13 results about "Density field" patented technology

The invention belongs to the technical field related to structure topological optimization, and discloses a multi-scale topological optimization method and equipment for a TPMS lattice structure, and the method comprises the steps: (1) a teacher network learns a cross-modal fusion relation between a geometric mode and a parameter mode of a TPMS type, and then the cross-modal fusion relation is distilled to a student network, so as to obtain a candidate performance matrix; (2) constructing a multi-objective optimization model containing rigidity maximization and space continuity, and further obtaining a type field; (3) performing density interpolation on the density field, then performing type boundary detection on the type field at the point level, and further performing density interpolation on non-boundary points to calculate and obtain a non-boundary point distance field value; and processing the obtained type weight by adopting a weighted mixing strategy to obtain a boundary point distance field, and compensating the boundary point distance field by adopting mixing degree adaptive compensation to obtain a boundary point distance field value so as to obtain a three-dimensional topological configuration. According to the invention, geometric smooth transition connection is realized.

The invention relates to a mountain stability evaluation method and device, computer equipment and a storage medium, and relates to the technical field of mountain stability. The evaluation method comprises the following steps: dividing a mountain into a plurality of voxels; an adjacent structure diagram and density field posterior probability distribution of the mountain are constructed, the adjacent structure diagram is constructed through muon data recorded by muon detectors arranged in the grotto, and the density field posterior probability distribution is probability distribution of all density models of the mountain under the muon data and the adjacent structure diagram; sampling from the density field posterior probability distribution according to the adjacent structure diagram to obtain a plurality of density field samples of the mountain; obtaining a density value of each voxel according to the plurality of density field samples; and according to the density values of the voxels, obtaining a stability abnormal region in the mountain. According to the mountain stability assessment method, the stability abnormal area in the mountain can be accurately determined, and the accuracy of mountain stability assessment is improved.

The present invention discloses a method and system for mechanical-thermal coupling simulation and structural optimization of ultra-complex structures. The method comprises: generating parameters and data structures required for subsequent simulation and optimization for a given model and design domain; constructing a background coarse grid, seed points and radii, modeling a porous structure based on a Voronoi structure, and precalculating its stress field; calculating a density field, performing thermodynamic simulation on the generated porous structure based on the density field, the background coarse grid, and a shape function; querying the stress matrix corresponding to each seed point based on the obtained stress field, performing eigendecomposition and extraction on the stress matrix, and obtaining a weight matrix corresponding to the seed point; using a topology optimization method, updating control parameters through a GCMMA method based on sensitivity, thereby updating the porous structure and optimizing material distribution, and continuously optimizing until the change in control parameters is less than a threshold. The method can quickly and effectively solve the topology optimization problem of anisotropic porous structures under multi-physics field conditions.

The invention discloses a Tokamak gas leakage detection method and system based on laser speckles, and the method comprises the steps: forming the laser speckles on a background plate, and collecting a speckle image reflected by the background plate in a gas leakage region of the outer wall surface of the Tokamak; after the speckle image is preprocessed, a displacement field is estimated on an overlapped grid window by adopting a multi-scale optical flow algorithm, and a density field is solved by adopting a rotation-field-free projection integration method according to the displacement field; monitoring the deviation of the density field relative to the leakage-free density field so as to determine a leakage cloud cluster formed by a plurality of abnormal pixel points, and positioning the coordinates of a leakage point based on the leakage cloud cluster; extracting spatial static features and time dynamic features according to the density field corresponding to the leaked cloud cluster, splicing the spatial static features and the time dynamic features, and inputting the spliced features into the trained form recognition network model to obtain a leakage form of the leaked cloud cluster; and calculating the volume of the leaked cloud cluster, and then estimating the gas leakage flow of the leaked cloud cluster by adopting a volume control method according to the volume.

