, ashless dispersants, oiliness improvers, antistatic agents, rust inhibitors, antifoam agents, viscosity index improvers, metal deactivators and solid lubricants.

[0027] Examples of components that make up a base oil include components that make up oils generally used as base oils, such as mineral oils, synthetic oils, animal and vegetable oils, and blends thereof.

[0028] Examples of mineral oils include paraffin-based crude oils, naphthene-based crude oils, intermediate-based crude oils, aromatic crude oils, distillate oils, and refined oils. Examples of paraffin-based crude oils include isoparaffin and paraffin. Examples of naphthene-based crude oils include naphthene.

[0029] Examples of synthetic oils include poly-α-olefins, polyisobutylene (polybutene), monoesters, diesters, polyol esters, silicate esters, polyalkylene glycols, polyphenyl esters, silicones, fluorocompounds, alkylbenzenes and GTL base oils.

[0030] Examples of animal or vegetable oils include vegetable oils and fats such as castor oil, olive oil, cocoa butter, sesame oil, rice bran oil, safflower oil, soybean oil, camellia oil, corn oil, rapeseed oil, palm oil, palm kernel oil, sunflower oil, cottonseed oil and coconut oil, as well as animal oils and fats such as beef tallow, lard, milk fat, fish oil and whale oil.

[0031] Examples of components that form an additive include an ester compound of a monohydric or polyhydric aliphatic carboxylic acid and a monohydric or polyhydric aliphatic alcohol.

[0032] Examples of the monovalent aliphatic carboxylic acid used for the synthesis of an ester compound include saturated aliphatic carboxylic acids such as methane acid, acetic acid, propionic acid, butyric acid, pentanoic acid, caproic acid, heptanoic acid, octanoic acid, nonanoic acid, decanoic acid, undecanoic acid, dodecanoic acid, tridecanoic acid, tetradecanoic acid, pentadecanoic acid, hexadecanoic acid, heptadecanoic acid, octadecanoic acid, nonadecanoic acid, icosanoic acid, henicosanoic acid, docosanoic acid, tricosanoic acid, tetracosanoic acid, pentacosanoic acid, hexacosanoic acid, heptacosanoic acid, octacosanoic acid, nonacosanoic acid, and triacontanoic acid, etc.

[0033] Examples of the polyvalent aliphatic carboxylic acid used for the synthesis of an ester compound include saturated aliphatic carboxylic acids such as oxalic acid, malonic acid, succinic acid, glutaric acid, adipic acid, pimelic acid, suberic acid, azelaic acid, sebacic acid, undecanedioic acid, dodecanedioic acid, tridecanedioic acid, tetradecanedioic acid, heptadecanedioic acid and hexadecanedioic acid.

[0034] Examples of the monohydric aliphatic alcohol used for the synthesis of an ester compound include methanol, ethanol, propanol, butanol, pentanol, hexanol, heptanol, octanol, nonoanol, decanol, undecanol, dodecanol, tridecanol, tetradecanol, pentadecanol and hexadecanol.

[0035] Examples of the polyhydric aliphatic alcohol used for the synthesis of an ester compound include ethylene glycol, propylene glycol, neopentyl glycol, glycerin, trimethylolethane, trimethylolpropane, pentaerythritol and sorbitan.

[0036] The machine learning potential 30 is an interatomic potential that uses a machine learning method that outputs energy from information about the structure of an atom. Examples of the machine learning potential include a neural network potential (NNP), a Gaussian approximation potential (GAP), a spectral neighbor analysis potential (SNAP), and a moment tensor potential (MTP). The NNP is an interatomic potential obtained by learning a relationship between a structure (three-dimensional coordinates) of a molecule, a crystal, or the like, and energy or force corresponding to an output using deep learning. Among them, NNP is preferable as a machine learning potential in view of the high flexibility of the neural network. Matlantis (registered trademark) can be used as the NNP. (prediction device)

[0037] Fig. 4 is a functional block diagram illustrating a configuration of the prediction device 10. As shown in Fig. 4, the prediction device 10 comprises a detector 11, a single-molecule information detector 12, a liquid structure generator 13, an optimized structure detector 14, a feature calculator 15, a predictor 16, and an output module 17.

[0038] The detector 11 detects information about molecules of interest that are molecules contained in a base oil and detects information about an additive.

[0039] The name, structural formula, and the like of the molecules of interest can be used as information about them. SMILES or the like can be used as structural formulas. SMILES is a description of a molecular structure using a character string. An example of a table listing structural formulas (SMILES) is shown in Fig. 5. As shown in Fig. 5, SMILES of molecules of interest are listed. The table containing the structural formulas of molecules of interest can be obtained from data in CSV format, spreadsheet software such as Excel, etc. The recorder 11 can enter a table in which SMILES of potentially interesting molecules are written, as in Fig. 5 shown.

[0040] SMILES can be obtained from a chemical database such as PubChem (a chemical database provided by NCBI in the United States).

[0041] As in Fig. 4, the single-molecule information detector 12 detects a 3D structure of the molecule of interest detected by the detector 11 as single-molecule information.

[0042] The single-molecule information acquirer 12 includes a 3D structure generator 121 and a structure optimizer 122.

[0043] As in Fig. 6, the 3D structure generator 121 generates a D3 structure of a molecule of interest in which the hydrogen atoms are added by the 3D structure generator 121 to the single molecule structure of the molecule of interest.

[0044] The structure optimizer 122 acquires an optimized structure in which the 3D structure is optimized as single-molecule information. The structure optimizer 122 arranges the atoms present in the molecule of interest at appropriate positions, taking into account the relationship between the coordinates and the energy of the atoms, and obtains a structure in which the 3D structure is most stable in terms of energy.

