90 results about "Poisson's equation" patented technology

The invention discloses a narrow space spraying trajectory planning method based on reinforcement learning, and the method comprises the steps: obtaining environment data through a LiDAR camera and a TOF camera, and obtaining a LiDAR point cloud and a TOF point cloud; carrying out registration fusion on the LiDAR point cloud and the TOF point cloud, and carrying out normal vector unification processing on each point of the fused point cloud; reconstructing by introducing a Poisson equation improved by shielding perception weight, and eliminating noise through Laplacian smoothing processing to obtain a target point cloud set; designing a state space, an action space and a reward function of reinforcement learning according to the planning of the spraying path trajectory, and determining a spraying path trajectory planning strategy; and based on the target point cloud set, adopting a spraying path trajectory planning strategy to carry out spraying trajectory planning. Sensor data fusion is realized through an improved iterative nearest point algorithm, and a shielding area is repaired in cooperation with a Poisson reconstruction algorithm, so that accurate correspondence between input parameters and output film thickness can be realized, and the control precision of the spraying thickness is remarkably improved.

The invention provides a flue-cured tobacco leaf image distortion correction enhancement method, and belongs to the technical field of flue-cured tobaccos. Distortion segmentation point detection is carried out on an obtained flue-cured tobacco leaf image, and an illumination change area caused by unevenness of a tobacco leaf surface is identified by using an illumination homogenization algorithm based on a Poisson equation; secondly, constructing a distortion jump point matrix to mark gray abrupt change positions caused by shadows and light reflection between adjacent areas, then decomposing and reconstructing multi-scale image features by adopting a Laplacian pyramid image fusion algorithm, and eliminating optical interference generated by surface fluctuation through a designed fusion rule; and finishing processing when the accuracy reaches a set range, otherwise, carrying out correction and optimization again, thereby solving the technical problem that the image quality and the subsequent analysis accuracy are influenced by shadow and light reflection generated during image acquisition due to unevenness of the tobacco leaves in the prior art.

The invention discloses a macro-micro hybrid simulation method for electrical equipment insulation defect analysis, which comprises the following steps: S1, establishing a gas defect discharge process simulation model, and describing microscopic particle state change and electric field distribution in a gas discharge process through a particle transport equation and a Poisson equation; the method comprises the steps of S1, establishing an electromagnetic propagation process simulation model, and describing the propagation process of an electromagnetic signal generated by gas discharge through a Maxwell equation set, and S3, performing joint simulation on the model, taking equivalent current of gas discharge as an excitation source, simulating propagation of the electromagnetic signal in power equipment through a finite time domain difference method FDTD, and recording a signal value of a receiver. According to the macro-micro hybrid simulation method for analyzing the insulation defects of the power equipment, the insulation defects of the power equipment are accurately analyzed and monitored, important support is provided for state monitoring and maintenance of the power equipment, and digital transformation of the monitoring technology of the power equipment is promoted.

The invention relates to a crown segmentation method based on a CSAF framework, and the method comprises the steps: integrating Fourier transform and a state space model through a GM-Mama module, and effectively improving the extraction capability of a fuzzy crown contour; a forgetting factor is dynamically regulated and controlled by using an MASA-Optimizer module in combination with a multi-agent negotiation mechanism, and continuous self-adaptive learning of dynamic changes of species and phenology is realized; a physical constraint based on a Poisson equation is introduced by means of an MPC-Poisson module, and spectral interference of non-target vegetation is significantly inhibited; according to the method provided by the invention, high-precision crown instance segmentation and counting can be realized in a cross-species and cross-phenological complex scene, accurate calculation of a vegetation index of a single plant level is supported, and a flexible and powerful method is provided for analysis of different types of forest crowns.