The invention relates to the technical field of detection devices, in particular to a repetition-frequency narrow-pulse electromagnetic gas valve jet flow detection device and method. Aiming at the problems that the traditional schlieren imaging system cannot accurately restore the real form and density field distribution of the jet flow of the repetition-frequency narrow-pulse electromagnetic gas valve, the reliability of detection data is influenced, and the accurate analysis and performance optimization of the jet flow characteristics are restricted, the invention provides the following technical scheme: the system comprises a modularized light source assembly; the modularized light source assembly is used for generating a coaxial point light source with adjustable light spot size; the reflection type light path assembly is used for converting the coaxial point light source into parallel light, and the parallel light penetrates through a jet flow field in the vacuum test chamber and then is focused. The real form and density field distribution of the repetition-frequency narrow-pulse electromagnetic gas valve jet flow can be accurately restored, the reliability of detection data is guaranteed, and powerful support is provided for accurate analysis of jet flow characteristics and performance optimization.

The application relates to the technical field of acoustic wave imaging, and discloses an acoustic wave environment imaging method based on an acoustic array. The method comprises the following steps: collecting multi-channel acoustic wave signals in a target space through a distributed acoustic array and extracting time-frequency domain characteristic parameters; constructing a space topology model containing a geometric constraint relationship between an acoustic wave reflection surface and an obstacle according to the parameters; calculating an acoustic wave energy attenuation gradient based on the model, dividing a dynamic partition of acoustic wave energy distribution, performing phase compensation on acoustic wave signals in the dynamic partition, and generating a compensated acoustic wave signal sequence; reconstructing a three-dimensional density field of acoustic wave energy in the target space and marking a non-continuous boundary; and finally, fusing the time-frequency domain characteristic parameters and the three-dimensional density field to generate a multi-modal imaging atlas of the acoustic wave environment. Through accurate processing of multiple links, the method realizes all-around and stereoscopic depiction of the acoustic wave environment, and provides a new technical path for fine perception of the acoustic wave environment.

The invention relates to the technical field of mixture statistics, and discloses a material pile dynamic system based on three-dimensional modeling, which comprises a modeling registration module for collecting and preprocessing three-dimensional point cloud data of a material pile and outputting a terrace reference surface and a height field; the segmentation identification module is used for carrying out retaining wall identification, dividing a material pile into a plurality of independent material pile areas and establishing a three-dimensional voxel grid model; the density solving module is used for constructing a density field model of the material pile and outputting a parameterized density field model and density field parameters; the quality fusion module is used for carrying out inversion updating on density field parameters in combination with external quality observation data; the weight output module is used for generating a final material pile total weight metering result; according to the method, the three-dimensional density field model is constructed, the vertical compaction effect, the horizontal separation effect and the time compaction effect are comprehensively considered to deal with the influence of the density of the material pile, the accuracy of weight metering is improved, the stability of parameter estimation is improved through density field parameter inversion, and stable operation under the complex working condition is guaranteed.

The disclosure provides a three-dimensional reconstruction method and device, a terminal and a storage medium. The three-dimensional reconstruction method comprises: acquiring a video including a human body and clothes worn by the human body; processing the video by using a trained neural network model to obtain a three-dimensional reconstruction model of the human body wearing the corresponding clothes, wherein processing the video by using the trained neural network model comprises: performing human three-dimensional reconstruction based on the video; performing clothes capturing and rendering based on the video; wherein performing clothes capturing and rendering based on the video comprises: obtaining a three-dimensional position of a clothes standard space point in a video frame of the video; obtaining a density field and a color field based on the three-dimensional position; extracting explicit geometric grid information from the density field; and generating a clothes model in the three-dimensional reconstruction model based on the density field, the color field and the explicit geometric grid information. The disclosure can effectively improve the detail performance of clothes and ensure that the geometric characteristics of clothes under complex shapes can be accurately captured through explicit geometric constraints.

The present invention relates to a method for reconstructing an asteroid gravitational field based on a neural density field with embedded mathematical knowledge, comprising: completing data point sampling based on an asteroid particle group model, constructing a function solver for converting spatial position to density; constructing a spatial sampling point function to obtain spatial density of spatial points at different positions; for spatial points, using an asteroid visual three-dimensional model to obtain celestial shell information, and screening spatial points inside and outside the celestial shell as prior information; screening the results of the density function solver based on the prior information; further using a partial differential equation solver to calculate the Laplace equation of the data point and using numerical integration to calculate the gravitational acceleration equation of the spatial point outside the celestial body; constructing a total loss function; performing error back propagation and optimizing network parameters at the same time; recalculating the density of spatial points inside the celestial body to obtain the internal density distribution of the asteroid; and further calculating the value of the gravitational acceleration at each spatial position to obtain the external gravitational field distribution of the asteroid.

