118 results about "Unstructured mesh" patented technology

The embodiment of the invention relates to the technical field of data modeling analysis, and discloses a tunnel rock stratum large deformation analysis method and system based on three-dimensional modeling, and the method comprises the steps: collecting initial three-dimensional point cloud data through three-dimensional laser scanning; constructing an initial three-dimensional geologic model containing spatial morphological characteristics and a rock stratum interface geometric topological relation through space coordinate matching and curved surface reconstruction; finite element unstructured mesh generation is carried out on the model, and a rock stratum mechanical analysis model is established in combination with physical and mechanical parameters; inputting construction parameters and process information, activating unit nodes in sequence, and solving a force balance equation to simulate stress distribution; identifying a potential deformation area through iterative calculation, and adjusting boundary conditions to predict a deformation trend; the three-dimensional space distribution cloud atlas, the time evolution curve and the position marks are fused to generate multi-view deformation early warning information, the accuracy and the visualization degree of tunnel rock stratum deformation analysis are improved, and support is provided for construction safety.

The invention discloses a coastal sound tomography data assimilation method based on stream function fitting. The method comprises the following steps: arranging a coastal sound tomography system, obtaining underwater sound data containing observation domain hydrological information, carrying out correlation peak extraction, and calculating to obtain a sound ray mutual return transmission time difference; scanning topographic information of an observation domain and temperature and profile data between stations, inputting the topographic information and the temperature and profile data into a sound velocity model for sound ray simulation, determining an actual propagation path of sound rays, and inverting path average flow; constructing a three-dimensional unstructured grid of the observation domain, and determining a projection relation between an actual propagation path of sound rays and the grid; establishing a state transition matrix by taking a sound ray path average flow as an observation vector of extended Kalman filtering and taking a coefficient matrix of a flow function least square fitting ocean numerical pattern calculation result as a state vector; and assimilating and updating by using an extended Kalman filtering algorithm, and obtaining an optimized flow field. According to the invention, the observation precision, the resolution and the spatial range of the coastal flow field can be obviously improved.

The invention discloses a space field prediction method, device and equipment based on a large model and a weighted loss function, and relates to the field of fluid space field dynamic prediction. The method comprises the following steps: carrying out unstructured grid division on an object model for generating a three-dimensional fluid space field, simulating to generate space field data of an irregular grid at a plurality of historical moments, and mapping the space field data into a regular grid; calculating a time sequence mutual information amount of any two historical moments, and reserving space field data of the historical moments with the time sequence mutual information amount greater than a time sequence dependency degree threshold so as to assist in forming a training data set; determining the weight of each regular grid according to the Euclidean distance between each regular grid and the analysis reference line, and obtaining a weighted loss function; training a hybrid convolutional network according to the training data set and the weighted loss function, and constructing a large space field prediction model; and an iterative prediction mechanism is adopted to obtain space field distribution at future moments. The precision, efficiency and adaptability of dynamic prediction of the fluid space field can be improved.

A worm and worm gear pair heat-fluid-solid coupling analysis method comprises the following steps that 1, a heat-fluid-solid coupling simulation analysis model of a worm and worm gear pair is built, and a three-dimensional geometric model is built based on the worm and worm gear meshing principle; 2, parameterized grid division is carried out on the simulation analysis model, a tetrahedral unstructured grid is adopted for a fluid domain, and a worm and worm gear tooth surface area is densified; 3, based on the fluid dynamics theory and an RNG # imgabs0 # turbulence model, establishing an oil-immersed lubricating oil gas two-phase flow transient simulation model, and solving fluid dynamic pressure and a convective heat transfer coefficient through a pressure-speed coupling algorithm; 4, establishing a three-dimensional transient dynamic simulation model based on a solid mechanics theory, combining a temperature-stress coupling field equation, adopting an explicit dynamic algorithm, and iteratively solving thermal deformation and stress strain of the worm and the worm gear through a finite element method; and 5, bidirectionally transmitting the pressure of the fluid domain, the convective heat transfer coefficient and the thermal deformation data of the solid domain through the fluid-solid interface, and carrying out iterative solution until convergence.