The invention discloses a multi-view three-dimensional reconstruction method for a slender object based on curve guidance. According to the method, geometric prior and a deep learning segmentation network are fused, two-stage foreground extraction, a curve-guided motion recovery structure (SfM) and a surface optimization module capable of differential Poisson reconstruction are provided, and a set of complete processing flow is established for the problem that a slender and weak-texture object is difficult to accurately reconstruct. The method comprises the following steps: acquiring a multi-view image through a consumer-level terminal and performing frame extraction to obtain an input data set; generating a high-quality foreground mask by utilizing depth estimation and semantic segmentation; performing curve-guided SfM initialization based on the foreground skeleton curve to realize joint estimation of the camera pose and the sparse three-dimensional curve; carrying out surface geometric optimization through a micro Poisson equation and differentiable grid rendering; and designing a multi-branch loss function fusing luminosity, mask, regularization and Gaussian rendering supervision to train the model. Finally, the method can output a three-dimensional reconstruction result of the slender structure with high geometric accuracy, strong structural integrity and smooth surface, and has good robustness and application value.

The invention relates to the technical field of data processing, in particular to an ocean transmission barrier searching method and system and a storage medium, and the method comprises the steps: carrying out the derivation of an earth turning flow field in a target sea area to obtain a differential equation of displacement, and carrying out the time integration of the differential equation to obtain the Lagrange trajectories of all grid point fluid infinitesimal elements in the target sea area; according to a partial derivative relationship between the velocity vector and a flow function, constructing a Poisson equation in combination with a geostationary flow field, and solving to obtain flow function distribution of the target sea area in a target time period; in the target time period, performing time integration on the flow function value of each fluid infinitesimal element in the target area along the Lagrange trajectory of the fluid infinitesimal element, and averaging to obtain a Lagrange average flow function distribution field; determining an LASF isoline of which the maximum line flow function standard deviation is smaller than a standard deviation threshold in the target time period, and determining a transmission barrier based on a substance line coinciding with the LASF isoline; according to the invention, the accuracy and reliability of transmission barrier identification in a complex marine environment are improved.

The invention provides a Gummel curve simulation method and system and electronic equipment, and relates to the technical field of semiconductor device emulation.The method comprises the steps that according to the actual structure of a variable-component SiGe HBT device, division is conducted in combination with a preset grid division strategy, and a discretized grid is obtained; according to the material attribute of the device, carrying out physical attribute binding on each grid point in the discretized grid to obtain a material definition result of the discretized grid; respectively carrying out conventional conductive type doping and component doping on the device to obtain spatial variation doping distribution of the device; combining a preset composite model and a preset mobility model to obtain a physical model set of the device; and a Poisson equation and a carrier continuity equation are solved in combination with numerical values, and a Gummel curve of the variable-component SiGe HBT device is obtained. According to the method, the actual current-voltage characteristics of the device can be accurately reflected, and the core requirement of high-precision simulation is met.

The invention discloses a cable surface defect monitoring method based on image recognition, and relates to the technical field of industrial nondestructive testing, and the method comprises the steps: carrying out the two-dimensional Fourier transform of a cable surface visible light image, extracting a spatial frequency corresponding to a spectrum main peak as a lattice basis vector, and building a space-time crystal model; performing spatial frequency mapping in a reciprocal space of the space-time crystal model, and positioning boundary points of a Brillouin region; analyzing the characteristic value distribution of the group representation matrix at the boundary points of the Brillouin region, identifying the space coordinate position of the symmetry defect, and marking the space coordinate position as a potential defect region; and based on the equivalent divergence distribution diagram, the three-dimensional defect form and depth distribution, executing space trajectory tracking on the identified symmetric defect, screening out isolated abnormal points caused by transient interference, and outputting a monitoring report. According to the method, the mapping relation between the divergence value and the defect depth in the equivalent divergence distribution diagram is established through the Poisson equation, and inversion of the three-dimensional shape and depth distribution of the potential defect area is achieved.

