198 results about "Molecular simulation" patented technology

The invention discloses a coal seam chemical outburst elimination method based on molecular dynamics, which belongs to the field of coal mine safety, and comprises the following steps: sampling and analyzing to obtain the element proportion, group composition and chemical bond connection mode of a coal sample, constructing a real coal molecular skeleton, and reducing and optimizing the molecular structure of coal by molecular simulation; the method comprises the following steps: adding a surfactant, setting the real gas content, gas pressure and environment temperature of a coal seam, establishing a micro model after the surfactant is injected into the coal seam, simulating the occurrence state and reaction process of methane molecules in the coal seam under the action of the surfactant, and determining the methane content in the coal seam by combining the migration characteristics of the methane molecules and the energy change of the system. And the concentration of the injected surfactant is optimized, so that the engineering application of large-range industrial water injection outburst elimination of the coal seam is developed. According to the method, through molecular simulation, the complex occurrence characteristics of the coal seam are considered, the concentration of the injected surface active agent is optimized, and the hydraulic measure efficiency is improved.

The invention relates to the field of pesticide composition screening, and discloses a stabilizer screening process for improving the low-temperature stability of a chlorantraniliprole and lufenuron suspending agent, which comprises the following steps: a) analyzing the low-temperature behavior of chlorantraniliprole and lufenuron based on molecular simulation, constructing a candidate stabilizer molecule library, and preliminarily screening candidate stabilizers; b) applying low-temperature stress to the chlorantraniliprole and lufenuron suspending agent sample containing the candidate stabilizer by using a microfluidic platform, and performing in-situ dynamic characterization to obtain dynamic data related to the low-temperature stability of the sample; and c) processing the dynamic data, and establishing an evaluation index representing the low-temperature stability performance of the candidate stabilizer. According to the invention, a cognitive bridge from a micromechanism to a mesoscopic phenomenon is constructed by integrating molecular simulation pre-screening and microfluidic high-throughput dynamic characterization. The problems of long research and development cycle and resource consumption caused by a large number of trial and error and experience judgment in a traditional screening method are effectively solved.

The invention discloses a molecular simulation method and device, a nonvolatile storage medium and electronic equipment. The method comprises the following steps: acquiring molecular system information of a current molecular configuration; calculating the first energy and the first gradient of the key atom region by adopting a variable component sub-feature solicitation solver algorithm, and calculating the second energy and the second gradient of the non-key atom region; wherein the first gradient is used for expressing the stress condition of each atom in the key atom area; the second gradient is used for expressing the stress condition of each atom in the non-key atom area; and updating the current molecular configuration according to the first energy, the first gradient, the second energy and the second gradient to obtain a target molecular configuration. According to the method and the device, the technical problem that the accuracy of the simulation result cannot be ensured under the condition that the computing resources are limited due to excessive computing resources consumed by the quantum chemical computing mode when the molecules are simulated by adopting the quantum chemical computing mode in the related technology is solved.

The invention provides a method for identifying a microscopic association body structure in a heavy oil molecule simulation trajectory, which comprises the following steps of: analyzing a molecular dynamics trajectory, identifying an aromatic ring structure in a heavy oil molecule component, calculating a geometric center of the aromatic ring structure, combining a periodic boundary condition, and applying an improved DBSCAN clustering algorithm to realize automatic identification and clustering of an association body; and performing post-processing optimization on the clustering result, merging the shared clusters to generate a final association body set, and analyzing the number, size distribution, morphological characteristics and dynamic evolution information of the association bodies based on the structural characteristics of the final association body set. The method can accurately capture pi-pi accumulation behaviors of heavy oil molecules, further reveals morphological characteristics and dynamic evolution laws of heavy oil microassociates and influences of the heavy oil microassociates on the apparent viscosity of heavy oil, provides theoretical support for heavy oil viscosity-causing mechanism analysis and viscosity reducer design, can complement experimental research, and has a wide application prospect. And efficient development and green utilization of a complex thickened oil system are promoted.