The invention belongs to the technical field of engineering simulation data processing, and particularly discloses a finite element grid graph structure construction method and system, a terminal and a medium. Comprising the following steps: analyzing original full-amount grid data exported by finite element simulation, segmenting unstructured grid data into a grid vertex coordinate set and a grid unit mark number set, and constructing a graph edge topological structure corresponding to a grid based on a unit mark number relationship; on the basis, loading physical field data and adopting a tolerance-based coordinate matching algorithm to realize accurate mapping of physical field labels and material attributes with grid nodes; and generating standardized finite element grid graph data which can be directly used for graph neural network processing. By means of the method, high-consistency and high-physical-reliability conversion from the finite element simulation data to the graph structure data is achieved, and the physical field modeling and simulation acceleration capacity based on graph learning is improved.

A three-dimensional temperature gradient field visualization method comprises the following steps: after analyzing a part model and storing a point-surface relationship and temperature information, calculating a data point temperature gradient and carrying out point cloud, grid and entity model visualization; after creating temperature field color mapping and calculating a temperature field contour line, representing a temperature gradient field based on a point icon, representing a temperature gradient field based on a streamline and representing a temperature gradient field based on texture; through design of a data visualization core algorithm and model building, visualization of three-dimensional data of various vector fields for an unstructured grid model is realized.

The invention provides a GPU-based parallel tetrahedral mesh optimization method, system and device and a medium, and relates to the technical field of numerical simulation. Comprising the steps of generating a triangular mesh covering a surface for a three-dimensional geometric model; generating an initial tetrahedral mesh based on the triangular mesh; performing a round of point smoothing operation on each tetrahedron by using a plurality of GPU threads; carrying out iterative circulation of the combination optimization operation, and simultaneously carrying out a round of point smoothing, edge deletion and multi-surface deletion operation on a plurality of initial tetrahedrons by utilizing a plurality of GPU threads; and when the edge deletion operation and the multi-surface deletion operation cannot further improve the grid quality, simultaneously performing a round of edge contraction and edge segmentation operation on the plurality of initial tetrahedrons by utilizing the plurality of GPU threads. According to the method, the large-scale grid with reliable quality is quickly generated based on the low-quality input grid, and the grid optimization rate is improved, so that the requirement of large-scale unstructured grid optimization in a three-dimensional space is met.

The invention discloses a silicon carbide continuous sintering furnace performance optimization method based on a dynamic grid technology. The method comprises the following steps that a three-dimensional simulation model of a continuous sintering furnace load silicon carbide workpiece is established, unstructured grid division is conducted, and grid encryption treatment is conducted on a heating area, the silicon carbide workpiece and a cooling water system inlet and outlet; performing three-dimensional transient heat transfer calculation, and setting a dynamic area by combining the dynamic grid model to obtain a dynamic temperature field distribution rule of the silicon carbide workpiece in the continuous moving heating process; under different heating belt structures and different temperature setting modes, the temperature difference between the center and the edge and the temperature difference between the front end face and the rear end face of the moving silicon carbide workpiece are compared and analyzed, a heating performance optimization scheme of an original furnace profile is determined, and therefore the temperature uniformity of the silicon carbide workpiece in the furnace is improved. The original furnace profile is optimized on the premise that the structural size of the furnace body is not changed, and the temperature uniformity of the silicon carbide workpiece is remarkably improved.

The invention discloses a multi-scale vortex recognition method based on topology analysis, and aims to solve the problems that an existing vortex recognition method depends on a single scale and is insufficient in adaptability, limited in generalization ability, poor in visualization effect and the like. According to the method, firstly, a merge tree is constructed to obtain a multi-scale topological structure of a flow field, candidate vortex regions are screened based on extreme point-saddle point branches, and multi-scale vortex extraction is carried out in combination with a scalar criterion, so that dependence of a traditional method on a fixed threshold is avoided. In order to adapt to unstructured grid data, the method enhances the applicability and robustness of the method in a complex flow environment through adjacency relation reconstruction. In addition, a region diffusion algorithm is adopted to identify the vortex region, and surface extraction and smoothing processing are combined to optimize the visualization effect, so that the boundary of the vortex structure is more continuous and natural. Experimental results show that the method can accurately extract vortex structures of different scales, shows good generalization ability on flow field data of different Reynolds numbers, and has higher adaptability on complex flow fields and unstructured grids compared with a traditional method. The method can be widely applied to turbulence research, aerodynamic optimization, meteorological analysis and engineering flow problems, and an efficient, robust and accurate vortex recognition technology is provided for fluid mechanics research.