The embodiment of the invention provides a multi-physical field collaborative simulation method and device for a high-speed silicon-based electro-optical modulator, and the method comprises the steps: building a physical model of the electro-optical modulator, carrying out the simulation under different bias voltage and temperature conditions, solving a Poisson equation, a carrier continuity equation and a drift diffusion transport equation, and obtaining the corresponding carrier concentration distribution; performing small-signal simulation analysis on the model to obtain corresponding PN junction electrical characteristic parameters; performing waveguide optical mode simulation on the model to obtain corresponding electric field mode distribution and effective refractive index, and calculating the effective refractive index and absorption coefficient change of the waveguide; constructing a traveling wave electrode model, simulating the traveling wave electrode model to solve an S parameter of a traveling wave electrode, and determining a distribution parameter of the no-load traveling wave microstrip line; a load traveling wave transmission line model is constructed in combination with PN junction electrical characteristic parameters, and based on the model, the frequency response characteristics and the modulation bandwidth of the electro-optical modulator at different temperatures are determined, so that the accuracy of a simulation result can be improved.

The present invention relates to a posterior error estimation method for a finite element numerical solution, and specifically to a posterior error estimation method for a finite element numerical simulation of an electrostatic field Poisson equation based on an artificial neural network. The method is used to solve the shortcomings of existing posterior error estimation methods for finite element numerical solutions of the electrostatic field Poisson equation, which require different method designs for different finite element grid types, resulting in high complexity of algorithms and programs. The present invention uses a pytorch software package to construct a neural network model based on the electrostatic field Poisson equation. After the neural network training is completed by the neural network model, it can be used to estimate the posterior error of the finite element numerical solution under different grids without the need to retrain the neural network. In particular, when the posterior error is only used for a grid adaptive algorithm, compared with the classical posterior error estimation method, the present invention saves computational effort because the process of constructing the posterior error is omitted.

The invention discloses a modeling method for a streamer discharge process in natural ester insulating oil by considering ionization in bubbles. The modeling method comprises the following steps: 1) constructing a two-dimensional non-axisymmetric model for simulating discharge of bubbles to streamers in the natural ester insulating oil under lightning impulse voltage; 2) introducing a plasma chemical reaction, and constructing a natural ester insulating oil stream discharge electrohydrodynamic model coupled with an ionization process of gas molecules in bubbles; and after a set voltage is applied to the two-dimensional non-axisymmetric model, the natural ester insulating oil streamer discharge electro-hydrodynamic model simulates a natural ester insulating oil streamer morphology evolution process. According to the method, the drift-diffusion-reaction process of the plasma in the bubble and the Poisson equation are subjected to self-consistent solution, and the plasma generation process in the bubble under the strong electric field is integrated into the existing streamer discharge model, so that the method is beneficial to understanding the liquid dielectric breakdown theory.

The application discloses a kind of extremely high resolution original drawing image layering segmentation method, equipment, medium and product, it is related to image segmentation field, the method includes obtaining extremely high resolution to be segmented original drawing image;Using SAM model to the original drawing image to be segmented is preliminarily segmented, and obtains preliminary segmentation result;Multiple initial mask regions are contained in preliminary segmentation result;Using GrabCut algorithm, the preliminary segmentation result is globally optimized, edge optimization and residual hole completion, and obtains the segmentation result after optimization;Using Poisson equation, the edge of the segmentation result after optimization is reconstructed, and obtains final segmentation result.The application can be segmented according to the needs of oneself to different granularity size.In the segmentation process, the application realizes maximum degree of automation using SAM model and GrabCut algorithm, in addition to this, the application optimizes the edge of last segmentation result, greatly improves the efficiency and accuracy of image segmentation.

The application discloses a method and device for eliminating mechanical vibration in background schlieren measurement, the method is to collect a photo of a flow field without fluid and a photo of a flow field with fluid, and displacement noise caused by vibration is recorded after the two photos are processed through a cross-correlation algorithm; the two photos collected from the target flow field contain displacement caused by density change in the flow field and displacement caused by vibration of the flow field; the displacement matrix C caused by density difference in the target flow field is obtained by subtracting the displacement matrix B recorded from the reference flow field from the displacement matrix A obtained through cross-correlation operation of the target flow field; Poisson equation is solved according to boundary conditions under test conditions, and the real density cloud picture of the target flow field is obtained according to the Dall-Glassstone formula. Meanwhile, the background plate with controllable light intensity provides different light intensities, which meets the requirement of imaging quality and widens application scenarios.