The invention relates to the technical field of machine learning potential molecular simulation, and provides a montmorillonite thermal conductivity prediction method based on NEP machine learning potential, and the method comprises the following steps: constructing a montmorillonite crystal structure model, carrying out first principle molecular dynamics calculation on the montmorillonite crystal structure model, introducing lattice distortion and atomic disturbance, and generating an initial structure set; performing static single-point DFT calculation on the initial structure set, and constructing a training data set; based on an NEP machine learning framework, utilizing the training data set to train and generate a potential function; optimizing the generation process of the potential function by adopting an active learning strategy; and performing thermal conductivity prediction verification on the potential function, and obtaining a thermal conductivity prediction result of the montmorillonite based on the potential function passing the verification. The defect that a traditional classical potential function depends on an empirical or semi-empirical model and is limited in precision and applicability is overcome, the problem that the dynamic behavior and the heat conduction mechanism of a complex material are difficult to accurately capture is solved, the calculation cost is reduced, and meanwhile the precision is kept.

The invention discloses a bubble initial temperature calculation method of an oil paper system and a related device, and the method comprises the steps: collecting operation state parameters of an in-service transformer during operation, carrying out the molecular simulation calculation according to the operation state parameters, and obtaining the micro-water aggregation rate of water molecules on an oil paper interface of an insulating oil paper system; guiding the flow rate of liquid water in the oil-paper system finite element model entering the insulation paper finite element model through the micro-water aggregation rate, and calculating the bubble initial temperature of the oil-paper system through the oil-paper system finite element model. The technical problems of low measurement efficiency and large measurement result error in the prior art are solved.

The invention relates to the technical field of molecular simulation, and provides a method for constructing a heavy component average molecular structure of thickened oil, which comprises the following steps: inputting parameters, generating a core aromatic ring unit, constructing an inner-layer aliphatic ring unit, constructing an outer-layer aliphatic ring unit, generating a side chain unit, and storing and outputting structural information. A plurality of molecular structures meeting experimental data are generated at a time through a traversal algorithm thought, composition characteristics of a heavy oil complex mixture are met, more available molecular structures are provided for molecular simulation, the accuracy and efficiency of molecular modeling are improved, the molecular structures completely meeting the experimental data can be rapidly generated, and the method is suitable for large-scale popularization and application. And only the mass spectrometer, the elemental analyzer, the nuclear magnetic resonance hydrogen spectrum and the infrared spectrum data need to be input, so that the experiment cost is reduced.

The invention relates to the technical field of molecular simulation, and discloses a molecular simulation analysis method of a functionalized covalent organic framework confinement ionic liquid microstructure, and the method specifically comprises the following steps: converting force field files of COF and anions and cations of ionic liquid into coordinate files by using a Python tool; the anions and the cations are confined to a pore channel formed by COF stacking to form a composite membrane, and the whole system is placed in a simulation box; data and input files are generated according to the potential file, and LAMMPS software is used for conducting molecular dynamics simulation on a set simulation box; the two-dimensional density, cation orientation and the like of the ionic liquid are analyzed through a Python post-treatment molecular dynamics track, and meanwhile the porosity and the free volume of the COF are calculated. According to the method, a regulation and control mechanism of the COF confinement ionic liquid can be obtained, and a theoretical basis is provided for design and preparation of a COF confinement ionic liquid membrane material.

The invention provides a method for predicting a polyimide stress threshold. The method comprises the following steps: 1) constructing a polyimide full-atom molecular model by adopting molecular simulation software; 2) performing mechanical tensile simulation on the molecular model to calculate a stress-strain curve; 3) calculating a creep curve of the molecular model under different constant stresses according to a stress-strain curve calculation result, and determining a mechanical failure interval according to the creep curve; and 4) selecting different stress values in a mechanical failure interval by using a dichotomy to carry out creep curve iteration, and finally determining a stress threshold. The method provided by the invention not only can efficiently predict the stress thresholds of the polyimides with different structures, but also can simulate and analyze the creep behavior of the polyimides in the tensile stress process, and provides a theoretical basis for guiding the structural design and mechanical property optimization of polyimide materials.

