51 results about "Potential energy surface" patented technology

The invention discloses a method for exploring a heterogeneous catalytic reaction network based on a deep potential energy surface model, and the method comprises the steps: carrying out the accurate prediction of the free energy change of all possible element reactions on a catalyst through the combination of transfer learning and a density functional theory, and constructing a reaction network diagram based on a graph theory method; an improved Dijkstra method is used for searching a reaction network diagram to determine a reaction mechanism, and the selectivity of the catalyst is explored based on a Markov chain method. According to the technical scheme provided by the invention, when the CO2 reduction reaction mechanism on the monatomic alloy is explored, the prediction error MAE of the electron energy of each intermediate is only 0.16 eV, the calculated amount is reduced by 90%, the dominant path of the CO2 reduction reaction is visually displayed, and a new normal form is provided for complex mechanism exploration and catalyst screening.

The invention relates to an intelligent state monitoring method and system for a high-voltage isolation switch operating mechanism, and relates to the technical field of power equipment fault diagnosis, and the method comprises the steps: enabling a collected real-time sequence to update a standard operation response potential energy surface, obtaining a real-time operation response potential energy surface, generating a phase response difference tensor through data projection, and obtaining a real-time operation response potential energy surface; and depicting deviation characteristics of actual and standard operation states. And then, in the constructed graph attention network, on the basis of a real-time sequence, monitoring parameters and video motion characteristics, constructing a multi-modal sequence state graph, and determining a state typing result through fine-grained recognition and interpretable analysis of a fault. And finally, introducing a cyclic fingerprint residual model, performing clustering optimization based on a real-time sequence, determining an abnormal deviation index, and combining the abnormal deviation index with a state typing result to determine a fault diagnosis result. Therefore, the accuracy of intelligent state monitoring of the high-voltage isolation switch operating mechanism can be effectively improved, and the method can be suitable for non-ideal environments, low dominant anomalies and other conditions.

The invention discloses a method for predicting the coverage degree of an H ion negative electrode in a Zn aqueous battery, and relates to the technical field of surface catalysis. In order to solve the defect that in the prior art, it is difficult to accurately predict the coverage degree of H ions on the surface of a Zn negative electrode on a multi-scale level, the technical scheme provided by the invention comprises the steps that a three-dimensional atom structure model of H atoms and a Zn electrode adsorption surface is constructed, formation energy and adsorption energy of the H ions and water molecules are calculated, and a reaction potential energy surface is established; constructing a reaction path and an electron transfer behavior, defining potential related parameters, and outputting reaction rate, current density and coverage data; establishing a phase field dilute material transfer model containing temperature, pH and pressure parameters, and completing material migration and charge distribution calculation; interface physical field parameters are fed back to the microdynamics model to form microcosmic-macroscopic bidirectional data interaction and realize self-consistent convergence; and outputting a coupling model result to obtain multi-scale prediction data. The method is suitable for active site identification and performance prediction in electro-catalytic reaction.

This application provides a method and apparatus for training a molecular multiconformation prediction model. The method includes: acquiring training data, which includes molecular structure representation samples and molecular conformation samples; processing the molecular structure representation samples and molecular conformation samples using the molecular multiconformation prediction model to be trained to obtain a training output; constructing a loss function value based on the training output and the molecular conformation samples; and, if the loss function value or the number of training rounds is within a preset range, using the trained molecular multiconformation prediction model as the molecular multiconformation prediction model. This application can achieve fast and accurate prediction of multiple three-dimensional conformations of molecules without analyzing their potential energy surfaces, reducing the computational resources and time consumed in resolving multiple three-dimensional conformations of molecules.

The invention provides an artificial neural network potential energy function development method for simulating a dynamic process of a molecular sieve packaged platinum cluster in an oxygen environment, which comprises the following steps of: performing potential energy surface sampling on a system of a Silicalites molecular sieve packaged platinum nanocluster and O2 and Pt-O structures by using a random surface walking global optimization method of LASP software; a part of structures are randomly extracted from a data set obtained through sampling, VASP software is used for structural optimization, and the obtained structures and data such as energy and force of the structures are added into a DFT training set. Neural network training is carried out on the DFT data set by means of machine learning, and an artificial neural network potential energy function G-NN is obtained; the artificial neural potential energy function obtained by training not only meets the expectation in calculation precision, but also has the advantage of magnitude order in calculation speed, and overcomes the defects of traditional molecular dynamics in calculation systems and calculation speed.