[0045] The structure optimizer 122 creates an optimized structure, preferably using the machine learning capability 30. The 3D structure generator 121 can increase the processing speed by using the machine learning capability 30 and thus reduce the time required to create an optimized structure.

[0046] As in Fig. 4, the liquid structure generator 13 generates a liquid structure by arranging a freely determined number (N) of units of single-molecule information acquired by the single-molecule information acquirer 12 within a simulation cell, as shown in Fig. 7 shown.

[0047] A simulation cell is defined to have the shape of an arbitrary parallelepiped. The size of a simulation cell is freely determined according to the type of lubricant or the molecule of interest, etc.

[0048] N is not particularly limited and can be any suitable number, for example, on the order of several dozen to several hundred. If N is too small, the accuracy of the feature calculated by the feature calculator 15 decreases. If N is too large, the computational load becomes excessive and the computation time increases.

[0049] The liquid structure generator 13 arranges a freely determined number (N) of units of single-molecule information in a simulation cell at different positions such that the N units of single-molecule information do not overlap.

[0050] It is advantageous if the fluid structure generator 13 defines a simulation cell such that the actual structure of a lubricant is reproduced by making the density in the simulation cell substantially equal to the actual density of a base oil.

[0051] The liquid structure generator 13 can generate a liquid structure by randomly arranging and rearranging a freely determined number (N) of units of single-molecule information in a simulation cell.

[0052] As in Fig. As shown in Figure 4, the optimized structure detector 14 detects a fluid-optimized structure optimized by relaxing the fluid structure generated by the fluid structure generator 13. Relaxation is a general structure optimization method, such as a steepest descent method or a conjugate gradient method, for detecting a minimum value of energy in a multidimensional space. Relaxation means, for example, that a sum or scalar of force vectors, or a stress tensor or the principal component of the stress tensor, or the like acting on the entire fluid structure are brought into agreement below a certain threshold with a predetermined pressure (external pressure), which will be described later.The fluid structure generated by the fluid structure generator 13 does not have a correct density corresponding to a real state because the single-molecule information is inserted into a simulation cell of an arbitrary size. If the density is incorrect, the feature calculator 15 is very likely to calculate a feature with low accuracy. The optimized structure detector 14 can generate a fluid-optimized structure adapted to a fluid structure with a correct density by optimizing the fluid structure.

[0053] When optimizing the liquid structure, the optimized structure acquirer 14 preferably creates the liquid optimized structure under the following two conditions.

[0054] (1) When the optimized structure detector 14 optimizes the fluid structure, the volume of the simulation cell is variable. The optimized structure detector 14 can generate a fluid structure at a predetermined pressure (external pressure) by applying an external force to the simulation cell.

[0055] (2) The optimized structure detector 14 generates a structure under condition (1) as a structure at absolute zero temperature. To consider the influence of temperature, it is preferable to perform a molecular dynamics simulation.

[0056] The fluid-optimized structure is generated under the two conditions described above. The simulation cell is defined to reproduce the density of the actual base oil. In other words, the simulation cell is defined such that the fluid structure has a density when relaxed, taking into account a predetermined pressure and a predetermined temperature.

[0057] Examples of molecular dynamics simulation include structural optimization (molecular mechanics method), molecular dynamics method, and Monte Carlo method.

[0058] To capture a structure at absolute zero temperature, the optimized structure capture device 14 makes the volume of a simulation cell variable. The temperature can be set to any suitable value. The optimized structure capture device 14 can create a liquid structure at a predetermined pressure (external pressure) by applying an external force to the simulation cell.

[0059] The temperature and pressure are preferably high, for example, when simulating a lubricant in a contact state. The high temperature is preferably, for example, 50°C to 200°C, and more preferably about 60°C. The high pressure is preferably, for example, 50 MPa to 2000 MPa, and more preferably about 500 MPa, simulating real contact, for example, contact between microprotrusions on one surface and microprotrusions on another surface.

[0060] The optimized structure detector 14 detects a liquid-stable structure of a molecule of interest, preferably using the machine learning capability 30. The optimized structure detector 14 can increase the computation speed by using the machine learning capability 30 and thus shorten the time required to detect a liquid-stable structure.

[0061] The feature calculator 15 calculates a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure and calculating a radial distribution function (RDF) representing a frequency ratio of one molecule to another molecule at each atomic distance. In other words, the feature calculator 15 may use a feature represented by a radial distribution function (for example, a slope, a center point, or the like of an approximate representation of a radial distribution function described later) represented based on an atomic distance between atoms of different molecules, instead of an atomic distance within a molecule.

[0062] An atomic distance between atoms of different molecules is a distance between two atoms, each contained in two molecules. Hereinafter, one molecule is referred to as "molecule A" and another molecule as "molecule B." An atomic distance between atoms of different molecules can be a distance between an "atom a" contained in molecule A and an atom present at a position corresponding to atom a in molecule B, or a distance between atom a in molecule A and an atom present at a position corresponding to an atom in molecule B other than atom a.

[0063] The feature calculator 15 may include a radial distribution function calculator 151, a converter 152, and a parameter extractor 153.

[0064] The radial distribution function calculator 151 calculates an atomic distance between atoms of different molecules from a liquid-optimized structure and calculates a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance.

[0065] For example, a radial distribution function is calculated by the following equation (1): [Equation 1] gA−B(r)=〈NA−B〉4πr2⋅Δr⋅ρB (in which A is a molecule, B is another molecule different from molecule A, r is a radial distance seen from atom a of molecule A, g A-B (r) the probability of the presence of atom b of molecule B at a radial distance r, seen from atom a of molecule A, <N A-B > the ensemble mean of the number of atoms b of molecule B present in a region from r - 1 / 2Δr to r + 1 / 2Δr as seen from atom a of molecule A, Δr is a spherical region and ρB is a number density of B.

