198results about How to "High simulation" patented technology

The invention discloses a probe card simulation method and system based on counter-acting force analysis, which are used in the field of wafer simulation test, and the probe card simulation method comprises the following steps: integrating multiple physical field working conditions to construct a probe card geometric model, and importing the probe card geometric model into finite element simulation analysis software; carrying out grid division on the geometric model of the probe card by utilizing the hybrid voxel tree model, adjusting the grid density according to the distribution requirement of the multi-physical field, and setting a simulation boundary condition by utilizing a preset boundary condition template according to the actual simulation requirement; starting a simulation calculation process based on the adjusted grid density and simulated boundary conditions; and providing a visual interface of physical field distribution and a structural stress state in the simulation calculation process, and performing iterative optimization on the performance of the probe card in the simulation process according to a reactive force distribution result. According to the method, the grid density is automatically adjusted to adapt to changes of a physical field, and it is ensured that higher simulation precision is obtained in a key area.

The invention provides a method and a device for generating a multiphase flow-containing solid rock three-dimensional model based on a diffusion model, and a medium, and can solve the problems that a multiphase pore and a rock skeleton structure cannot be generated at the same time and only a single-phase structure or a pure skeleton image can be reconstructed in the prior art. According to the method, the watershed algorithm is introduced to accurately segment the multiphase features, Gaussian noise is used as diffusion model input, and the model gradually learns' de-noising 'from the noise to obtain a real and complex multiphase rock image in training, so that a high-fidelity multiphase distributed rock internal image can be obtained only by inputting random noise in an inference stage.

The invention discloses a three-dimensional embossment type ceramic tile and a preparation method thereof, the three-dimensional embossment type ceramic tile sequentially comprises a base layer, a ground coat layer, a three-dimensional layer, a glaze layer and a polishing layer from bottom to top, the base layer is located at the bottom layer of the ceramic tile and has a thickness of 8-10mm, the ground coat layer is formed by applying ground coat on the base layer and has a thickness of 0.5-1mm, and the three-dimensional layer is located at the bottom layer of the ceramic tile and has a thickness of 0.5-1mm. The three-dimensional layer is a pattern layer formed after ink-jet printing is carried out on the ground glaze layer, the thickness of the pattern layer is 1.5-2.5 mm, the glaze surface layer is formed after overglaze is applied to the three-dimensional layer, the thickness of the glaze surface layer is 0.4-0.8 mm, and the polishing layer is formed after polishing treatment is carried out on the glaze surface layer. According to the three-dimensional embossment type ceramic tile, the ultra-white ground glaze, the multi-layer ink-jet printing three-dimensional layer and the cover glaze layer are combined by adopting multi-layer laminated glaze, so that the three-dimensional embossment type ceramic tile can clearly embody the texture and texture of raw stone, has better visual impact force and artistic expressive force, and improves the three-dimensional effect and artistic ornamental value of the ceramic tile; the requirements of the market on high-end decorative ceramic tiles are met.

The invention discloses a method based on combination of a Voigt function and a deep neural network, which is used for reconstructing a Raman spectrum of a deep substance, solves the technical problems of complicated implementation process and higher implementation threshold in the prior art, and specifically comprises the following steps: step 1, generating a training data set; 2, training a deep neural network to obtain a Raman spectrum reconstruction model; step 3, collecting a spatial offset Raman spectrum of the double-layer sample; and step 4, carrying out baseline correction processing on the collected Raman spectrums to obtain corrected Raman spectrums, splicing the corrected Raman spectrums into input data, and inputting the input data into the Raman spectrum reconstruction model to obtain reconstructed pure Raman spectrums. The method is high in adaptability and simple in operation step, and can be used for automatically processing the Raman spectrum and eliminating the spectrum peak interference of the packaging material, so that the pure Raman spectrum of the deep sample can be quickly, simply and conveniently obtained.

The invention belongs to the technical field of seepage tests, and relates to a seepage test device and method for a concrete and bedrock cementation surface under a seepage corrosion condition, the device comprises axial loading devices, a water sealing device is arranged between the axial loading devices, a sample is arranged in the water sealing device, a tensile stress loading device is arranged on the side surface of the sample, and the tensile stress loading device is arranged on the side surface of the sample. The upper end and the lower end of the water sealing device are connected with a water seepage measuring system through water pipes, the water seepage measuring system is connected with a seepage water pressure loading device, and the seepage water pressure loading device, the tensile stress loading device and the axial loading device are electrically connected with the upper computer; the permeation water pressure loading device accurately controls water pressure and can provide high and low water pressure loading environments, and the water sealing device solves the problems that a conventional cement-based material corrosion device cannot seal high water pressure and cannot develop permeation corrosion characteristics of a concrete-bedrock interface under the action of different hydraulic gradients.