The invention discloses an unstructured mesh generation method applied to casing heat flow field numerical calculation, which relates to the technical field of engineering, and comprises the following steps: obtaining the structural characteristics of a casing, and constructing a casing finite element analysis model; preprocessing the finite element analysis model of the casing pipe, identifying geometric defects and carrying out geometric shape division; performing grid drawing based on the fluid area and the solid area of the casing finite element analysis model to obtain a first grid and a second grid; and based on the first grid and the second grid, casing pipe swept grid data are obtained, and a three-dimensional swept grid is formed. According to the unstructured grid drawing method for sleeve heat flow field numerical calculation based on the sweeping method, the grid subdivision efficiency is improved, the calculation complexity of sleeve heat flow field analysis is effectively reduced, and powerful technical support is provided for design, optimization and safe operation of power equipment.

The invention provides a grid generation method and device for a nuclear power fuel assembly and a computer readable medium, and the method comprises the steps: receiving a geometric file of the nuclear power fuel assembly, and generating a hybrid grid of the nuclear power fuel assembly according to a grid generation strategy and the geometric file, the grid generation strategy comprises the steps that a structured grid is used in a boundary layer of the nuclear power fuel assembly, and an unstructured grid is used in an area outside the boundary layer; the generated hybrid grid is subjected to self-adaptive refinement processing, the self-adaptive refinement processing comprises the steps of recognizing a high-gradient area and increasing the grid density in the high-gradient area, and the high-gradient area refers to the area where the change rate of fluid parameters in space is larger than a preset threshold value when fluid flows along the wall face of the nuclear power fuel assembly. According to the method, the grid density is only increased in a high-gradient region, and relatively thick grids are kept in other regions, so that the resolution of a key region is ensured, and the waste of computing resources is effectively reduced.

The invention discloses a magnetotelluric sounding measurement method for multidirectional aggregation observation, and the method comprises the steps: building a novel device layout: designing a reasonable measurement point layout scheme according to geological features and electromagnetic response distribution; processing data and establishing an unstructured grid: acquiring an electromagnetic field component, completing data acquisition, and inputting the processed electromagnetic field data into an unstructured grid model; impedance tensor calculation: on the basis of the unstructured grid, performing impedance calculation by utilizing multi-component electromagnetic field data and combining an interpolation idea, and constructing an extended impedance tensor of an electromagnetic field component to obtain an impedance value component; and three-dimensional inversion and visualization: performing three-dimensional inversion through a vector finite element method to obtain an inversion result. According to the method, richer observation data are provided for deep geologic structure inversion.

The invention relates to a novel stable node integral calculation method based on hourglass control, and the method comprises the steps: dividing a problem domain into a triangular or tetrahedral mesh, and dividing a mesh integral domain into a plurality of non-overlapping node integral domains based on nodes in the triangular or tetrahedral mesh; performing node domain smoothing on the node integral domain based on the triangular / tetrahedron strain component of the node to obtain a smooth domain; hourglass control between a node and an adjacent node is constructed in the smooth domain, and hourglass force between the node and the adjacent node is calculated; and superposing the hourglass force borne by the node with the original internal force calculated by node integration to obtain the internal force of a stable node integration method, and solving a discrete motion equation in explicit dynamics based on the internal force of the stable node integration method to obtain a dynamic response. According to the method, the calculation precision based on the unstructured grid is greatly improved, and the advantages that the number of integration points of node integration is small and the grid distortion resistance is high are effectively kept.

The invention discloses a coastal disaster forecasting method and system. The method comprises the following steps: acquiring data of a target near-shore area, and carrying out data preprocessing and unstructured grid division; generating an open sea drive-near-shore response sample library by using the coupling numerical model; extracting the dominant lag step number of each open-sea factor for each node near the shore through mutual information analysis, performing time rearrangement on key factors, splicing with static geographic features, and constructing a structured input tensor; training a graph attention space-time fusion model taking the unstructured grid as a graph structure; in real-time forecasting, an implicit state assimilation mechanism based on observation errors is introduced, and the model state is dynamically corrected to achieve self-adaptive rolling forecasting. According to the invention, high-precision and high-efficiency near-shore disaster real-time early warning is realized.