The invention belongs to the technical field of computational chemistry and artificial intelligence crossing, and provides a molecular potential energy surface prediction method based on deep learning. The method comprises the following steps: inputting a structure file of an unoptimized molecule, and reading and predicting energy and force of a current molecular structure by utilizing a machine learning potential model; and then, judging whether the structure is converged or not according to a prediction result by adopting a GeomeTRIC algorithm, and if not, adjusting coordinates to generate a new structure and performing circular prediction until the structure is converged. According to the method, the quantum chemistry and the deep learning technology are combined, the calculation efficiency can be greatly improved, complex inter-atomic interaction in molecules can be fully captured, and the prediction accuracy is ensured; and meanwhile, batch processing and automatic generation of a quantum chemistry software interface are supported, and the method is particularly suitable for high-throughput molecular structure optimization in the fields of drug discovery, material science and chemical product design.

The invention relates to the technical field of computational chemistry and drug screening, and discloses a fluorinated derivative screening method based on computer-aided design, which comprises the following steps: acquiring an initial binding conformation of fluorine-containing molecules and a target system, and constructing a follow-up local orthogonal coordinate system by taking the axial direction of a fluorocarbon bond as a reference; projection coordinates of the environmental particles relative to the fluorocarbon bonds are calculated and mapped as a spherical coordinate index, a preset anisotropic potential energy surface grid is called, and a potential energy value is calculated through interpolation; the method comprises the following steps: generating an anisotropic correction term, monitoring the modulus length of the energy gradient in real time, and when the modulus length of the energy gradient exceeds a sensitive threshold value, adaptively and nonlinearly reducing a sampling step length and executing high-frequency local sampling in a sensitive area. The problem of false negative activity prediction caused by insufficient sampling when a traditional equal-step algorithm is used for processing anisotropic distribution of fluorine atom electron clouds is solved, and the recognition precision and screening efficiency for weak directional interaction such as halogen bonds are improved.

This invention discloses a multi-objective adaptive control method for a power generation, grid, load, and storage system in an industrial park, relating to the field of energy management technology. The method first constructs a generalized state space of physical equipment and virtualized production tasks, mapping the production tasks to a virtual energy storage model and constructing a congestion risk potential surface function. Then, it calculates and decomposes the total regret value of the strategy, dynamically updates the objective function weights based on external regrets, and corrects the virtual model parameters online based on internal regrets. Subsequently, it constructs a collaborative optimization proposition using the updated weights and parameters, incorporating congestion risk into the optimization objective, and solves for the optimal control sequence. Finally, it decouples the control sequence into physical charging / discharging commands and production operation commands. This invention achieves compatibility between discrete production and continuous energy control through virtualization mapping and utilizes a two-layer regret mechanism to achieve adaptive adjustment of strategies and parameters, improving the system's economy and robustness.

The application discloses a calculation method of a reaction mechanism of cyanamide prepared in an alkaline environment and application thereof in byproduct control, and the calculation method comprises the following steps: step 1, constructing a molecular initial model of a reaction system containing a hydroxyl water molecular cluster, performing structural optimization and vibration analysis tasks on the model by using a density functional theory method, and calculating the molecular configuration to obtain a reasonable configuration without virtual frequency; step 2, searching for a reaction transition state based on the reasonable configuration of step 1, and verifying the reasonableness of the transition state; step 3, calculating the free energy in an alkaline aqueous solution environment, and drawing a potential energy surface diagram to determine the reaction rate control step; and step 4, selecting the reaction calculation hydrogen dissociation energy and hydrogen combination energy values of the reaction rate control step containing an intramolecular hydrogen migration process. The calculation method can simulate and calculate phenomena that are difficult to observe by experimental means, obtain reaction information of each elementary reaction process in an actual reaction, obtain the essential law of the reaction mechanism in the alkaline environment, and provide a theoretical reference for process improvement to reduce byproduct generation and promote the selectivity of the main reaction.