The invention discloses a method for screening an MOFs material with excellent CH4 adsorption separation performance based on high-throughput calculation, a method for improving the screening efficiency and a preparation method. The screening method comprises the following steps: constructing a screening database of MOFs material structure information; calculating physical and chemical characteristic parameters by utilizing computational chemistry and material simulation software, and further setting a screening standard aiming at CH4 adsorption; carrying out preliminary screening through a high-throughput screening technology; predicting the competitive adsorption behavior of the primarily selected material in the mixed gas by means of a molecular simulation technology, and performing secondary screening; and finally, determining the final MOFs material in combination with an actual application environment. According to the method, the high-performance MOFs material is rapidly and accurately screened out from a large number of materials, the research and development period is effectively shortened, the development cost is reduced, an efficient and feasible solution is provided for resource utilization of ultralow-concentration CH4 in coal mine windblown gas, and energy utilization and environmental protection are assisted.

The invention provides a molecular simulation method and device for interaction behaviors of vulcanized rubber powder and asphalt, and relates to the technical field of molecular simulation, and the method comprises the following steps: respectively constructing a rubber powder molecular model and an asphalt molecular model through molecular simulation software; establishing a rubber powder modified asphalt system model according to the rubber powder molecular model and the asphalt molecular model; distributing molecular force fields to the rubber powder molecular model, the asphalt molecular model and the rubber powder modified asphalt system model; performing energy minimization processing and relaxation operation on the rubber powder modified asphalt system model, and calculating thermodynamic parameters and particle motion trails of the rubber powder modified asphalt system model; and evaluating the interaction effect of the vulcanized rubber powder and the asphalt according to the thermodynamic parameters and the particle motion trail of the rubber powder modified asphalt system model. According to the molecular simulation method, the interaction between the vulcanized rubber powder and the asphalt is systematically explained from the molecular level.

The invention belongs to the technical field of molecular simulation, and discloses a method and system for obtaining the heat conductivity coefficient of polyimide composite aerogel. The method comprises the following steps: establishing an initial structure of polyimide composite aerogel; establishing an initial structure model of the polyimide composite aerogel; establishing a nitrogen molecular structure model, and determining a heat transfer model of the polyimide composite aerogel containing nitrogen molecules; setting simulation calculation parameters, and carrying out molecular dynamics simulation calculation; calculating mean square displacement; and calculating the heat conductivity coefficient of the polyimide composite aerogel in the heat transfer model of the polyimide composite aerogel containing the nitrogen molecules according to the mean square displacement. According to the method, gaseous heat transfer is simulated on the polyimide composite aerogel, mean square displacement of a system at different temperatures is analyzed, and the heat conductivity coefficient of the aerogel can be directly obtained. The result can provide a certain reference for the heat transfer micromechanism of various materials, and effective thermal protection is obtained.

The invention discloses an interactive catalysis method and platform based on molecular simulation and artificial intelligence. According to the method, intelligent design of a catalyst is achieved through data classification, condition optimization, structure modeling and mechanism analysis. Catalytic data uploaded by a user is subjected to standardization processing and then is divided into two types: a data set A is used for recommending an optimal reaction condition by a multi-objective optimization algorithm; a three-dimensional model is constructed after a catalyst structure of the data set B is optimized through quantum chemistry software, a plurality of intermediates and transition state structures are optimized, and geometric / electronic descriptors are calculated. And constructing a dual-model collaborative prediction system of the model E and the model F through a feature screening and dimension reduction technology, and finally screening the high-performance catalyst by utilizing a model F driven molecule generation algorithm. According to the method, flexible adaptation to specific application scenes can be achieved, the urgent requirements of the industry for efficient and accurate catalyst performance prediction and intelligent design are met, cost reduction and efficiency improvement of the industry are effectively assisted, and the research and development process of the catalyst is accelerated.