The invention relates to the technical field of medical instruments, in particular to a multi-layer ultra-light hernia patch which comprises a large-mesh electrostatic spinning tissue ingrowth layer, a mesh woven base material layer and an anti-adhesion layer which are sequentially connected in a stacked mode. Wherein after the large-mesh electrostatic spinning tissue ingrowth layer is implanted into a human body, a cell microenvironment can be better simulated, cell adhesion, differentiation and migration are facilitated, uniform distribution of newborn cells in the fiber direction of an electrospinning membrane is promoted, granulation tissues uniformly ingrow in a defect part, and the electrospinning fiber membrane and the tissues are cooperatively repaired; secondly, the middle layer can provide supporting performance and mechanical performance of the patch in the early stage for the patch, the defects of the electrospinning fiber membrane in the supporting performance and the mechanical performance are overcome, the long-term implantation amount of the patch after later degradation is greatly reduced, and the foreign body sensation is obviously reduced; the anti-adhesion layer also adopts the electrostatic spinning technology, and the anti-adhesion effect is achieved by selecting materials and regulating and controlling the porosity, the aperture and the fiber diameter of the electrospun fiber membrane.

The invention discloses a construction method and application of thymus organoid, and the method comprises the following steps: differentiating pluripotent stem cells into thymus organoid under a suspension culture condition through combined and matched use of small molecule compounds and small molecule proteins, according to the method disclosed by the invention, the thymus organoid of which the structure and the function are closer to those of natural thymus can be efficiently obtained.

The invention discloses a fast universal passivity macro-modeling method based on Hamiltonian matrix pure virtual feature values. The method comprises the steps that n-port passive network parameters obtained through simulation or measurement serve as an original data set; rational fitting is carried out on the passive data set, and a macro model is established; calculating a pure virtual characteristic value of a Hamiltonian matrix M of the macro model, and judging the passivity of the macro model; carrying out passivity repair on the initial model according to a passivity judgment result; and inputting the passive macro model into a circuit simulator for simulation. According to the method, only the pure virtual feature value of the Hamiltonian matrix is calculated to achieve rapid passivity detection of the macro model, the method is suitable for macro modeling of symmetric and asymmetric models, the method has the characteristic of universality, and when the method is applied to integrated circuit design, the convergence of circuit simulation can be improved, the macro modeling time is shortened, the circuit simulation is accelerated, and the circuit simulation efficiency is improved. And the design and market period of electronic products is shortened.

The invention discloses a hovering four-wing flapping-wing aircraft, and relates to the technical field of aircrafts, the hovering four-wing flapping-wing aircraft comprises a fixed rack, a power source and a gear set located in the middle of the fixed rack, two pairs of wings arranged at the two ends of the fixed rack, a connecting gear set and a connecting rod mechanism of the two pairs of wings, and an attitude control structure movably connected with the two pairs of wings respectively; one end, far away from the transmission shaft, of the crank is hinged to a connecting rod, one end, far away from the crank, of the connecting rod is hinged to a rocker, the middle of the rocker is bent and rotates on the outer side of the fixed rack, one end of the rocker is connected with the connecting rod, the other end of the rocker is movably connected with a flapping wing rod, and the tail end of the flapping wing rod is fixedly connected with a wing; the connecting rod mechanism is a five-connecting-rod mechanism composed of a crank, a connecting rod, a rocker, a fixed rack and a flapping wing rod. The performance is improved by simplifying a transmission system and an attitude control structure of an existing ornithopter and optimizing a connecting rod mechanism, better attitude control and bionic flight effects are achieved, and the performance of the ornithopter is greatly improved.