The invention discloses a tug system maintenance operation risk supervision method and system based on the Internet of Things, and relates to the technical field of the Internet of Things, and the method comprises the steps: collecting multi-physical-quantity data, accessing the Internet of Things, carrying out the preprocessing of original data, and generating standard input; inputting the preprocessed data to a coupling resolver, capturing characteristics among physical quantities, and dividing a coupling state; generating risk factors based on abnormal coupling, and constructing a first safety potential energy surface and a second safety potential energy surface; comparing the two types of potential energy surface data, and outputting execution admission judgment of the maintenance step; and collecting the track of the maintenance personnel, comparing with the path template, and triggering the behavior lock if deviation occurs. According to the invention, through a cooperation mechanism of multi-physical quantity coupling analysis, potential energy surface admission determination and personnel trajectory supervision, real-time risk identification and safe controllable execution of the whole tug maintenance operation process are realized.

The invention discloses a method for searching a minimum energy path from a discrete potential energy surface and a related product, the method is applied to a discrete potential path optimizer, and the method comprises the steps that an energy coordinate data set in an input file is read, and each energy coordinate data in the energy coordinate data set comprises a three-dimensional coordinate and an energy value corresponding to the three-dimensional coordinate; the energy coordinate data set is simplified based on a preset energy cutoff value, a simplified energy coordinate data set is obtained, and the energy value corresponding to each three-dimensional coordinate in the simplified energy coordinate data set is smaller than or equal to the energy cutoff value; searching energy coordinate data corresponding to a minimum energy path from the simplified energy coordinate data set based on a path connection condition between a preset starting point coordinate and a preset end point coordinate; and sorting the energy coordinate data corresponding to the minimum energy path based on preset grid precision to obtain the minimum energy path corresponding to the discrete potential energy surface. Therefore, the minimum energy path can be searched from the discrete potential energy surface.

A machine learning potential energy surface model construction method comprises the following steps: collecting data for training a model, the data comprising data of one or more systems, the data of each system comprising atomic coordinates of the system and corresponding properties of the system; performing data inspection and data storage on the collected data; constructing a machine learning potential energy surface model, comprising: establishing a feature engineering; obtaining a final data set for training the model through the feature engineering; constructing a machine learning model; and training the machine learning model using the final data set to obtain the machine learning potential energy surface model. By using a data-driven method, the present application avoids manual feature extraction, and the model autonomously iterates learning from the data set, thereby avoiding the introduction of human bias. The model of the present application has very strong scalability, can be scaled to larger systems, and is not limited by the properties of the target system, and simultaneously supports multi-phase systems and periodic / non-periodic systems.

The present invention proposes an outlier-aware adaptive mixed-precision quantization method and system for an interatomic potential energy surface model. By using mixed-precision quantization with different data widths for coefficients of different-order terms, the outlier-aware quantization method proposed in the present invention can effectively improve the efficiency of quantization representation. The present invention also proposes a mechanism and method for systematically controlling the errors in polynomial representation caused by quantization, and proposes a method for adaptively selecting the optimal quantization precision for mixed-precision quantization of coefficients of different-order terms under the constraints of error control. By analyzing the data of the polynomial coefficient table of the interatomic potential energy model, the present invention adaptively provides an optimal mixed-precision quantization scheme and performs outlier-aware mixed-precision quantization operations on the polynomial coefficient table based on this scheme, thereby improving the inference scale and speed of the interatomic potential energy model.

The application discloses a kind of optimization method and device for identifying Cu-Zn alloy cluster structure potential energy surface, comprising: randomly generating several Cu-Zn alloy clusters, constitute population structure library n pop ; develop shell script that can efficiently call Gaussian16 quantum chemistry software, carry out first local structure optimization to n pop ; innovative surface atom self-adapting reorganization strategy is proposed, iteratively move cluster surface atoms to self-adapting generated empty space sites, execute second stage structure optimization;Through crossover, variation operation, in combination with similarity detection and potential energy comparison, extract the optimal structure of binary Cu x Zn y Cluster.Using the technical scheme of the application, the local and global spatial search capability of genetic algorithm is optimized and improved, and the stability and electronic structure of Cu-Zn alloy cluster can be efficiently analyzed by calling quantum chemistry software.