[0066] Fig. Figure 8 shows an example of a radial distribution function illustrating a relationship between atom a of molecule A and atom b of molecule B. A radial distribution function is obtained in which a probability density of the presence of molecule B increases from a constant state of 0 toward 1 as the distance from atom a of molecule A to atom b of molecule B increases, as shown in Fig. 8, and the value of the vertical axis for a Fig. 8, the horizontal axis area remains essentially constant at 0.95.

[0067] A radial distribution function gives the probability that atom b of molecule B exists at a distance r from atom a of molecule A, as in Fig. 9. In other words, a radial distribution function represents the frequency with which atom b of molecule B is positioned at a distance r from atom a of molecule A and correlates with an intermolecular distance between atom a of molecule A and atom b of molecule B.

[0068] A radial distribution function can be represented (plotted) in several dimensions instead of two.

[0069] The converter 152 converts the radial distribution function obtained from the radial distribution function calculator 151 into an approximate expression. The converter 152 can obtain an approximate expression of the radial distribution function by fitting the radial distribution function to a logistic function of the following equation (2) or the like. [Equation 2] y=a1+e−b(x−x0)+c (where y is a probability density for the presence of molecule B, a and c are coefficients, b corresponds to a slope of the approximate expression, x is a distance (Å) from an atom in molecule A to an atom in molecule B, and x0 is the position of the midpoint of the slope of the approximate expression.)

[0070] The parameter extractor 153 extracts a parameter of the approximate expression of the radial distribution function converted by the converter 152, and thereby calculates a feature indicating an intermolecular space. The parameter is preferably the center point or the slope of the approximate expression of the radial distribution function. In other words, when the converter 152 obtains the approximate expression of the radial distribution function using the logistic curve of the above equation (2), the position of the center point of the approximate expression of the radial distribution function is used as the parameter x0 and the slope of the approximate expression is used as the parameter b in the logistic curve of the equation (2). Therefore, as shown in Fig. 8, the center point or slope of the approximate expression of the radial distribution function can be used as a parameter of a feature indicating an intermolecular space, and the center point or slope of the approximate expression of the radial distribution function correlates with the atomic distance between atoms of different molecules and correlates with the degree of density of the oil film.

[0071] The center point of an approximate expression corresponds to the distance between molecules. The larger the center point, the farther the molecules are from each other, which means that a relatively thin oil film is formed and the adsorption time until an additive is adsorbed to the surface is short.

[0072] The slope of an approximate expression corresponds to the distance between molecules. The smaller the slope, the farther the molecules are from each other, which means that a relatively thin oil film is formed and the adsorption time until an additive is adsorbed to the surface is short.

[0073] The smaller a center point of the approximate expression or the larger a slope, the closer the molecules are to each other, which means that a relatively dense oil film is formed and the adsorption time until an additive is adsorbed to the surface is long.

[0074] Therefore, the adsorption time of an additive becomes shorter as a center point of an approximate expression becomes larger or a slope becomes smaller. If an additive can be adsorbed onto the sliding surface of the sliding part of an industrial machine or the like in a shorter time, the effect of the additive can be exerted and the sliding surface of the sliding part can be protected from strong friction, thereby achieving effects such as low friction and low wear. Therefore, it is preferable to adjust an oil film to have a loose ("sparse") structure by increasing a center point of an approximate expression or decreasing a slope to increase an intermolecular space.

[0075] As in Fig. 8, the parameter extractor 153 may use, as a feature indicating an intermolecular space, a first peak (a local maximum value that occurs first when viewed from short distance to long distance), a second peak (a local maximum value that occurs second when viewed from short distance to long distance), or a value on the horizontal axis at which the differential value of the radial distribution function asymptotically approaches substantially zero.

[0076] The feature calculator 15 uses a feature that correlates with an atomic distance between different molecules (for example, a slope, a midpoint, or the like of an approximate expression of a radial distribution function) instead of an atomic distance in a molecule as a feature that includes an intermolecular space between molecules of interest contained in the base oil. The feature calculator 15 can specify an intermolecular space by obtaining a radial distribution function that focuses only on an atomic distance between molecules, since completely different features appear depending on the type of molecule. For example, in the case of a radial distribution function of an ordinary normal alkane liquid, as shown in Fig. As shown in Figure 10, many CH bonds and CC bonds are present in a molecule, so sharp peaks derived from the CH bonds and CC bonds are present in a region where the distance between atoms in a molecule is short. In a region where the distance between molecules is large, a peak or the like related to an intermolecular pair is present, but such a peak is relatively weak compared to the atomic distance in a molecule and hardly noticeable. If the atomic distance in a molecule, such as a CH bond or a CC bond, is excluded, as in Fig. As shown in Figure 11, a radial distribution function is obtained in which only the distance between atoms of different molecules appears. A radial distribution function that only represents the distance between atoms of different molecules, such as isoparaffin, paraffin, and naphthene, has a completely different shape depending on the type of molecule and is therefore effective for specifying a space between atoms of different molecules.

[0077] The feature calculator 15 may use various physical properties such as viscosity, flash point, diffusion coefficient, and thermal conductivity of the base oil in addition to the center point or slope of the approximate expression of the radial distribution function as a feature indicating an intermolecular space.

[0078] The predictor 16 predicts a density level of an oil film formed from molecules of interest based on a size of a feature indicating an intermolecular space calculated by the feature calculator 15, such as the center point or slope extracted from the radial distribution function, specifically by measuring whether the feature of the molecules of interest is larger or smaller than a feature of a currently used base oil.