This invention discloses a method for reconstructing magnetoacoustic electromagnetic particle images based on time-reversal. This method utilizes the method of moments discretization, combining the voltage-time inversion formula with the theory of the magnetoacoustic electromotive force problem for magnetic particles. It reconstructs the spatiotemporal discrete signals of the first and second derivatives of the voltage acquired by the electrodes into an equivalent field source distribution within the imaging region. A unit current is injected across the electrodes, and the potential distribution is obtained by solving the reciprocal potential Poisson equation. Combining the field source and the reciprocal potential distribution, the discrete divergence field of the magnetic nanoparticle concentration can be calculated, and then expressed as a five-point difference form of the Poisson equation. Finally, the Gauss-Seidel iterative method is used to solve this equation to obtain an image of the magnetic nanoparticle concentration distribution. This invention effectively reduces computational complexity and storage requirements, significantly improves imaging speed, and expands the application potential of this technology in the field of medical imaging.

The application provides a method of coupling momentum potential theory and pressure wave decomposition and application thereof, and belongs to the field of fluid mechanics and aeroacoustics. The method first acquires flow field data, performs Helmholtz decomposition on the momentum density field based on the momentum potential theory to obtain momentum densities of fluid dynamics, acoustics and entropy components; then, the momentum densities of the components are taken as independent source terms to construct independent Poisson equations for describing fluid dynamics pressure, acoustic pressure and entropy pressure; and finally, the equations are solved to obtain pressure wave fields of the physical components. The application establishes a physical mapping between momentum modes and pressure modes, gets rid of the dependence on filtering threshold or propagation speed criterion in the prior art, realizes automatic and physically consistent separation of acoustic pressure (true sound) and fluid dynamics pressure (pseudo sound) in complex turbulent flow, and is suitable for noise mechanism research, source positioning and noise reduction design.

The invention discloses a three-dimensional semiconductor device computer numerical simulation method considering a quantum effect. The method comprises the following steps: constructing a drift diffusion model and a quantum correction DG model equation set; discretizing the equation set by adopting an FVM discretization method to obtain a discrete equation set; to-be-solved variables of an electron continuity equation and a hole continuity equation in the discrete equation set are modified, so that discrete formats of a Poisson equation, the electron continuity equation and the hole continuity equation of the drift diffusion model are modified, and calculation of quasi Fermi potential energy in a DG equation is modified; and solving the corrected discrete equation set. According to the three-dimensional semiconductor device computer numerical simulation method considering the quantum effect, a finite volume method discretization format is adopted to discretize a quantum correction DG model, and a quantum correction DG equation and a drift diffusion DD equation are solved through Newton iteration coupling; and the convergence of the quantum correction DG model in complex three-dimensional device simulation can be obviously improved.

The invention discloses a sea target depth map recovery method based on a polarization image and a regularization technology, and belongs to the field of polarization image depth recovery, and the method comprises the following steps: obtaining an original polarization image of a target; based on the four-channel polarization image, according to an illumination estimation model and a Stokes parameter under mirror reflection and diffuse mixing illumination conditions, respectively obtaining zenith angle and azimuth angle images of the target; carrying out polarization information calculation by adopting an analog polarization method to obtain a non-light-intensity polarization image of the target; determining a target surface normal by referring to the polar coordinate system of the target surface normal and using a light source dot product principle, and further obtaining a normal vector field image; gradient information of a normal vector field image is converted into a large discretization sparse matrix under a Poisson equation, on the basis of solving the Poisson equation through a biconjugate gradient method, an adaptive fusion regularization formula is proposed to stably solve the Poisson equation, and depth image recovery of the marine target is completed.