The invention discloses a molecular simulation and flavor perception-based beer umami peptide efficient identification method, and relates to the technical field of food flavor analysis and bioinformatics. The method comprises the following steps: performing liquid-liquid extraction and mass spectrum analysis on a beer sample to obtain a polypeptide sequence; screening the umami active peptide based on a UMPred FRL model; the method comprises the following steps: constructing a T1R1 / T1R3 receptor homologous model, and screening a peptide fragment of which the binding energy is lower than-25kcal / mol by adopting a CDOCKER semi-flexible docking mode; screening a stable peptide of which the delta Gbind is lower than-50kcal / mol through molecular dynamics simulation and MM / GBSA calculation; and finally, the umami characteristic is verified by combining an electronic tongue and a sensory test. The screened peptide fragment can be used for improving umami perception and taste balance of beer.

The invention discloses a method and system for generating a three-dimensional pore atom model based on preset target pore parameters, and relates to the technical field of petroleum engineering and molecular simulation, and the method comprises the steps: receiving the target pore parameters, and preparing an initial model; on the basis of the target pore parameters, a three-dimensional pore channel center path and pore channel section dimensions changing along the path are generated in the initial atomic structure model in a programmed mode; removing atoms to form an initial pore channel; and carrying out atomic-scale structure optimization on the model with the initial pore channel, and checking and removing atoms which do not meet a preset coordination condition to obtain a three-dimensional pore atom model with a stable structure. The method can efficiently and controllably generate the micro-pore model which meets the requirements in the statistical sense and is stable in structure based on high-level parameters, and provides a high-quality initial structure for researches such as molecular dynamics simulation and the like.

The invention discloses an interface thickness parameter acquisition method based on a microcosmic full-atom composite model, and belongs to the field of molecular simulation. The method comprises the following steps of: firstly, constructing a periodic model of a polymer A and a periodic model of a polymer B by utilizing MS software, and after the periodic models are combined, obtaining a global energy minimization total-atom composite model through geometric optimization and dynamic equilibrium optimization; finally, atomic concentration distribution data points of the full-atomic composite model are calculated in an MS software Perl script module through the secondary programming technology, an atomic concentration distribution curve is drawn, and smooth processing is conducted; setting a judgment benchmark of the interface thickness of the full-atom composite model, and judging whether the interface thickness exists in the full-atom composite model or not; and defining an interface thickness calculation formula of the full-atom composite model, and calculating the interface thickness of the full-atom composite model with the interface thickness, so as to quickly, efficiently, accurately and quantitatively calculate the thickness result of the interface layer of the polymer composite material. In addition, the method has the advantages of being convenient to operate and the like.

The invention discloses a molecular simulation-based method for estimating CO2 / CH4 adsorption capacity in a shale reservoir, which comprises the following steps of: firstly, selecting different mineral components from the shale reservoir, and determining reservoir temperature and pressure corresponding to each mineral component; the method comprises the following steps: constructing molecular models of different mineral components in a shale reservoir and a to-be-detected gas molecular model by using molecular simulation software, and carrying out geometric optimization; setting temperature and pressure conditions conforming to the actual reservoir in the simulation environment, and introducing gas molecules to perform molecular dynamics simulation so as to research the competitive adsorption behavior and obtain adsorption quantity data; based on simulation data, an adsorption capacity calculation expression comprehensively considering the coupling influence of multiple factors such as temperature and pressure is established, and the actual shale reservoir gas adsorption capacity is rapidly and accurately estimated. Through molecular simulation, the limitation that a traditional experiment is long in period, high in cost and difficult to separate multi-factor influences is overcome, and an efficient and reliable theoretical tool is provided for shale gas reserve prediction and carbon dioxide geological storage benefit evaluation.