The invention discloses a geographic scene intelligent computing system and method, and belongs to the field of geographic space information. The invention discloses a geographic scene intelligent calculation system. A geographic scene comprises a scene model, geographic data and model calculation. The scene model comprises a geographic model, geographic knowledge and an object space, and the geographic model performs modeling for a geographic process on a geographic scene to form a geographic process analysis and dynamic simulation calculation process; the geographic data comprises a model data set and a basic data set, the model data set is used for calculating a data set list required by a geographic model on the geographic scene, and the basic data set is used for visualized display of the geographic scene; the model calculation comprises data input, data output and a calculation process, and the calculation process of the geographic model comprises a calculation process, a calculation node and a calculation state; according to the method, the problem of cross-layer and multi-element combined geographic space comprehensive analysis and calculation in the prior art is solved, and multi-geographic-element combined complex geographic space intelligent calculation is realized.

The invention discloses a camouflage material, a hyperspectral camouflage coating and a preparation method and application thereof, and belongs to the technical field of camouflage materials, and the molecular formula of the camouflage material is Co2Cr3TiLiNiO10. The camouflage material has the performance of simulating spectral characteristics of natural yellow and green plants and has a good bionic function, crystals of the camouflage material are of a spinel structure, and the camouflage material has the characteristics that the lightness value L is relatively low, the hue is relatively dark, a green camouflage coating is easier to prepare, the camouflage material has the characteristics of good aging resistance, good thermal stability and the like, and the camouflage material can be used as a green camouflage coating. The reflectivity of the infrared plateau area is higher, the camouflage performance is easier to realize, and the spectral characteristics of green vegetation can be better simulated.

The invention is suitable for the field of social network analysis, and provides a social public opinion hypernetwork key public opinion element identification method based on a structural hole, and the method comprises the steps: collecting public opinion event related data; constructing a social public opinion hypernetwork model of five layers of subnets; constructing a conversion hypergraph, and constructing a conversion line graph according to the conversion hypergraph; and calculating the sum of scores of all hyperedges formed by node participation, and calculating the final score of the node. According to the method, the development situation of public opinion propagation is researched by utilizing a multilayer super network model, a structural hole theory is introduced into a method for identifying public opinion elements, and a conventional hyperedge sorting method is improved to a certain extent; compared with the traditional method, the method provided by the invention can effectively and accurately identify various key public opinion elements in the social public opinion super network, and can analyze by combining with the node importance index of the super network, so that the key public opinion information can be interpreted from multiple dimensions and multiple perspectives.

The invention belongs to the technical field of water drive reservoir development, and particularly relates to a parallel sand-filled pipe model with an inter-pipe fluid streaming function as well as a preparation method and application of the parallel sand-filled pipe model. The invention provides a parallel sand-filled pipe model with an inter-pipe fluid streaming function. The model comprises a plurality of sand-filled pipes with different permeability and interlayer connecting pipes 4, each sand-filled pipe comprises a sand-filled pipe body 1 with inter-pipe interconnection holes 3 and a filler 2 filled in the sand-filled pipe body 1; and the plurality of sand-filled pipes with different permeability are connected in parallel by using interlayer connecting pipes 4 through the pipe interconnection holes 3. The parallel sand-filled pipe model has an inter-pipe fluid streaming function and can simulate the seepage condition of a heterogeneous reservoir in a layer, and the simulation result is close to the actual condition.

The invention discloses a deepwater solid material lifting experiment system. A material distributing mechanism comprises a clear water bin, a circulating bin, a material distributing bin and a material guiding trolley; the calibration mechanism comprises a calibration box which is connected with an outlet of the distribution bin; the feeding mechanism comprises a storage bin and a spiral conveyor connected with an outlet of the storage bin, and an inlet of the storage bin is connected with the circulating bin and the calibration box; the lifting mechanism comprises a plurality of vertical lifting pipes, each vertical lifting pipe comprises a plurality of combined pipe sections, and each combined pipe section comprises a supporting section, a measuring section and an observing section; the power mechanism comprises a water pump and an air compressor, an inlet of the water pump is connected with the clean water bin, the storage bin and a water source, an outlet of the spiral conveyor and an outlet of the water pump are connected with an inlet of the vertical lifting pipe, an air-entrapping connector is arranged on the side face of the vertical lifting pipe, and an outlet of the air compressor is connected with the air-entrapping connector; an outlet of the vertical lifting pipe is located in the material guiding trolley. According to the invention, hydraulic, pneumatic and hybrid material lifting experiments can be carried out, and the gas-liquid-solid three-phase flow state can be better monitored through the combined pipe section.