[0079] The output module 17 outputs the prediction result of the density of the oil film formed from the base oil consisting of the molecules of interest within the simulation cell, which is predicted by the predictor 16, by display, transmission or the like. (Hardware configuration of the prediction device 10)

[0080] Next, an example of a hardware configuration of the prediction device 10 will be described. Fig. 12 is a block diagram illustrating a hardware configuration of the prediction device 10. As shown in Fig. As shown in Fig. 12, the prediction device 10 is configured by an information processing device (computer) and may be physically configured as a computer system including a central processing unit (CPU; a processor) 101 as an arithmetic processing unit, a random access memory (RAM) 102 and a read-only memory (ROM) 103 as main storage devices, an input device 104 as an input device, an output device 105, a communication module 106, an auxiliary storage device 107 such as a hard disk, and the like. These are connected to each other via a bus 108. The output device 105 and the auxiliary storage device 107 may be provided externally.

[0081] The CPU 101 controls the overall operation of the prediction device 10 and performs various types of information processing. For example, the CPU 101 can predict the oil film density by performing a prediction process or executing a prediction program, which will be described later and is stored in the ROM 103 or the auxiliary memory 107.

[0082] The RAM 102 may include a non-volatile RAM that is used as the working area of ​​the CPU 101 and stores main control parameters and information.

[0083] The ROM 103 stores a basic input / output program and the like. The prediction program may be stored in the ROM 103.

[0084] The input device 104 is an input device such as a keyboard, a mouse, an operation button, a touch panel, or a display screen, which receives information input by a user as a command signal and outputs the command signal to the CPU 101.

[0085] The output device 105 is a display device such as a monitor display, a speaker, a printing device such as a printer, or the like. In the output device 105, for example, information such as the prediction result of the oil film density is displayed on a display device such as a monitor display, and a screen to be displayed is updated according to an input operation via the input device 104 or the communication module 106.

[0086] The communication module 106 is a data transmission / reception device such as a network card, and functions as a communication interface that receives information from an external data recording server or the like and outputs analysis information to another electronic device.

[0087] The auxiliary storage device 107 is a storage device such as a solid-state drive (SSD) and a hard disk drive (HDD), and stores, for example, various data, files, and the like required for the operation of the prediction device 10.

[0088] Each function of the prediction device 10 is realized by reading predetermined computer software (including a prediction program) from the main storage device such as the RAM 102 or the auxiliary storage device 107, and by executing the computer software by the CPU 101 to read and write to the main storage device such as the RAM 102 and / or the auxiliary storage device 107 or the like, and to operate the input device 104, the output device 105, and the communication module 106.

[0089] Therefore, every part of the Fig. 4 is realized by the cooperation of software and hardware by a processor executing predetermined computer software (including a prediction program) previously stored in a computer having the prediction device 10.

[0090] A computer program that performs at least some of the functions of the Fig. 4, can be installed in a memory of one or more computers. The CPU 101 of one or more computers can read a computer program installed in the central unit into a main memory and execute the computer program, thereby implementing the functions of the respective parts of the Fig. 4 shown prediction device 10.

[0091] The Fig. The prediction device 10 illustrated in Figure 4 may be implemented by one or more CPUs 101. Herein, CPU 101 may refer to one or more electronic circuits disposed on a single chip, or to one or more electronic circuits disposed on two or more chips or two or more devices. In the case where multiple electronic circuits are used, each electronic circuit may communicate by wired or wireless means.

[0092] The functions of the parts of the Fig. 4 may be executed by one computer or by a plurality of computers in a distributed manner. In the case where the functions of the respective parts of the Fig. 4 may be executed in a distributed manner by a plurality of computers, the plurality of computers may send and receive data over a communications network comprising a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or the Internet.

[0093] The prediction program may be stored, for example, in the main storage device or the auxiliary storage device 107 included in the computer. The prediction program may be stored in a computer connected to a communication line such as the Internet, and part or all of the prediction program may be downloaded via the communication line. Furthermore, the prediction program may be configured to be provided or distributed via a communication line.

[0094] The prediction program may be recorded (or installed) in the computer in a state where part or all of the prediction program is stored in a portable storage medium such as an optical disk such as a CD-ROM and a DVD-ROM, a semiconductor memory such as a flash memory, or the like. <vorhersageverfahren>

[0095] A prediction method according to the present embodiment will be described. The prediction method according to the present embodiment can be performed using the prediction system 1 described above. Therefore, the description of the already described matters will be partially omitted.

[0096] Fig. 13 is a flowchart illustrating the prediction method according to the present embodiment. As shown in Fig. 13, the prediction method according to the present embodiment is a prediction method for predicting a density degree of an oil film formed from a base oil consisting of molecules of interest.

[0097] In the prediction method according to the present embodiment, the detector 11 detects information about molecules of interest contained in the base oil and detects information about an additive (detecting step, step S11).

[0098] As described above, the name and structural formula of the molecules of interest can be used as information about the molecules of interest, and SMILES or the like can be used as the structural formula as described above.

[0099] Next, the single-molecule information acquirer 12 acquires a 3D structure of the molecules of interest acquired in the acquisition step (step S11) as single-molecule information (single-molecule information acquisition step, step S12).

[0100] The single-molecule information acquisition step (step S12) may include a 3D structure generation step (step S121) and a structure optimization step (step S122).

[0101] In the step of generating a 3D structure (step S121), the 3D structure generator 121 adds hydrogen to the single-molecule structure of the molecules of interest to generate a 3D structure of the molecules of interest in which hydrogen is added to the single-molecule structure of the molecules of interest (see Fig. 6).