The invention discloses a diacetylene topological chemical reaction method which comprises the following steps: spin-coating a diacetylene solution on a substrate to form a coating film; and irradiating the coating film by adopting an infrared light source, and reacting to generate polydiacetylene. Compared with side reaction and degradation caused by the fact that high-energy rays enable the material to enter an electron excitation state and material fragmentation caused by severe spatial change of molecules under high polymerization, mid-infrared light vibration excitation in the method serves as a means capable of triggering molecular vibration under an electron ground state, and under the mild condition, the mid-infrared light can be used as a means capable of triggering molecular vibration under the electron ground state. Therefore, various bad effects generated in the electron excitation state can be avoided.

This invention provides a method for the controllable synthesis of high-quality quantum dots based on pre-nucleated clusters that reach a minimum potential energy surface through chemical self-assembly, belonging to the field of semiconductor nanomaterials technology. The quantum dots of this invention are prepared by reacting a pre-nucleated sample of semiconductor material with pre-prepared quantum dots to obtain high-quality quantum dots grown heterogeneously (core-shell structure) or homogeneously. This method, based on the pre-nucleated clusters, allows for the control of quantum dot size growth, narrowing the quantum dot distribution and avoiding the need to control the bonding temperature of the semiconductor material. This method improves the accuracy of quantum dot size control, and the quantum dots prepared by this method also exhibit significant luminescent properties, effectively expanding the applications of quantum dots and showing great promise.

A method for training and subsequently executing a machine learning network of both an autoencoder and a machine learning model in the context of machine learning inter-atomic potential is disclosed. The systems described herein are configured to embed atomic locations and categories of a given atomic system, and apply them to an autoencoder to learn auxiliary attributes, and to a machine learning model to learn local energy. The auxiliary attributes are then used to generate an auxiliary Hamiltonian description. Attributes, such as the total energy of the atomic system, are determined by combining both the auxiliary Hamiltonian description and the local energy. By processing machine learning via both an autoencoder and a machine learning model, such methods ensure that both long range and short range effects are considered while also appropriately achieving realistic discontinuities and / or transitions within potential energy surfaces.

The invention relates to the cross field of computational chemistry and artificial intelligence, in particular to a multiphase chemical reaction transition state certainty generation method and system based on physical constraint and adaptive optimization and a storage medium. The method is executed by a computer system and comprises the following steps: acquiring initial structure data of reactants and products; calling a nonlinear initial guess module to execute path search based on the coarse-grained potential energy surface so as to generate a nonlinear initial guess path; a constraint optimal transmission module is called to generate a candidate transition state TS structure; the invention provides a multiphase chemical reaction transition state certainty generation method and system based on physical constraint and adaptive optimization and a storage medium, which not only keep high efficiency of a machine learning method, but also realize accurate, robust and reliable simulation of complex and real chemical scenes.

The invention belongs to the field of molecular simulation, and particularly relates to a method for constructing a reactive force field based on a multi-pole moment model. Comprising the following contents: a charge penetration effect, a polarization effect, a charge transfer effect, Van der Waals interaction and a three-body potential energy form suitable for a chemical reaction; a dynamic electric multi-pole moment method suitable for chemical reaction; the invention relates to a method for constructing a potential energy surface data set with six degrees of freedom. The invention provides a simple and effective force field model for researching chemical reaction and a parameter optimization method for users. According to the method, a potential energy surface data set composed of a large number of conformations with six degrees of freedom can be generated, and a data basis is provided for model development and parameter optimization. The method is mainly used for researching the chemical reaction process and the mechanism thereof, and has a wide application prospect in the research fields of chemistry, material science and the like.