The invention relates to a scheduling processing method and device, computer equipment, a computer readable storage medium and a computer program product, and relates to the technical field of cloud computing. According to the invention, the problem of pressure measurement flow concentration can be avoided to a certain extent, so that the network request flow condition in reality can be better simulated. The method comprises the steps that a node recommendation value of each node in a cluster is obtained for each scheduling unit copy of a pressure measurement task, the node recommendation value is determined based on the deployment state of a node pool where the node is located for the pressure measurement task, and a target node of the scheduling unit copy is determined according to the node recommendation value of each node.

The invention discloses a total phosphorus concentration prediction method in a sewage treatment process, electronic equipment and a medium, and belongs to the field of sewage treatment. The method comprises the following steps: collecting various index data in various sewage treatment processes to obtain a data set; after the obtained data set is preprocessed, characteristic variables related to total phosphorus are screened; a corresponding phosphate concentration prediction model based on the physical information technology is established for the ferric salt delivery process and is used for predicting the phosphate concentration; combining the phosphate radical concentration obtained by prediction with the screened variables to jointly construct a total phosphorus prediction model; the total phosphorus prediction model is trained, and the trained total phosphorus prediction model is used for predicting the total phosphorus concentration in the sewage treatment process. According to the method, part of known scientific rules and empirical formulas in the sewage treatment field are creatively integrated into a data-driven prediction model, the theoretical basis of a traditional model and the self-learning ability of a machine learning model are combined, and the prediction accuracy and reliability of the model are improved.

The invention relates to a sea surface wind speed inversion method and system based on a GNSS-R forward model, and belongs to the technical field of ocean remote sensing and environment monitoring. The method comprises the following steps: firstly, inputting four types of CYGNSS data to obtain a modeling DDM and a Jacobian matrix, secondly, carrying out feature extraction and data quality screening to obtain a plurality of features related to sea surface wind speed, then carrying out space-time matching on the features and the wind speed, inputting the features into an inversion model, and finally obtaining that the distribution of the DDM generated by a DDM forward model and the distribution of the observation DDM are relatively concentrated, and the overall simulation effect is good. According to the method, the problems that the neural network lacks physical interpretability and the inversion process is complex are solved, and a new thought is provided for later inversion of soil humidity and other indexes.

The invention discloses a picture object detection method for an X-ray security inspection machine. The method comprises the following steps: acquiring a picture data set of the X-ray security inspection machine, generating a high-quality X-ray image sample through a generative adversarial network, expanding the data set, and optimizing a training process of an adversarial network generator through reinforcement learning. A multi-task learning model is constructed, object detection, object segmentation and object category identification are jointly trained, and under the condition of no labeled data, feature representation is generated through self-supervised learning to serve as an initial weight to be applied to training of an object detection model; the invention discloses a generative adversarial network. A traditional data enhancement method may have limitation in diversity and authenticity of generated samples. The complex condition in the actual environment can be better simulated through the sample generated by the GAN, so that the robustness of the model in the actual application is improved. And reinforcement learning optimization: utilizing reinforcement learning to optimize the training process of the GAN, and dynamically adjusting the generation strategy, so that the generated image samples are more diversified and real.

The invention discloses a method for improving cell growth and expression by changing an iron form in a culture medium, and relates to the technical field of cell culture, the culture medium comprises a basic culture medium, a supplemented culture medium and an iron source; wherein the iron source is FeCl3, the basic culture medium and the supplementary culture medium are commercially available culture media, the iron source is not added or removed, and the culture media further contain a three-dimensional culture material; the redox balance of the culture medium can be maintained, oxidative stress damage of cells is prevented, hydrolysis or precipitation is not prone to occurring, the preparation process of the culture medium is simplified, and the requirements of cell growth and expression are met.

The invention relates to the technical field of organoid culture, in particular to a pulmonary fibrosis organoid model and a construction method and application thereof. The pulmonary fibrosis organ-like model is obtained by co-culturing lentivirus stably transfected alveolar epithelial cells, human embryonic lung fibroblasts MRC-5 and macrophages. According to the invention, macrophages are innovatively introduced, a ternary co-culture system of lentivirus stably transfected alveolar epithelial cells, human embryonic lung fibroblasts MRC-5 and macrophages is constructed, and a key'epithelial injury-immune infiltration-fibroblast activation 'pathological triangle in a pulmonary fibrosis pathogenesis process can be more comprehensively simulated; the finally obtained pulmonary fibrosis organ-like model provides an efficient tool for pulmonary fibrosis mechanism research and research, development and screening of anti-pulmonary fibrosis drugs.