The invention provides a parameterization construction method of a recycled concrete three-dimensional mesoscopic numerical model. The parameterization construction method comprises the following steps: 1) calculating the particle number and the particle size of recycled concrete aggregate and the thickness of an old mortar layer; 2) constructing a geometric model of the recycled concrete three-dimensional random spherical aggregate test piece; (3) constructing a geometric model of the recycled concrete three-dimensional random convex polyhedral aggregate test piece; (4) carrying out unstructured grid division; and 5) construction of a recycled concrete projection network model. According to the method, the problem that the geometric space structure of the recycled concrete is complex is solved, and the established geometric model can truly reflect the actual grading, content and shape of the aggregate in the recycled concrete; according to the unstructured mesh generation method aligned with the interface, the problems that mesh generation is difficult, and modeling is difficult due to the fact that an interface transition area is too thin are solved, and the generated mesh model has extremely high unit quality.

The invention discloses a gas turbine blade root wheel groove stress field reconstruction method and system, and relates to the technical field of stress field reconstruction, and the method comprises the following steps: collecting a plurality of sets of design variables, a plurality of sets of measuring point data and graph data corresponding to a plurality of finite element models under different molded line design schemes; performing finite element calculation on the finite element model to obtain a plurality of first stress fields and a plurality of second stress fields; the multiple sets of design variables and the multiple sets of measuring point data serve as input, the multiple first stress fields serve as output, and a convolutional neural network is trained; and training the graph convolutional neural network by taking the plurality of groups of design variables, the plurality of groups of measurement point data and the plurality of groups of unstructured grid data as inputs and the plurality of second stress fields as outputs. According to the method, finite element calculation consuming long time is not needed in the application stage, real-time simulation can be achieved, for different network structures, the most skilled data forms can be processed, and the stress field reconstruction efficiency under the blade root wheel groove profile design is improved.

The invention discloses an aircraft grid generation method and device based on conformal mapping and topological segmentation, equipment and a storage medium, and relates to the technical field of computational fluid mechanics, and the method comprises the steps: generating a surface unstructured grid according to an aircraft model, carrying out the conformal mapping to a parameter domain plane grid through Ricci flow, and calculating the compression ratio of the surface unstructured grid; if the compression ratio is smaller than a first threshold value, extracting a contour line corresponding to a second threshold value; positioning a triangular grid corresponding to the isoline discrete points in the parameter domain plane grid by using a spatial partitioning algorithm, inserting the discrete points into the grid according to the relationship between the vertex distance and the edge distance, forming a closed internal boundary through a one-dimensional boundary recovery operation, and determining a corresponding region; the area is inversely mapped back to the three-dimensional surface, the surface grid is updated, the mapping and judging steps are repeated until the compression ratio meets the requirement, and the parameter domain plane grid is set as a target aircraft grid, so that the efficiency of generating the aircraft grid is improved.

The invention discloses an aero-engine complete machine full-three-dimensional performance simulation method and device and a storage medium. The method comprises the steps that firstly, a complete machine full-three-dimensional geometric model of all main watershed components from an inlet to an outlet of the aero-engine is constructed under a unified coordinate system and processed, and a single-channel model is formed through periodic simplification; grid division is carried out by adopting a hybrid grid strategy, structured grids are divided for the rotating part, and unstructured grids are divided for the complex structure; secondly, in a single universal CFD solver, specific turbulence and combustion models are configured, a Stage interface is adopted to carry out data transmission between components, high-precision numerical simulation is executed, and solving is carried out based on a strict convergence criterion; and finally, extracting performance parameters from a simulation result for evaluation. According to the method, the integral high-precision simulation of the whole machine in a single solver environment is realized, the problem of low parameter transmission precision among parts in a traditional method is effectively solved, and the engineering-level simulation precision can be achieved.

The invention relates to the technical field of marine environment numerical simulation, in particular to a near-shore typhoon wave refined simulation method based on terrain adaptive grids and parameter dynamic correction, and the method comprises the steps: obtaining basic data and simulation parameters of a target sea area, constructing a computational domain and an initial grid, and constructing a target sea area; the initial grid is an unstructured grid fused with multiple physical constraint factors; performing optimization processing on the initial grid based on a numerical stability criterion to obtain a final calculation grid meeting a calculation convergence requirement; establishing a bidirectional coupling model, and applying the final computational grid to the bidirectional coupling model; carrying out localization iteration calibration on key physical parameters of the bidirectional coupling model to obtain an optimal parameter set; and carrying out typhoon wave process simulation calculation and constructing a typhoon wave feature database by utilizing the final calculation grid and the optimal parameter set, and realizing high-precision simulation and prediction of the typhoon wave process by constructing an adaptive optimization grid, a bidirectional coupling model and a parameter localization calibration mechanism.