The invention discloses an automatic sparse data labeling method and system based on a graph neural network, and the method comprises the steps: constructing a space-time diagram structure, carrying out the embedding and mapping of traffic flow and speed features through a Grassmann manifold, and generating node features and edge weights; constructing a traffic flow field based on the node features and the edge weights, decomposing the traffic flow field into diffused gradient features and direction-changing curl features, and updating the edge weights; designing a drawing convolution mechanism, and fusing scalar potential and vector potential convolution kernels to generate a propagation kernel; constructing a space-time propagation mechanism through a long and short-term memory network to update node features; a dynamic action range is introduced, and a propagation range is adjusted by combining node density and a traffic flow field; the traffic flow field is used as a confidence index, traffic flow information of marked nodes is spread to sparse nodes through a Poisson equation, and high-precision automatic marking is achieved; the method overcomes the limitation that a traditional method neglects nonlinear dynamic and static propagation, can improve the precision and robustness of sparse node labeling, and is suitable for peak and emergency scenes.

The invention discloses an ADSS optical cable potential measurement method and device based on a CNN-PINN fusion network, and a medium, and the method comprises the steps: S1, collecting a space electric field vector around an optical cable and environment temperature and humidity parameters, and constructing a three-channel tensor; s2, extracting a high-dimensional feature vector; s3, a PINN physical constraint module is embedded into a Poisson equation, field intensity gradient constraint and dielectric compensation; s4, outputting a three-dimensional space potential distribution field and an electric corrosion risk level; according to the method, CNN and PINN are fused and improved, and the bottleneck of the traditional technology is thoroughly broken through: high-gradient distortion is captured through UNet, key area perception is enhanced through CBAM, and the measurement error is reduced compared with that of a traditional contact sensor through PINN physical constraint; environmental interference is decoupled through three channels, humidity drift is suppressed through dielectric compensation, and reconstruction precision stability in a strong interference environment is guaranteed through frequency domain noise reduction; and the output layer generates three-dimensional potential field distribution, calculates an electric corrosion risk level integrating a space gradient and a time-varying characteristic in real time, and supports a corrosion protection decision.

This invention belongs to the field of image generation, specifically relating to a method and system for generating 360° images of subway vehicles based on image stitching. The method includes: acquiring source images collected along the subway vehicle; obtaining effective matching point pairs through feature extraction matching and spatial topological similarity filtering; selecting the camera pair with the strongest geometric constraints as a benchmark; using incremental triangulation to recover the camera pose and 3D point cloud; in global bundle adjustment, optimization is performed by combining reprojection error and linear regularization terms based on the geometric principal axes of principal component analysis, with the regularization term weights adjusted according to the average distance from the point cloud to the principal axes; after optimization, the source images are projected onto a cylindrical canvas, and the optimal image gradient is selected based on gradient information entropy to construct a gradient field; a seamless 360° panoramic image is generated by solving the Poisson equation. This invention can effectively eliminate mismatches, suppress 3D reconstruction offset, and generate high-quality seamless panoramic images.

The invention discloses a depth map adjusting method for relieving 2D (two-dimensional) to 3D (three-dimensional) video display crosstalk. According to the method, an initial depth map is obtained through monocular depth estimation, a spatial weight map of a crosstalk sensitive area is generated in combination with a screen spatial position, a motion weight map is generated based on an inter-frame optical flow, and a gradient magnitude map of the initial depth map is calculated; fusing the multiple weight maps to generate a self-adaptive smooth weight map, performing attenuation optimization on a gradient field of the initial depth map, and performing reconstruction by solving a Poisson equation to obtain an optimized depth map; and finally, left and right eye views are generated based on the optimized depth map, and a 3D video stream is synthesized and output. The method does not need to depend on hardware modification, can intelligently target a crosstalk high-risk area, retains the stereoscopic impression of a central area to the greatest extent while effectively relieving crosstalk, is adaptive to various monocular depth models and 3D display equipment, has the advantages of accurate crosstalk suppression, excellent stereoscopic vision experience, high compatibility and the like, is suitable for various 2D-to-3D video processing scenes, and has a wide application prospect. And the 3D watching comfort of the user is improved.