This invention provides an adaptive margin optimization method, system, and electronic device based on speaker verification. The method includes: inputting speech training data including various speech durations into a speaker verification model; determining the loss function of the speaker verification model; adaptively optimizing the margin parameters of the loss function based on the speech durations in the speech training data and a preset target margin for each speech duration; and training the speaker verification model using the margin parameters of the adaptively optimized loss function to determine the acceptable training difficulty of the speaker verification model. This invention utilizes training speech of varying lengths to better simulate real-life scenarios. Through adaptive optimization and fine-tuning of the margin, adjusting the margin according to the duration and similarity of each speech, this method achieves good speaker verification performance for speech of different durations in real-world scenarios.

The application provides a VR detection lens, which comprises, in sequence from an object side to an imaging surface along an optical axis, an aperture, a first lens group with positive refractive power, a second lens group with negative refractive power, and a third lens group with positive refractive power; the lens satisfies the following conditions: 0.8<=|f20 / f|<=1.5, 33<=|f30 / f|<=40, and 4.0<=|f40 / f|<=5.0, wherein f is the focal length of the detection lens, f20 is the focal length of the first lens group, f30 is the focal length of the second lens group, and f40 is the focal length of the third lens group. The application has higher resolution and a larger field of view, can well simulate human eye observation, and realizes batch rapid detection of VR products.

This invention relates to the fields of biomedical technology and biological model construction technology, and discloses an ANCA-associated vasculitis glomerular endothelial cell injury model, its construction method and application, and the application of BACH1 inhibitors. The method includes: (1) activating neutrophils sequentially with tumor necrosis factor α and immunoglobulin G to obtain a culture medium containing activated neutrophils; wherein the immunoglobulin G is immunoglobulin G isolated and purified from the plasma of patients with ANCA-associated vasculitis; (2) co-culturing the culture medium containing activated neutrophils with human glomerular endothelial cells to obtain the cell injury model. The cell injury model provided by this invention can verify the role of neutrophil adhesion characteristics in endothelial cell injury, and at the same time, the cell injury model can better simulate the immune microenvironment of ANCA-associated vasculitis glomerular endothelial cells in vivo.

The application discloses a kind of hardness correlation liver cancer lung premetastatic niche mouse model construction method, it includes the following steps: (1) in vitro enrichment respectively on the low hardness substrate of hardness 6kPa and the high hardness substrate of hardness 16kPa, the condition culture supernatant of mouse liver cancer cell Hepa1-6 that is cultured and grows, respectively named mL-CM and mH-CM;(2) mL-CM and mH-CM are respectively injected into two groups of mice using tail vein injection mode, every 1 day injection 1 time, until the 26th day, during continuous monitoring the recruitment of BMDCs in fresh lung tissue, while with the induction endpoint, i.e. 26th day premetastatic niche mark feature detection includes lung tissue matrix remodeling, immunosuppression, angiogenesis, vascular permeability and inflammation etc..The method has the advantages of short modeling cycle, easy operation, controllable induction condition, can comprehensively and accurately evaluate the common characteristics of liver cancer lung premetastatic niche formation etc..

The invention provides a simulated guide tube rod and a simulated nuclear reactor, and the simulated guide tube rod comprises a core rod which comprises a boss, a middle rod, a connector and a core rod groove; part of the core rod is located in the outer pipe, the outer pipe comprises a pipe part and an outer pipe groove for containing the boss, and first gaps are formed between the inner wall of the pipe part of the outer pipe and the outer wall of the core rod and between the inner wall of the outer pipe groove and the outer wall of the core rod; the end plug comprises a first part and a second part, the first part is matched with the core rod groove, the second part is located below the outer pipe, and the diameter of the second part is larger than or equal to that of the outer pipe; the sealing filler, the inner wall of part of the outer pipe groove, the boss and the second part jointly form a first space, and the sealing filler is located in the first space and suitable for sealing the first space. The simulated guide tube rod and the simulated nuclear reactor provided by the invention can help to solve the problem of insulation between the guide tube rod and a peripheral electrified part, and the performance of the guide tube rod is further improved.