This invention relates to an optical measurement method for aero-engine thrust using background schlieren technology. The method utilizes the principle that the refractive index gradient of light in the measured flow field is proportional to the airflow density, transforming changes in the density gradient in the flow field into changes in relative light intensity on a recording plane. An LCD display screen and a high-speed camera for acquiring background schlieren images are respectively arranged on both sides of the observation section of the outlet flow field to be measured. When the density of the observation area changes, light rays from the background pattern are refracted through the areas of density change. The background image is processed using a cross-correlation calculation method to calculate the displacement of points on the background image between every two images, converting it into a digital image field for calculation to obtain the velocity and density fields. These are then combined with the thrust calculation formula to obtain the engine thrust. This invention does not affect the internal structure of the outlet flow field to be measured, is easy to operate, has a large measurement range, fast response speed, high measurement accuracy, and lower cost.

The invention discloses an underground space structure three-dimensional modeling method and system driven by muon imaging data, and relates to the technical field of underground space detection, and the method comprises the steps: collecting muon detection data transmitted and scattered in an underground space target area, and obtaining transmission and scattering imaging data; global scanning is carried out on transmission data, a density field preliminary distribution model is established, scattering data event clustering and trajectory reconstruction are carried out, and a local supplementary model is generated; performing iterative optimization on the two models to obtain a three-dimensional modeling parameter matrix, and combining external physical field data to form a multi-source fusion system; and based on the system, executing a three-dimensional reconstruction algorithm to establish an underground space structure model. According to the method, the technical problems of incomplete modeling and limited precision caused by the fact that the existing underground space modeling technology does not combine muon transmission imaging and scattering imaging to obtain detection data are solved, and the technical effects of realizing three-dimensional modeling of underground space from global to local and single to multi-source fusion and improving modeling comprehensiveness and precision are achieved.

The invention relates to the technical field of flow field measurement and three-dimensional reconstruction, and discloses a single-view schlieren map three-dimensional density field reconstruction method based on multi-mode aerodynamic embedding. The method comprises the following steps: acquiring a single-view schlieren / background schlieren image and surface pressure distribution of a target object, and establishing a three-dimensional reconstruction domain and projection mapping; schlieren features and pressure tokens are extracted through a schlieren encoder and a pressure encoder respectively, a three-dimensional query point performs cross attention retrieval on the two types of tokens on the basis of projection alignment, and adaptive fusion is performed through learnable gating; and fusing the features, outputting a density field through a three-dimensional decoder, and introducing a micro-projectable consistency constraint and a structure retention constraint to optimize a reconstruction result. According to the method, the surface pressure is used as strong boundary prior, the depth ambiguity of single-view reconstruction is relieved, and the density field space positioning precision and the structure fidelity are improved.

The invention belongs to the technical field of computational fluid mechanics, and discloses a pool boiling bubble three-phase interface identification and contact angle measurement method, which comprises the following steps of: reading simulation data of a multiphase flow simulation model for preprocessing to obtain a density field and a boiling pool wall surface geometric equation; presetting density thresholds, and extracting contour point sets of each layer of the gas-liquid two phases under different density thresholds from the density field in combination with layered contour analysis; screening a target contour point set representing a bubble main body in each layer of contour point set, and fitting the target contour point set to obtain a complete bubble contour elliptic equation; determining a target contour equation for positioning a three-phase contact point; the target contour equation and the boiling pool wall surface geometric equation are combined, and three-phase contact points are solved and determined; the problems of low three-phase contact point identification precision and insufficient contact angle measurement result reliability caused by poor adaptability to weak gradient and irregular bubble morphology in the prior art are effectively solved.