The present invention discloses a T-type Halbach plug-in axial flux motor and an analytical method thereof. The motor structure is a double-stator single rotor, and a plug-in T-type Halbach magnetic pole array is provided between the rotor and each stator, and the magnetic pole array includes a plurality of magnetic poles. In the method, a subdomain model method is first used to divide the motor into three subdomains, and the magnetization intensity components of different subdomains are respectively given to establish a double Fourier decomposition expression of the total magnetization intensity. Secondly, according to the Poisson equation and the Laplace equation in a three-dimensional coordinate system, the separation of variables method is used to calculate the general solution expression of the scalar magnetic potential of each subdomain. Then, according to the interface connection condition of the Ampere circuit law and the flux continuity principle, the unknown coefficients in the scalar magnetic potential are solved, and based on the relationship between the magnetic field intensity and the magnetic flux density and the scalar magnetic potential, the general expressions of the magnetic field intensity and the magnetic flux density are calculated. Then, the Carter coefficient is introduced to correct the magnetic flux density to obtain the corrected magnetic flux density expression.

Embodiments of the present application disclose an image processing method and related equipment. The method can be applied to a scene of removing other subjects irrelevant to a first subject in an image or a video. The method comprises: acquiring a first image and a second image; determining a first region of the first image and a second region of the second image; constructing a Poisson equation based on a target region, the target region comprising the first region and the second region; and repairing the first region and the second region based on the Poisson equation to obtain a first target image and a second target image, the first target image and the second target image not comprising a second subject. Other subjects irrelevant to the first subject in multiple images can be removed to obtain a target image with the first subject and a clean background. Moreover, multiple repaired target images are obtained by repairing target regions based on the Poisson equation constructed based on the target regions. Compared with the prior art which needs to construct a Poisson equation for each image, the image repair rate can be improved.

The application discloses a kind of cable branch box electromechanical collaborative test method, its steps are: using random mechanical excitation force in external environment as passive test source, three-dimensional acceleration mechanical vibration wave and metal shielding ground line wideband weak current of box shell are synchronously collected, and double-mode time sequence is generated;Based on coherent integration algorithm, filter out power frequency induction component, extract induced displacement current and calculate electromechanical conversion coupling coefficient matrix;Using Fourier transform fitting dielectric relaxation response rate, output frequency-dependent dielectric relaxation characteristic vector;Response rate is mapped to different physical depth of insulating layer by combining radial propagation sound path difference model, and radial space charge concentration gradient parameter is output;Based on Poisson equation, reconstruct micro electric field stress distortion curve by reverse integration;Extract maximum stress peak value and compare with dendritic breakdown threshold value, and automatically output adaptive adjustment instruction.The application does not need to apply electrical excitation additionally, and can passively perceive internal space charge distribution and electric field stress distortion of insulating medium.

The invention discloses a method and a system for calculating a three-dimensional synthetic electric field of a direct-current transmission project, and belongs to the technical field of high-voltage direct-current transmission. The method comprises the following steps: establishing a calculation model of a three-dimensional synthetic electric field of the direct-current power transmission project; initializing space charge density, solving a potential Poisson equation and a current continuity equation of the three-dimensional synthetic electric field of the direct-current power transmission project by utilizing a finite element method based on the calculation model and the initialized space charge density, and obtaining a solving result; judging whether the relative error of the solving result meets an error limit value or not, and if not, continuing to iteratively solve the potential Poisson equation and the current continuity equation; if yes, whether the Kaptzov hypothesis meets the preset condition or not is judged, if not, the surface charge density of the conductor is corrected, the potential Poisson equation and the current continuity equation continue to be iteratively solved, and if yes, iteration is terminated, and all calculation is finished. The method can be applied to calculation of any three-dimensional composite electric field in a direct-current transmission project.

The application provides a numerical simulation method of a short channel negative capacitance surrounding gate field effect transistor, constructs a source-drain current model of a short channel negative capacitance surrounding gate field effect transistor NC-GAAFET with the structure of MFIS; the source-drain current model of a conventional GAAFET is analyzed, the electrostatic potential obtained by solving a cylindrical two-dimensional Poisson equation is obtained; the source-drain current I ds is obtained according to the boundary conditions near the source region and the drain region, and the drain-induced barrier lowering effect parameter is calculated; the source-drain current model of the conventional short channel GAAFET is coupled, and the source-drain current model of the NC-GAAFET is obtained by combining the flexoelectric effect; the subthreshold swing of the NC-GAAFET is calculated; step 6: repeating steps 2 to 5, adjusting different ferroelectric layer thicknesses, structure parameters and flexoelectric coefficients, and obtaining electrical properties under different conditions.