[0102] In the structure optimization step (step S122), the structure optimizer 122 acquires an optimized structure in which the 3D structure is optimized as single-molecule information. In the structure optimization step (step S122), a structure in which the 3D structure is most stable in terms of energy is obtained by the structure optimizer 122 arranging the atoms constituting the molecule of interest at appropriate positions, taking into account the relationship between the coordinates and the energy of the atoms.

[0103] In the structure optimization step (step S122), the structure optimizer 122 preferably acquires an optimized structure using the machine learning potential 30. In the structure optimization step (step S122), the time required to generate the optimized structure can be shortened by using the machine learning potential 30.

[0104] Next, the liquid structure generator 13 arranges N units of single-molecule information, which are a freely selected number, acquired in the single-molecule information acquisition step (step S12), within the simulation cell as shown in Fig. 7, whereby a liquid structure is generated (the liquid structure generation step, step S13).

[0105] N is not particularly limited and can be any suitable number, for example on the order of several dozen to several hundred.

[0106] In the liquid structure generation step (step S13), the liquid structure generator 13 arranges a freely determined number (N) of units of single-molecule information in a simulation cell at different positions such that the N units of single-molecule information do not overlap.

[0107] In the step of generating a liquid structure (step S13), it is preferable that the liquid structure generator 13 defines a simulation cell so that a density of the base oil is reproduced in the simulation cell.

[0108] In the liquid structure generation step (step S13), the liquid structure generator 13 may generate a liquid structure by randomly arranging and rearranging a freely determined number (N) of units of single-molecule information in a simulation cell.

[0109] Next, the optimized structure detector 14 obtains a liquid optimized structure optimized by relaxing the liquid structure generated in the liquid structure generation step (step S13) (the optimized structure acquisition step, step S14).

[0110] The fluid structure generated in the fluid structure generation step (step S13) does not have a correct density corresponding to a real state because the single-molecule information is inserted into a simulation cell of an arbitrary size. If the density is incorrect, the feature calculation step (step S15) is very likely to result in a low-accuracy feature calculation. The optimized structure acquisition step (step S14) can generate a fluid-optimized structure adapted to a fluid structure with a correct density by optimizing the fluid structure.

[0111] In the optimized structure detecting step (step S14), the optimized structure detecting means 14 may generate the liquid optimized structure under the following two conditions when optimizing the liquid structure.

[0112] (1) When the optimized structure detector 14 optimizes the fluid structure, the volume of the simulation cell is variable. The optimized structure detector 14 can generate a fluid structure at a predetermined pressure (external pressure) by applying an external force to the simulation cell.

[0113] (2) The optimized structure detector 14 generates a structure under condition (1) as a structure at absolute zero temperature. To consider the influence of temperature, it is advantageous to perform a molecular dynamics simulation.

[0114] To capture a structure at absolute zero temperature, the optimized structure capture device 14 makes the volume of a simulation cell variable. The temperature can be set to any suitable value. The optimized structure capture device 14 can generate a fluid structure at a predetermined pressure (external pressure) by applying an external force to the simulation cell.

[0115] The temperature and pressure are preferably high, for example, when simulating a lubricant in a contact state. The high temperature is, for example, preferably 50°C to 200°C, and more preferably about 60°C. The high pressure is, for example, preferably 50 MPa to 2000 MPa, and more preferably about 500 MPa, simulating real contact, for example, contact between microprotrusions on one surface and microprotrusions on another surface.

[0116] The optimized structure acquisition step (step S14) acquires a liquid-stable structure of a molecule of interest, preferably using the machine learning capability 30. The optimized structure acquirer 14 can shorten the time required to acquire a liquid-stable structure by using the machine learning capability 30.

[0117] Next, the feature calculator 15 calculates a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure and calculates a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance (feature calculation step, step S15).

[0118] The feature calculation step (step S15) may include a radial distribution function calculation step (step S151), a conversion step (step S152), and a parameter extraction step (step S153).

[0119] In the radial distribution function calculation step (step S151), the radial distribution function calculator 151 calculates an atomic distance between atoms of different molecules from a liquid-optimized structure and calculates a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance.

[0120] For example, a radial distribution function is calculated by the following equation (1): [Equation 3] gA−B(r)=〈NA−B〉4πr2⋅Δr⋅ρB (in which A is a molecule, B is another molecule different from molecule A, r is a radial distance measured from atom a of molecule A, g A-B (r) the probability of the presence of atom b of molecule B at a radial distance r, seen from atom a of molecule A, <N A-B > the ensemble mean of the number of atoms b of molecule B present in a region from r - 1 / 2Δr to r + 1 / 2Δr, seen from atom a of molecule A, Δr is a spherical region and ρ B is a number density of B.

[0121] In the conversion step (step S152), the converter 152 converts the radial distribution function obtained in the radial distribution function calculation step (step S151) into an approximate expression. In the conversion step (step S152), the converter 152 can obtain an approximate expression of the radial distribution function by fitting the radial distribution function to a logistic function of the following equation (2) or the like. [Equation 4] y=a1+e−b(x−x0)+c (where y is the probability density for the presence of molecule B, a and c are coefficients, b corresponds to the slope of the approximate expression, x is the distance (λ) between an atom in molecule A and an atom in molecule B, and x0 is the position of the midpoint of the slope of the approximate expression.)