This invention discloses a method, system, and medium for data mapping between a solid geometric model and an unstructured mesh, relating to the field of nuclear reactor core coupling programming technology. By converting the CSG geometric entity into a coupling intermediate mesh (unstructured mesh), and using the coupling intermediate mesh as an intermediary between the CSG geometric entity and other professional unstructured meshes, high-precision coupled data transmission and coupled calculation between the CSG geometric entity and other professional unstructured meshes can be achieved through mesh mapping and other methods, based on the geometric correspondence between the CSG geometric entity and the coupling intermediate mesh.

The application provides a structure grid-based unstructured grid division method and an electronic device. The method determines a calculation domain range of a combustion detonation simulation task, performs surface grid division on a geometric model to obtain a geometric surface representation grid, generates a structure grid according to the geometric surface representation grid and the calculation domain range, performs simulation calculation on the structure grid, determines a grid encryption parameter according to a simulation calculation result of the structure grid, generates an unstructured grid through the grid encryption parameter, the calculation domain range and the geometric surface representation grid, realizes the structure grid-based unstructured grid division, can replace a trial-and-error iterative optimization or a complex dynamic self-adaptive process of the unstructured grid, improves the generation efficiency of the unstructured grid and guarantees the generation quality of the unstructured grid, balances the calculation efficiency while meeting the high-precision requirement, and is suitable for a large-scale parallel environment.

The invention discloses a CFD-based automobile air wind resistance friction simulation system, and relates to the technical field of CFD simulation, the CFD-based automobile air wind resistance friction simulation system comprises a monitoring center, and the monitoring center is in communication connection with a data acquisition module, a data processing module, a CFD simulation module, a real-time monitoring visualization module and a deep learning optimization module; the method comprises the following steps: acquiring vehicle shape geometric data, vehicle running state data and environmental condition data of a to-be-tested vehicle, and constructing a vehicle geometric model and initial simulation conditions; performing unstructured grid processing on the vehicle geometric model to obtain unstructured grid data; obtaining a wind resistance coefficient and a friction force index according to the unstructured grid data and an initial simulation condition, and further judging whether the wind resistance coefficient and the friction force index of the to-be-tested automobile meet requirements or not; according to the wind resistance coefficient and the friction force index, constructing a CFD simulation result training deep learning model, and further obtaining a wind resistance friction optimization coefficient; and the simulation efficiency of the wind resistance friction is improved.

The invention discloses a complex mountain wind power plant wake flow simulation method and system based on computational fluid mechanics, and relates to the technical field of computational fluid mechanics and engineering.The complex mountain wind power plant wake flow simulation method comprises the steps that a wind power plant reference frame model is built based on digital elevation model data of the complex mountain terrain where a target wind power plant is located and basic parameters of wind power facilities; then constructing a three-dimensional unstructured grid computational domain of the target wind power plant; taking a three-dimensional unstructured grid computational domain as a basic environment, generating an inlet turbulence condition, and setting a wind turbine actuating line model by utilizing computational fluid mechanics simulation software based on wind power facility basic parameters; solving the wind turbine actuating line model to obtain a simulation data set; and verifying the simulation data set based on the actual observation information, and carrying out quantitative analysis simulation on the wake flow development condition of the target wind power plant based on the verified simulation data set. According to the invention, the technical problems of high data acquisition requirements and low flexibility and stability of traditional numerical simulation are solved.

The invention discloses a cycloid type static pressure oil cavity structure design method which comprises the steps that a first fluid domain model corresponding to a plane type oil cavity and a second fluid domain model corresponding to a cycloid type oil cavity are extracted, grid discretization processing is conducted on the first fluid domain model, and unstructured grid division is conducted on the second fluid domain model; respectively simulating fluid motion states of the first fluid domain model and the second fluid domain model, and analyzing oil film bearing capacity and oil film pressure field distribution corresponding to the planar oil cavity and the cycloid oil cavity according to simulation results; determining the oil film bearing force difference and the oil film pressure field distribution difference between the plane type oil cavity and the cycloid type oil cavity; and optimizing the preset structure of the cycloid oil cavity according to the oil film bearing force difference and the oil film pressure field distribution difference. According to the design method of the cycloid type static pressure oil cavity structure, the problems that an oil film of a static pressure oil cavity is prone to fracture, pressure fluctuates, and machining stability and precision are low are solved.