The invention discloses an insulator sample internal charge transport simulation method and system based on a Nebula simulator, and the method comprises the steps: mapping a continuous or non-grid particle accumulation behavior into a charge density source item on a finite difference grid through a finite grid drift diffusion module after obtaining scattering simulation data of the Nebula simulator; then, a strong coupling iteration mechanism is used for simultaneously solving a Poisson equation and carrying out drift diffusion calculation in each micro time step so as to accurately and efficiently iteratively calculate the electric field evolution process in the insulator, and the charge density, the electric field and the dynamic evolution process in the insulator are fed back to a Nebula simulator; therefore, the technical problem that an existing Nebulla simulator neglects an internal electric field and cannot effectively describe the long-term accumulation and evolution process of charges in a material is solved, and the simulation precision of electron beam measurement and imaging is improved.

The application discloses a large-view-field three-dimensional density field reconstruction method and a measuring system based on an air-based multi-platform array, and belongs to the technical field of flow field measurement.The air-based multi-platform array is an airborne platform carrying at least five cameras;the air-based multi-platform array synchronously collects natural background images without flow field disturbance and disturbance images with flow field disturbance, and the natural background images without flow field disturbance are taken as reference images;the air-based multi-platform takes the reference images under the field angles of the cameras as the reference, estimates the displacement of each pixel point in the corresponding disturbance image through an optical flow algorithm, and obtains the displacement field of each disturbance image;the NeDF model is constructed through a multi-layer perception mechanism, the three-dimensional refractive index gradient is reconstructed from the displacement field based on the NeDF model;the three-dimensional Poisson equation is solved by using the three-dimensional refractive index gradient field, and the three-dimensional refractive index distribution is obtained;the Gladstone-Dale equation is solved by using the three-dimensional refractive index field, and the three-dimensional density field of the flow field is obtained.The application realizes high-fidelity three-dimensional density field reconstruction of a super-large-range flow field by using a natural background.

The application provides a transistor interface state model optimization method, a computing device and a storage medium. The method comprises the following steps: constructing a transistor interface state model; if the defect type is an acceptor type defect; calculating the ionized acceptor type interface state defect concentration according to an acceptor type ionization probability model; adding the acceptor type interface state defect concentration to a Poisson equation to obtain an acceptor type Poisson equation; replacing the Poisson equation with the acceptor type Poisson equation to obtain an acceptor type transistor interface state optimization model; if the defect type is a donor type defect; calculating the ionized donor type interface state defect concentration according to a donor type ionization probability model; adding the donor type interface state defect concentration to the Poisson equation to obtain a donor type Poisson equation; replacing the Poisson equation with the donor type Poisson equation to obtain a donor type transistor interface state optimization model. The beneficial effect of the scheme is that the transistor interface state model is optimized, and the base current simulation result obtained by using the optimization model is more accurate.

The invention provides a density field topological optimization method and system for coupling a derivable symbol distance function, and belongs to the technical field of topological optimization design, and the method comprises the steps: predefining a density boundary threshold value, carrying out the interpolation calculation according to the density threshold value, and obtaining a zero horizontal plane of a symbol distance field; solving a Poisson equation to obtain a symbolic distance function by taking the symbolic distance value of the zero horizontal plane as a Dirichlet boundary condition and a temperature field gradient; on the basis of the obtained sign distance function, the average curvature of the structure boundary is obtained through calculation, and a curvature constraint function is constructed and comprises a positive curvature constraint and a negative curvature constraint; sensitivity information of design variables to the constraint function and the target function is obtained, a standard optimization solver is used for solving an optimization problem, and an optimal solution meeting the convergence condition is used as an optimal topological structure after multiple steps of iteration.