[0122] In the parameter extraction step (step S153), the parameter extractor 153 extracts a parameter of the approximate expression of the radial distribution function converted by the converter 152, thereby calculating a feature indicating an intermolecular space. The parameter is preferably the center point or the slope of the approximate expression of the radial distribution function (see Fig. 8). In other words, in the conversion step (step S152), in the case where the approximate expression of the radial distribution function is obtained using the logistic function of the above equation (2), the position of the center point of the approximate expression of the radial distribution function is used as the parameter x0 and the slope of the approximate expression is used as the parameter b in the logistic function of the equation (2). Therefore, as shown in Fig. 8, the center point or slope of the approximate expression of the radial distribution function can be used as a parameter of a feature indicating an intermolecular space, and the center point or slope of the approximate expression of the radial distribution function is correlated with the atomic distance between atoms of different molecules and correlated with the degree of density of the oil film.

[0123] As in Fig. 8, in the parameter extraction step (step S153), as a feature indicating an intermolecular space, a first peak (a local maximum value that appears first when viewed from short distance to long distance), a second peak (a local maximum value that appears second when viewed from short distance to long distance), or a value on the horizontal axis at which a differential value of the radial distribution function asymptotically approaches substantially zero of the radial distribution function may be used.

[0124] In the feature calculation step (step S15), the feature calculator 15 calculates a feature of the molecules of interest contained in the base oil using a feature correlated with the atomic distances between different molecules (for example, a slope, a center point, and the like, the approximate expression of the radial distribution function). In the feature calculation step (step S15), the feature calculator 15 can specify an intermolecular space by obtaining a radial distribution function that focuses only on an atomic distance between molecules, since completely different features appear depending on the type of molecule.

[0125] In the parameter extraction step (step S153), the feature calculator 15 may use various physical properties such as viscosity, flash point, diffusion coefficient, and thermal conductivity of the base oil in addition to the center point or slope of the approximate expression of the radial distribution function as a feature indicating an intermolecular space.

[0126] Next, the predictor 16 predicts a density degree of the oil film formed from the molecules of interest based on the feature calculated in the feature calculation step (step S15) (the prediction step, step S16).

[0127] Next, the output module 17 outputs, by display or the like, the prediction result of the density of the oil film formed from the base oil composed of the molecules of interest within a simulation cell, which is predicted in the prediction step (step S16) (output step, step S17).

[0128] As described above, the prediction system 1 comprises the prediction device 10, and the prediction device 10 comprises the detector 11, the single-molecule information detector 12, the liquid structure generator 13, the optimized structure detector 14, the feature calculator 15, and the predictor 16. The prediction system 1 detects a 3D structure of molecules of interest (see Fig. 6) as single-molecule information by the single-molecule information detector 12, generates a liquid structure in which the 3D structure of the molecules of interest is arranged within the simulation cell by the liquid structure generator 13 (see Fig. 7), and acquires a fluid-optimized structure in which the fluid structure is optimized by the optimized structure acquirer 14. In the prediction system, the feature calculator 15 calculates an atomic distance between atoms of different molecules from a fluid-optimized structure and calculates a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance. In the prediction system 1, the predictor 16 predicts a density degree of an oil film formed from the molecules of interest based on the calculated feature.

[0129] In the prediction system 1, a density degree of an oil film can be predicted by calculating a feature from a radial distribution function obtained from a fluid-optimized structure of molecules of interest, and thus, a passage state of an additive in the oil film can be specified. In the case where an atomic distance between molecules in an oil film formed from a base oil consisting of molecules of interest is large, gaps are likely to be generated. Since an oil film has a looser structure, an additive can migrate through the gaps in the oil film, and therefore, the additive can be quickly adsorbed onto a sliding surface of a sliding portion of an industrial machine or the like with which the oil comes into contact, and the effect of the additive, such as tribological properties, can be easily expressed.Therefore, the prediction system 1 can shorten the time required to select a base oil that produces an effect of an additive by predicting a density degree of an oil film formed from molecules of interest using a feature indicating an intermolecular space of the molecules of interest.

[0130] For example, when predicting the density level of an oil film formed from a base oil composed of molecules of interest using a diffusion / adsorption simulation commonly used in the prior art, it takes a long time (for example, on the order of several days to several months) to calculate a feature from a radial distribution function even for one type of base oil, and thus it takes longer to predict the density of an oil film. The prediction system 1 can calculate a feature of a type of base oil in a short time, for example, on the order of several tens of minutes to several hours, and thus predict the density level of an oil film in a shorter time.

[0131] The prediction system 1 includes the predictor 16, and thus, it is possible to reduce the effort and time required to predict, from molecules of interest, a characteristic related to a density of an oil film formed from a base oil composed of molecules of interest. In practice, in the manufacture of a lubricant, in order to produce a lubricant that meets the desired performance according to the type and use of the lubricant, a lubricant is prepared by combining various base oils and additives in a trial, and a density level check, etc., is performed. Since performing these steps determines the type of base oil that exhibits the effect of an additive, the composition of a base oil and an additive, and the like, a large amount of labor is required, and the cost burden is high due to the manufacture of various base oils and additives.The prediction system 1 can efficiently predict, from molecules of interest, a feature indicating an intermolecular space of the molecules of interest while reducing the effort, and thus reduce the effort in predicting a density degree of an oil film formed from a base oil composed of the molecules of interest.