This invention relates to the field of engineering machinery safety technology, specifically disclosing a load stability prediction system for telescopic boom forklifts based on the force exerted by the boom. The system collects motion sensing data and stress wave sensing data at the boom hinge; calculates a cross-spectral coherence function based on the motion data to generate a hinge coherence disorder index; performs time-frequency transformation on the stress wave data and automatically identifies energy impact patches, extracting their peak energy, center frequency, and duration; and generates a transient collision intensity spectrum index based on sensitive frequency band mapping and aggregation; constructs a time-series comprehensive feature vector from the two indices, inputs it into a pre-trained multi-layer gated recurrent unit structure for nonlinear dynamic system identification, and outputs a fused stability assessment state quantity; maps this state quantity to a virtual potential energy surface for multi-step forward extrapolation, calculates the dynamic stability margin value, and issues an early warning when the margin is lower than an adaptive threshold; this invention achieves early and accurate prediction of nonlinear jump instability.

The invention discloses a torsion regulation-based ultrafast broadband optical limiting material design method, a material and application, and relates to the field of optical limiting materials.The torsion regulation-based ultrafast broadband optical limiting material design method comprises the steps that structure optimization is conducted on double-branched-chain polycyclic aromatic hydrocarbon molecules through a computer based on the density functional theory, then the vibration frequency of each molecule is calculated, and the vibration frequency of each molecule is calculated; ensuring that the molecular configuration of the double-branched-chain type polycyclic aromatic hydrocarbon molecule after structure optimization is at the lowest point of a potential energy surface, and obtaining the double-branched-chain type molecule with a polycyclic conjugated structure, wherein the dihedral angle between a core group and a side chain group is 90 degrees; and adjusting a pi bridge between the core group with the polycyclic conjugated structure and the side chain group, so that a dihedral angle between the core group and the side chain group is gradually changed from 90 degrees to 0 degree, thereby obtaining the ultrafast broadband optical limiting material. The research of the invention clearly clarifies the anti-saturated absorption enhancement mechanism of the two-branch anthracene derivatives, and lays a foundation for the development of novel organic optical limiting materials.

A method for training and subsequently executing a machine learning network of one or more machine learning (ML) models is disclosed. The systems described herein are configured to embed atomic locations and categories of a given atomic system and apply them to ML model (s) to learn charge transfer attributes and local energy. By constructing atomic charges from the learned charge transfer attributes, local and global charge neutrality is ensured. Atomic charges are then used to generate an auxiliary Hamiltonian description. Attributes, such as the total energy of the atomic system, are determined by combining the auxiliary Hamiltonian description and the learned local energy. By determining the total energy from the local energy described and learned by the auxiliary Hamiltonian, such a method ensures that both long range and short range effects are considered while also appropriately achieving true discontinuities and / or transitions within potential energy surfaces.

Methods for a machine learning network that trains and subsequently executes one or more machine learning (ML) models are disclosed. The system described herein is configured to embed atomic positions and species of a given atomic system and apply those to ML model(s) to learn charge transfer properties and local energies. By constructing atomic charges from learned charge transfer properties, both local and global charge neutrality is ensured. The atomic charges are then used to generate an auxiliary Hamiltonian description. By combining both the auxiliary Hamiltonian description and the learned local energies, properties such as total energy of the atomic system are determined. By determining total energy from the auxiliary Hamiltonian description and the learned local energies, such methods ensure that long and short range effects are accounted for, while also appropriately enabling for realistic discontinuities and / or transitions within the potential energy surface.

This invention discloses a material force field fitting method based on potential energy surface matching, comprising: S1 obtaining the corresponding chemical structural formula of the target material; S2 constructing a chemical structural model of the connecting bonds in the chemical structural formula based on the atomic hybridization state of the chemical structural formula; S3 performing preliminary structural optimization on the chemical structural model to obtain the energy-minimizing structure; S4 calculating the chemical structural parameters of the target material using the DFT method based on the energy-minimizing structure; S5 obtaining the Morse bond stretching potential parameter using the Morse algorithm based on the chemical structural parameters and saving it in the molecular simulation file; S6 repeating S2-S5 until the calculation of all chemical structural models is completed, and integrating all molecular simulation files into the force field file; S7 performing molecular simulation based on the force field file obtained in S6 to calculate the mechanical properties of the target material. This method, by introducing the Morse bond stretching potential parameter, makes the simulation results of fiber materials more consistent with reality.