The application discloses a kind of urban flood simulation methods based on unstructured grid coupling SWMM and FVCOM model, adopt inspection well simplified method to characterize the exchange process of surface water flow and underground pipe network water flow, unstructured grid FVCOM is main program, SWMM is called by its subprogram.First, by using SWMM model and unstructured grid FVCOM model given basic data, the model of underground pipe network water flow and surface water flow is established respectively;Establish the topological relationship of SWMM pipe network node and FVCOM grid space;At the time step that needs to be coupled, obtain the head of pipe network water flow and surface water head of inspection well node pipe network, and calculate the exchange flow, according to the updated head and flow, simulate surface water and pipe network water flow respectively;Time progression, finally complete the whole urban flood simulation process.The application can accurately simulate the coupling effect of urban pipe network water flow and surface water flow, realize the simulation of urban flood with tortuous coastline or complex building profile.

The present invention provides a parametric method for constructing a two-dimensional mesoscopic numerical model of recycled concrete, comprising the following steps: 1) calculating the particle count and particle size of recycled concrete aggregate and the thickness of the old mortar layer; 2) constructing a geometric model of a two-dimensional random circular aggregate specimen of recycled concrete; 3) constructing a geometric model of a two-dimensional random convex polygonal aggregate specimen of recycled concrete; 4) performing unstructured meshing; and 5) constructing a projected mesh model of recycled concrete. This method effectively addresses the complex geometric spatial structure of recycled concrete. The resulting geometric model can truly reflect the actual gradation, content, and shape of the aggregate in the recycled concrete. It also effectively addresses the difficulty of meshing, resulting in a mesh model with extremely high unit quality.

The invention discloses an unstructured grid adaptive sub-mesoscale hybrid parameterization method and system, and the method comprises the steps: calculating a sea surface horizontal density gradient defined on an edge in a sea model of an unstructured grid, correcting the sea surface horizontal density gradient according to an effective resolution ratio and a frontal surface scale, and carrying out the calculation of the submesoscale hybrid parameterization of the submesoscale hybrid parameterization of the submesoscale hybrid parameterization of the submesoscale hybrid parameterization of the unstructured grid. Calculating a sea surface horizontal buoyancy gradient; calculating the sea surface level Ekman buoyancy flux according to the sea surface level buoyancy gradient and the sea surface wind stress; and if the sea surface level Ekman buoyancy flux is a negative number, further calculating a mixing coefficient in the boundary layer. According to the method, the influence of the symmetrical unstable process on boundary layer mixing can be effectively introduced into the global unstructured grid ocean mode, and the simulation effect of the depth of the mixing layer under coarse resolution is remarkably improved.

The invention relates to the technical field of biomolecule electrostatic characteristic prediction, and discloses a partial differential equation solving method based on a finite volume element method. According to the method, an unstructured grid matched with a biomolecular interface is constructed, a solution of a Poisson-Boltzmann equation (PBE) is separated into an analytical solution G, a linear interface problem solution psi and a nonlinear interface problem solution by adopting a decomposition technology, and a non-homogeneous flux condition at the interface is processed by utilizing a controlled volume decomposition technology; linear and nonlinear problems are solved through a finite volume element method, and final electrostatic potential distribution and solvation free energy are obtained through assembling. The method solves the problems of insufficient precision and low efficiency when a traditional numerical method is used for processing a complex interface non-homogeneous flux condition, has second-order L2 convergence and first-order H1 convergence, and can be efficiently applied to the biophysical research fields such as protein solvation free energy prediction and the like.

The embodiment of the invention provides a hydrogen fuel flow simulation CFD post-processing method, device, equipment and medium, and relates to the technical field of numerical simulation, and the method is realized through the following steps: executing hydrogen fuel flow CFD simulation based on an aero-engine, and generating unstructured grid data; converting the unstructured grid data into a plurality of simulation result files, sorting grid geometric information and numerical value distribution data in the simulation result files, generating a time snapshot matrix, calculating an adjoint matrix S through the time snapshot matrix, and then obtaining a mode corresponding to each feature vector of the hydrogen fuel flow field; and through a mode corresponding to each feature vector of the hydrogen fuel flow field, calculating to obtain a magnification gi and a mode coefficient amplitude ba, and carrying out dynamic reconstruction and prediction on the hydrogen fuel flow field through a mode superposition method. According to the scheme, the flow field state at any moment can be reconstructed and predicted through a modal superposition method, and the post-processing efficiency is improved.