[0132] In the prediction system 1, the single-molecule information acquirer 12 may include the 3D structure generator 121 and the structure optimizer 122. The 3D structure generator 121 generates a 3D structure in which hydrogen is added to molecules of interest contained in a base oil, and the structure optimizer 122 optimizes the 3D structure containing the molecules of interest contained in the base oil, and an optimized structure can thereby be obtained as single-molecule information. The 3D structure generator 121 can generate a 3D structure that appropriately represents a length, a size, an inclination, and the like of molecules contained in a base oil, and the structure optimizer 122 can generate a structure that appropriately reflects the relationship between the coordinates and the energy of molecules of interest contained in a base oil and is most stable in terms of energy.The prediction system 1 can more accurately provide a radial distribution function with the feature calculator 15 by using an optimized structure of molecules of interest obtained by the structure optimizer 122 as single-molecule information. Therefore, the prediction system 1 can more accurately calculate a feature of molecules of interest from a radial distribution function and thus more accurately predict a density of an oil film formed from a base oil composed of the molecules of interest. Therefore, the prediction system 1 can increase the accuracy in selecting a base oil that exhibits an effect of an additive.

[0133] Prediction system 1 can use machine learning capability 30 as structure optimizer 122. Thus, prediction system 1 can shorten the time required for structure optimizer 122 to create an optimized 3D structure of a molecule of interest. Therefore, prediction system 1 can shorten the time required to predict a density level of an oil film and thus shorten the time required to select a base oil that produces an additive effect.

[0134] In the prediction system 1, a simulation cell can be defined by the fluid structure generator 13 such that a density of a base oil is reproduced in the simulation cell. Thus, the prediction system 1 can approximate a fluid structure of molecules of interest to a state in which the molecules of interest are actually used as oil, for example, as a lubricant, and thus approximate a radial distribution function calculated by the feature calculator 15 to the state of the lubricant actually used. Therefore, the prediction system 1 can calculate a feature from a radial distribution function to correspond to a state of an actually used base oil of molecules of interest, and thus predict a density of an oil film formed from the base oil with higher accuracy. Therefore, the prediction system 1 can more appropriately select a base oil that exhibits an effect of an additive.

[0135] The prediction system 1 can use the machine learning capability 30 as the optimized structure detector 14. Accordingly, the prediction system 1 can shorten the time required to generate a fluid-optimized structure in the optimized structure detector 14. Therefore, the prediction system 1 can shorten the time required to predict a density level of an oil film and thus shorten the time required to select a base oil that produces an effect of an additive.

[0136] In the prediction system 1, the feature calculator 15 may include the radial distribution function calculator 151, the converter 152, and the parameter extractor 153. The feature calculator 15 may obtain a radial distribution function through the radial distribution function calculator 151, obtain an approximate expression by fitting the radial distribution function obtained through the converter 152, and use a parameter of the approximate expression of the radial distribution function for a feature based on the radial distribution function through the parameter extractor 153. Accordingly, the prediction system 1 can more easily and accurately calculate a feature from a radial distribution function, and thus predict an oil film density more easily and accurately. Therefore, the prediction system 1 can select a base oil that exhibits an effect of an additive with ease and increased accuracy.

[0137] The prediction system 1 can use a logistic function as an approximate expression of a radial distribution function in the parameter extractor 153, and a parameter can have a slope or a midpoint of the logistic function. Thus, the parameter extractor 153 can easily determine a slope or a midpoint of the approximate expression of the logistic function. Therefore, prediction system 1 can specify a feature in parameter extractor 153 more easily and accurately, and thus predict an oil film density more easily and accurately. Therefore, prediction system 1 can more easily and appropriately select a base oil that produces an additive effect.

[0138] As described above, the prediction system 1 can be used in selecting a base oil that exhibits the effect of an additive and is thus suitable for manufacturing a lubricant or the like. Since lubricants used in industrial machinery and the like are used under harsh conditions such as high pressure, high speed, high load, and high temperature, it is important to select a lubricant that can provide the desired lubricating performance even under such environmental conditions. The prediction system 1 is used in selecting a base oil that exhibits the effect of an additive, and thus it is possible to select a lubricant appropriately according to the use of various lubricants used for industrial machinery and the like.

[0139] The embodiments have been described above, however, they serve merely as examples, and the present invention is not limited to the embodiments. The above embodiments can be implemented in various other forms, and various combinations, omissions, substitutions, changes, and the like can be made without departing from the spirit of the invention. These embodiments and modifications thereof are included within the scope and spirit of the invention and are included in the invention described in the claims and the scope of their equivalents.

[0140] The embodiments of the present invention are, for example, as follows: <1> Forecasting system comprising: a detector configured to detect information concerning molecules contained in a base oil at normal temperature of 20°C and normal pressure of 1.013 × 10 5 Pa is liquid, to detect molecules of interest; a single-molecule information acquirer configured to acquire a 3D structure of the molecule of interest as single-molecule information; a fluid structure generator configured to generate a fluid structure by defining a simulation cell having an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information within the simulation cell; an optimized structure detector configured to detect a fluid optimized structure optimized by relaxing the fluid structure; a feature calculator configured to calculate a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from the liquid-optimized structure and calculating a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance; and a predictor configured to predict a density level of an oil film formed from molecules of interest based on the feature. <2> The forecast system according to <1> , wherein the single-molecule information acquisition device comprises: a 3D structure generator configured to generate a 3D structure of the molecule of interest by adding hydrogen to the molecule of interest; and a structure optimizer configured to acquire an optimized structure obtained by optimizing the 3D structure as single-molecule information. <3> The forecast system according to <2> , where the structure optimizer uses machine learning potential. <4> The prediction system according to one of the points <1> until <3> , where the fluid structure generator defines the simulation cell so that a density of the base oil is reproduced in the simulation cell. <5> The prediction system according to one of the points <1> until <3> , where the Optimized Structure Collector uses machine learning potential. <6> The prediction system according to one of the points <1> until <5> , where the feature calculator has: a radial distribution function calculator configured to calculate the radial distribution function; a converter configured to convert the radial distribution function into an approximate expression; and a parameter extractor configured to calculate the feature by extracting a parameter of the approximate expression. <7> The forecast system according to <6> , where the approximate expression is a logistic function and the parameter has a slope or a midpoint of the logistic function. <8> A prediction program that causes a computer to do the following: a detection step for detecting information concerning molecules contained in a base oil stored at normal temperature of 20°C and normal pressure of 1.013 × 10 5 Pa is liquid, as molecules of interest; a single-molecule information acquisition step for acquiring a 3D structure of the molecules of interest as single-molecule information; a liquid structure generation step for generating a liquid structure by defining a simulation cell having an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information within the simulation cell; an optimized structure detecting step for detecting a liquid optimized structure optimized by relaxing the liquid structure; a feature calculation step for calculating a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure, and calculating a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance; and a prediction step for predicting a density level of an oil film formed from the molecules of interest based on the feature. <9> A prediction technique in which a computer is configured to perform the following: a detection step for detecting information concerning molecules contained in a base oil stored at a normal temperature of 20°C and a normal pressure of 1.013 × 10 5 Pa is liquid, as molecules of interest; a single-molecule information acquisition step for acquiring a 3D structure of molecules of interest as single-molecule information; a liquid structure generation step for generating a liquid structure by defining a simulation cell having an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information within the simulation cell; an optimized structure detecting step for detecting a liquid optimized structure optimized by relaxing the liquid structure; a feature calculation step of calculating a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure, and calculating a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance; and a prediction step of predicting a density degree of an oil film formed from the molecules of interest based on the feature.