The invention discloses a calculation method of a correction map, electronic equipment, a storage medium and a program product, and relates to the technical field of molecular dynamics. The method comprises the following steps: performing dynamic simulation on target protein molecules to obtain a target dynamic model; the target dynamic model is composed of a plurality of square blocks; determining a target atom chain formed by connecting a target number of atoms in the target kinetic model; determining a target reference atom in each target atom chain, determining a target block where the target reference atom is located, and executing a correction map calculation task of the target atom chain based on a target calculation component of the target block; and in response to a completion instruction of the calculation task of each target atomic chain, obtaining a calculation result of the correction map of the target protein molecule. According to the scheme, calculation of the correction map of the protein structure can be achieved, and help is provided for ensuring that the force field can more accurately reproduce the potential energy surface derived by quantum and keeping the real protein conformation as time goes on.

The present invention discloses a method for optimizing vacancy-type defect structures at interfaces. The main steps include: obtaining the lattice position and affected region with minimum vacancy formation energy, deleting atoms in the region that may be affected by vacancy occupation; dividing the affected region after the atoms are deleted into a grid; inserting atoms into the divided grid; optimizing the occupied positions of the inserted atoms on the grid; and relaxing the optimized occupied positions to obtain the vacancy ground-state atomic structure. This optimization method does not restrict the atomic lattice position and can search for vacancy structures at all positions in a certain spatial region. It can effectively access the potential energy surface characteristics near the interface and obtain the ground-state structure of the vacancy.

The present application relates to the intersection field of computational chemistry and artificial intelligence, and in particular to a multi-phase chemical reaction transition state certainty generation method based on physical constraints and adaptive optimization, a system and a storage medium; The method is executed by a computer system, and the method comprises the following steps: obtaining initial structure data of reactants and products; calling a nonlinear initial guess module to perform path search based on a coarse-grained potential energy surface to generate a nonlinear initial guess path; and calling a constraint optimal transport module to generate a candidate transition state TS structure; The present application provides a multi-phase chemical reaction transition state certainty generation method based on physical constraints and adaptive optimization, a system and a storage medium, which not only maintains the high efficiency of the machine learning method, but also realizes accurate, robust and reliable simulation of complex and real chemical scenes.

The application relates to a solid-state electrolyte prediction screening method and system based on an improved bond valence site energy method, which comprises the following steps: obtaining the crystal structure of a solid-state electrolyte, constructing a three-dimensional calculation grid of a migration ion in a unit cell by using the crystal structure, calculating a global potential energy surface by using an improved BVSE model, and determining a migration energy barrier; the improved BVSE model is obtained by introducing a Coulomb repulsion term between migration ions in an original BVSE model; identifying migration ion sites based on the global potential energy surface, calculating the site energy of each migration ion site, obtaining characteristic Coulomb repulsion energy, and performing equivalent surface analysis and simple harmonic vibration analysis to obtain an equivalent surface surrounding volume and a site vibration frequency; and comparing the migration energy barrier, the characteristic Coulomb repulsion energy, the equivalent surface surrounding volume and the site vibration frequency with preset screening criteria to screen out potential solid-state electrolyte materials. The application can promote high-throughput screening of solid-state electrolyte materials.

The application provides a method for evaluating explosive safety by neural network potential calculation, comprising the following steps: S1, obtaining a microstructure image of a binder explosive by in-situ transmission electron microscopy (TEM); S2, establishing a molecular model of an energetic material and a molecular model of a binder according to the obtained microstructure image and modifying the microstructure image; S3, performing neural network training on data obtained by simulation, fitting a potential energy surface of the molecular model of the energetic material and the molecular model of the binder; S4, simulating evolution of the explosive with time under the action of a load; and S5, analyzing simulation results. The method for evaluating explosive safety by neural network potential calculation can evaluate all explosives that have been applied and have application potential by using the binder explosive real microstructure safety calculation method based on the neural network potential, greatly promotes application of new explosives, and brings great convenience to ammunition safety engineering personnel.