A method and system for calculating the free energy difference between a target state and a reference state. The method includes determining one or more intermediate states using a coupling parameter, performing molecular simulations to obtain ensembles of micro-states for each of the system states, and calculating the free energy difference by an analysis of the ensembles of micro-states of the system states. The method can be particularly suited for calculating physical or non-physical transformation of molecular systems such as ring-opening, ring-closing, and other transformations involving bond breaking and / or formation. A soft bond potential dependent on a bond stretching component of the coupling parameter and different from the conventional harmonic potential is used in the molecular simulations of the system states for the bond being broken or formed during the transformation.

The invention relates to the technical field of safe conveying of oil and gas pipelines, in particular to a polypeptide type hydrate kinetic inhibitor screening method based on molecular simulation, which comprises the following steps: selecting target protein related to hydrate growth, and designing a polypeptide sequence; performing polypeptide pre-screening; evaluating the binding affinity and binding mode of the polypeptide and the target protein; performing MD simulation on the screened high-affinity polypeptide; verifying the binding affinity of the polypeptide and the target protein; and collecting all experiment and simulation data for statistical analysis, and optimizing the polypeptide sequence. The structure and function of the target protein are systematically analyzed by using a bioinformatics database and tools, it is ensured that the selected target plays an important role in hydrate formation, and meanwhile, a polypeptide sequence designed based on the target protein structure can effectively improve the binding capacity; and in the pre-screening and docking analysis process of the polypeptide, dynamic parameters are simulated and extracted by molecular dynamics, so that the interaction between the polypeptide and the target protein can be deeply understood.

The invention discloses a method for screening and extracting DES of tea saponin based on molecular simulation and machine learning, and relates to the technical field of tea saponin extraction.The method comprises the steps that tea saponin, cellulose, a hydrogen bond donor, a hydrogen bond receptor and a DES molecular model are constructed and optimized, and a molecular polarity index is obtained; the method comprises the following steps: generating a topology, obtaining a trajectory and energy through MD simulation, and calculating E (TeaS-DES), E (TeaS-Cell) and SI; and measuring the extraction rate, combining the molecular polarity index, SI, process parameters and the like to construct a data set to train a machine learning model, predicting the extraction rate, and locking the optimal DES and process. According to the method, DES screening can be directly guided through simulated energy and structure descriptors, the number of experiments is reduced, and the research and development period is shortened.

The present invention efficiently calculates molecular energy corresponding to a plurality of interatomic distances.　An information processing device (10) estimates, for interatomic distances (14a, 14b, 14c), estimated execution times (15a, 15b, 15c) and estimated repetition counts (16a, 16b, 16c) of an algorithm that calculates molecular energies using quantum circuit data, on the basis of molecular information (13). The information processing device (10) determines repetition counts (18a, 18b, 18c) for the interatomic distances (14a, 14b, 14c) on the basis of a restriction time (17), the estimated execution times (15a, 15b, 15c), and the estimated repetition counts (16a, 16b, 16c). The information processing device (10) calculates, by executing the algorithm, molecular energies (19a, 19b, 19c) for the interatomic distances (14a, 14b, 14c) on the basis of the repetition counts (18a, 18b, 18c).

The invention belongs to the field of computational material science, artificial intelligence algorithm and nano composite material performance prediction, and particularly relates to a conductivity prediction system and method based on three-dimensional voxelization representation and a graph neural network, and the system comprises a voxelization module, a graph construction module, a graph neural network prediction module and an interpretability analysis module which are sequentially connected according to a process. The voxelization module converts three-dimensional molecular simulation structure data into a discrete local density field; the graph construction module generates a voxel graph structure; the graph neural network prediction module adopts an end-to-end design, is composed of a cascade convolution attention layer and the like through incremental feature learning, and outputs a conductivity prediction result; the interpretability analysis module reveals a structure-performance relationship. The method comprises the steps of data acquisition, voxelization processing, voxel graph structure construction and input model prediction and analysis. According to the method, the conductivity of the nanocomposite can be efficiently and accurately predicted, and powerful support is provided for material design and optimization.