[0141] This application claims priority to Japanese Patent Application No. 2022-199612, filed with the Japan Patent Office on December 14, 2022, the entire contents of which are hereby incorporated by reference. LIST OF REFERENCE SYMBOLS 1 forecast system 10 Forecast device 11 authors 12 single-molecule information recorders 13 Fluid Structure Generators 14 Optimized Structure Capturers 15 feature calculators 16 Predictor 17 Output module 20 storage 30 Machine Learning Potential 121 3D Structure Generator 122 Structure Optimizers 151 Radial Distribution Function Calculator 152 converters 153 Parameter Extractor QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] JP 2022-158124

[0004] JP 2022-199612

[0141] Cited non-patent literature

[0000] M. Konishi and H. Washizu, Tribology International, 2020, Volume 149, Article 105568

[0006] < / vorhersageverfahren> < / vorhersagesystem>

Claims

[1] Forecasting system, comprising: a detector configured to detect information concerning molecules contained in a base oil at normal temperature of 20°C and normal pressure of 1.013 × 10 5 Pa is liquid, to detect molecules of interest; a single-molecule information detector configured to detect a 3D structure of the molecule of interest as single-molecule information; a liquid structure generator configured to generate a liquid structure by defining a simulation cell having an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information in the simulation cell; an optimized structure detector configured to detect a fluid-optimized structure optimized by relaxing the fluid structure; a feature calculator configured to calculate a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from the liquid-optimized structure and calculating a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance; and a predictor configured to predict the degree of density of an oil film formed by the molecules of interest based on the feature. [2] The prediction system according to claim 1, wherein the single-molecule information detector a 3D structure generator configured to generate a 3D structure of the molecule of interest by adding hydrogen to the molecule of interest; and a structure optimizer configured to acquire an optimized structure obtained by optimizing the 3D structure as the single-molecule information. [3] The prediction system of claim 2, wherein the structure optimizer uses a machine learning capability. [4] The prediction system according to claim 1, wherein the fluid structure generator defines the simulation cell so that a density of the base oil is reproduced in the simulation cell. [5] The prediction system of claim 1, wherein the optimized structure detector uses a machine learning capability. [6] The prediction system of claim 1, wherein the feature calculator comprises: a radial distribution function calculator configured to calculate the radial distribution function; a converter configured to convert the radial distribution function into an approximate expression; and a parameter extractor configured to calculate the feature by extracting a parameter of the approximate expression. [7] Prediction system according to claim 6, wherein the approximate expression is a logistic function and the parameter has a slope or a midpoint of the logistic function. [8] Prediction program that causes a computer to do the following: a detection step for detecting information concerning molecules contained in a base oil stored at normal temperature of 20°C and normal pressure of 1.013 × 10 5 Pa is liquid, as molecules of interest; a single-molecule information acquisition step for acquiring a 3D structure of the molecules of interest as single-molecule information; a liquid structure generation step for generating a liquid structure by defining a simulation cell having an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information within the simulation cell; an optimized structure detecting step for detecting a liquid optimized structure optimized by relaxing the liquid structure; a feature calculation step for calculating a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure and calculating a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance; and a prediction step of predicting a density degree of an oil film formed from the molecules of interest based on the feature. [9] A method of prediction in which a computer is configured to perform the following: a detection step for detecting information concerning molecules contained in a base oil stored at a normal temperature of 20 °C and a normal pressure of 1.013 × 10 5 Pa is liquid, as molecules of interest; a single-molecule information acquisition step of acquiring a 3D structure of the molecule of interest as single-molecule information; a liquid structure generation step for generating a liquid structure by defining a simulation cell having an arbitrary parallelepiped shape and arranging an arbitrary number of units of the single-molecule information within the simulation cell; an optimized structure detecting step for detecting a liquid optimized structure optimized by relaxing the liquid structure; a feature calculation step for calculating a feature indicating an intermolecular space by calculating an atomic distance between atoms of different molecules from a liquid-optimized structure and calculating a radial distribution function representing a frequency ratio of one molecule to another molecule at each atomic distance; and a prediction step of predicting a density degree of an oil film formed from the molecules of interest based on the feature.

Images