The invention discloses a fruit fly larva tumbling motion modeling and optimizing method based on a material point method. Aiming at a collaborative design challenge of a larva rolling motion model and the problems that a traditional method depends on manual data measurement and is time-consuming in training, a neural mechanical model is designed in combination with biological characteristics of drosophila larvae, and an artificial neural network based on deep learning is constructed as a model controller for training; meanwhile, a hyper-parameter optimization method based on a TPE algorithm is designed. According to the method, the method is embedded into a microprogrammable framework, and high integration and efficiency optimization of the method are achieved. The method can effectively simulate and optimize the rollover motion behavior of the drosophila larva under limited computing resources. The soft body model with the biological characteristics of the drosophila larvae has the advantages of flexible movement and adaptability to complex environments. According to the method, the optimal control strategy in the rollover process of the drosophila larvae is explored, interaction between the drosophila larvae and a complex environment is simulated, and important reference is provided for design of a soft robot similar to the drosophila larvae in reality.

The utility model discloses an electrolytic bath bipolar plate micropore flow channel performance detection device. A bipolar plate assembly is arranged on a bottom plate; the cover plate is arranged at the top of the bipolar plate assembly and is made of a transparent material; the isolation plate is arranged between the two adjacent bipolar plates, airflow simulation units are arranged on the sides, facing the bipolar plates located below the isolation plate and the cover plate, of the isolation plate and the cover plate, and airflow can be provided for the anode sides of the corresponding bipolar plates through the airflow simulation units; the jet flow form detection unit and the jet flow height detection unit are respectively used for detecting the jet flow form and the jet flow height of the jet flow formed on the bipolar plate below the cover plate; and the jet flow pressure detection unit is used for detecting the impact acting force of the jet flow formed on the bipolar plate below the jet flow pressure detection unit. By simulating the real working condition of the bipolar plate, the performance of the microporous flow channel under different design parameters can be fed back more truly, and a reliable guarantee is provided for optimization of the design parameters of the microporous flow channel.

This invention relates to the field of drilling fluid testing equipment technology, specifically to an evaluation device and method for simulating the dynamic pollution of drilling fluids by CO2. The evaluation device includes a sealed pressure vessel, an injection assembly, a circulation system, a control module, a CO2 injection volume detection module, a first monitoring module, a second monitoring module, and an information processing and analysis system. The information processing and analysis system can establish a first evaluation model based on drilling fluid performance parameters before and after pollution, establish a second evaluation model related to CO2 consumption based on drilling fluid pollution index parameters, and establish an overall pollution degree evaluation model related to CO2 injection pressure, pollution time, and pollution temperature based on the first and second evaluation models. The evaluation device of this invention can achieve a comprehensive and dynamic assessment of the CO2 pollution process of drilling fluids and can simulate the complex effects of CO2 on drilling fluids under real-world pollution conditions.

The present invention discloses a system suitable for reservoir rock mass engineering geomechanics test, including a stress control unit, a temperature control unit and a rock sample holder. The stress control unit includes a rigid pressurizing mechanism and a flexible pressurizing mechanism. The rigid pressurizing mechanism uses a pressure head and a pressure block to load rigid pressure on the rock mass, and the flexible pressurizing mechanism uses a pressurizing medium to apply flexible pressure to the rock mass. The present invention adopts two methods of rigid pressurization and flexible pressurization to better simulate the stress state under actual geological conditions and ensure pressurization uniformity. The temperature control unit uses a heating mechanism and a cooling mechanism to control the temperature of the rock mass. At the same time, the insulation mechanism uses an inner spacer and an outer spacer for double-layer thermal insulation, which can ensure that the rock mass reaches and maintains the temperature conditions of the test, providing a guarantee for the test. The present invention also provides a method suitable for reservoir rock mass engineering geomechanics test.

The application relates to a large eddy simulation-based atmospheric flow field reconstruction method, which comprises the following steps: acquiring turbulent flow data of a target region and performing normalization processing; inputting the normalized turbulent flow data into a resolution promotion module to extract high-resolution time characteristics and predict time characteristics at the next moment, and acquiring reconstructed high-resolution data, wherein the resolution promotion module is obtained through training of turbulent flow data and corresponding resolution-promoted turbulent flow data; performing numerical simulation on the reconstructed high-resolution data by using a large eddy simulation model to construct a multi-layer resolution grid structure; and if the resolution accuracy of a local region in the multi-layer resolution grid structure is lower than a preset threshold, adjusting the grid density through an adaptive grid refinement technique to acquire a three-dimensional atmospheric flow field reconstruction result. The application improves the accuracy and time continuity of flow field simulation under complex terrain.