Provided is a system for predicting compatibility between different kinds of resin materials, the method being capable of predicting compatibility such as mutual solubility and interfacial adhesion between constituent materials in a blend material prepared by mixing a plurality of resin materials. The system is characterized by comprising: a three-dimensional molecular model creation unit that creates a three-dimensional molecular model of each of a plurality of resin materials using, as input information, information obtained by digitizing the chemical structures of the resin materials; a molecular entanglement model creation unit that, using the three-dimensional molecular model, temperature, and pressure as input information, creates a molecular entanglement model of different kinds of resin materials in which two kinds of the three-dimensional molecular models of each of the resin materials are combined; a mutual diffusion coefficient calculation unit that executes a molecular simulation on the basis of the molecular entanglement model and calculates a mutual diffusion coefficient between the different kinds of resin materials; a fracture energy calculation unit that performs a molecular simulation on the basis of the molecular entanglement model and calculates a fracture energy; a trajectory saving unit in which the time evolution of the coordinates of the molecular entanglement model in the process of calculating the fracture energy is saved; and a simulation result output unit that outputs the mutual diffusion coefficient and the fracture energy as numerical information and outputs the time evolution of the coordinates as a moving image.

The invention belongs to the field of molecular dynamics, and provides a molecular dynamics adjacency list construction optimization method, a molecular dynamics adjacency list construction optimization system and a super computer platform in order to solve the problem that the cache hit rate is reduced when a traditional adjacency list is constructed. The molecular dynamics adjacency list construction optimization method comprises the following steps: mapping atomic coordinates in each physical space to three-dimensional integer grid coordinates; carrying out one-dimensional mapping coding on the three-dimensional integer grid coordinates of all the atoms and sorting all the atoms; clustering all the sequenced atoms to obtain a plurality of atom clusters and initializing sharing states in the atom clusters; constructing a candidate neighbor region of each atomic cluster based on the initialized sharing state in each atomic cluster; and uniformly traversing the candidate neighbor regions of each atomic cluster according to a vectorization instruction, constructing an adjacency list of each atom of each atomic cluster, and writing the adjacency list into a global adjacency list, so that larger-scale and higher-efficiency molecular simulation is realized.

The invention discloses an ACE (angiotensin converting enzyme) inhibitory peptide with antihypertensive activity and application thereof, and belongs to the technical field of active peptide biology, amino acid sequences of the ACE inhibitory peptide are EYPVK and SYPVK, the two peptides are obtained through module replacement design, the ACE inhibitory activity of the two peptides is remarkably improved compared with that of an original peptide LYPVK, IC50 values are respectively reduced to 14.82 + / -0.82 mu M and 12.83 + / -0.79 mu M from 117.14 + / -1.40 mu M, and the inhibitory activity is improved by 13.5% and 25.1%. Molecular simulation analysis shows that the interaction force between the two and ACE is stronger, and a potential modular effect exists between sequences. Besides, the polypeptide has remarkable blood pressure lowering activity on human umbilical vein endothelial cells, plays a role by regulating release of NO and ET-1, has the activity superior to that of original peptides, has the advantages of good digestive enzyme stability, small molecular weight and good water solubility, and is suitable for developing related products for lowering blood pressure.

The invention discloses a method for analyzing a fiber and resin interface area based on molecular simulation, and belongs to the field of building a fiber and resin interface model based on a molecular simulation technology, and carrying out geometric optimization and dynamic balance on the interface model; counting and presenting the fiber surface conformation of the interface model; constructing a model of independent resin molecular chains and fibers, and explaining reasons of conformational changes of the resin molecular chains; based on the radial distribution function and the mean square displacement curve, analyzing the interface structure of the interface model and the activity of molecules; and constructing an interface model for interface nanoindentation simulation, and analyzing the stress bearing capacity of different areas of the interface and the motion displacement condition of atoms. According to the method, conformation and performance analysis of the fiber and resin interface are realized, favorable support can be provided for design, optimization and performance prediction of the fiber reinforced composite material, and the method has relatively good practicability.

This invention relates to a method for constructing a machine learning potential energy model based on hierarchical active learning, comprising the following steps: constructing a database of background solvent molecules and lithium salt molecules; initializing a machine learning potential energy model committee; outer active learning automatically exploring the chemical species composition space of the electrolyte, generating electrolyte formulations and initial structures; determining whether a structure needs to be included in the labeling range based on uncertainty calculations; automatically constructing a liquid phase environment based on classical force field simulation methods; performing high-precision simulations based on first-principles calculations and labeling the energy and force values ​​of the structures; inner active learning driving the machine learning potential energy model to explore the configuration space and label the energy and force values ​​of structures with high uncertainty; obtaining a first-principles structure database of the real liquid phase environment through several iterations; and training to obtain the final high-performance machine learning potential energy model. This invention systematically and comprehensively samples the chemical species composition space of the electrolyte and the molecular simulation configuration space based on hierarchical active learning, thereby ensuring the predictive performance and simulation accuracy of the machine learning potential model on unknown and complex electrolyte systems.

The application discloses a method for predicting polypeptides based on machine learning, comprising: predicting the binding site of PEX5 of the receptor; constructing a polypeptide sequence and determining the initial state of the polypeptide; generating the space required for polypeptide and PEX5 binding site collision simulation; performing polypeptide and PEX5 binding site collision simulation; performing data preprocessing; learning the known classified polypeptides through sklearn and outputting the combination of neural networks with high accuracy; reconstructing the neural network from sklearn through the neural network of Pytorch, learning the known classified polypeptides, predicting the unknown classified polypeptides, and outputting the confidence. The application can make accurate prediction of polypeptides by using various computer simulation methods such as molecular simulation and combining machine learning, can sufficiently reduce the order of magnitude of wet experiments, and thus reduce the cost and improve the efficiency. The method can also be used for screening the interaction of other functional polypeptides, drug polypeptides and corresponding target points.

The present invention relates to the technical field of heavy oil development, and particularly relates to a method for predicting the composition and yield of supercritical multi-component thermal fluids for heavy oil thermal recovery. By using a molecular simulation platform to simulate the reaction of reaction molecular systems with different organic matter concentrations, simulation results similar to those of physical experiments can be obtained at an economic cost and time cost lower than that of physical experiments, and the composition and yield data of supercritical multi-component thermal fluids under different reaction conditions can be obtained. Moreover, a relationship model between the composition and yield and many reaction condition factors is established through a machine learning modeling method, which is convenient for studying the influence of each reaction condition factor on the composition and yield of supercritical multi-component thermal fluids. At the same time, the purpose of predicting the composition and yield of supercritical multi-component thermal fluids according to the levels of reaction condition factors can be achieved. In addition, the method combining molecular simulation and machine learning has strong plasticity, and more factors can be added according to the research process and requirements to further improve the accuracy and applicability of the prediction model.

The invention discloses a molecular dynamics simulation method for electron-induced gas desorption, and belongs to the technical field of molecular simulation. The method comprises the following steps: firstly, constructing an aluminum oxide molecular model, and then simulating adsorption equilibrium of hydrogen on the surface of aluminum oxide at different temperatures and air pressures by adopting a giant regular Monte Carlo method to obtain a stable adsorption configuration; then, based on the configuration, an electron force field is adopted to introduce explicit electron particles to simulate electron beam bombardment, and an electron-induced gas desorption kinetic process is simulated under a molecular dynamics framework; and finally, performing quantitative analysis on the desorption product. Through full-atom molecular dynamics simulation, coherent research from gas adsorption to electron-induced desorption is realized, the limitation that related experimental measurement is difficult and a micromechanism is difficult to reveal under a high-voltage condition is overcome, and an effective simulation tool is provided for deeply understanding a physical mechanism of dielectric surface gas desorption-induced discharge.