150 results about "Mixed finite element method" patented technology

The computational efficiency of Finite Element Methods (FEM) on parallel architectures is typically severely limited by sparse iterative solvers. Standard iterative solvers are based on sequential steps of global algebraic operations, which limit their parallel efficiency, and prior art techniques exploit sophisticated programming techniques tailored to specific CPU architectures to improve performance. The inventors present a FEM Multigrid Gaussian Belief Propagation (FMGaBP) technique that eliminates global algebraic operations and sparse data-structures based upon reformulating the variational FEM into a probabilistic inference problem based upon graphical models. Further, the inventors present new formulations for FMGaBP, which further enhance its computation and communication complexities where the parallel features of FMGaBP are leveraged to multicore architectures.

A variable discretization method for general multiphase flow simulation in a producing hydrocarbon reservoir. For subsurface regions for which a regular or Voronoi computational mesh is suitable, a finite difference / finite volume method (“FDM”) is used to discretize numerical solution of the differential equations governing fluid flow (101). For subsurface regions with more complex geometries, a finite element method (“FEM”) is used. The invention combines FDM and FEM in a single computational framework (102). Mathematical coupling at interfaces between different discretization regions is accomplished by decomposing individual phase velocity into an averaged component and a correction term. The averaged velocity component may be determined from pressure and averaged capillary pressure and other properties based on the discretization method employed, while the velocity correction term may be computed using a multipoint flux approximation type method, which may be reduced to two-point flux approximation for simple grid and permeability fields.

The invention discloses a finite element method-based storage battery thermal management analysis and optimization method. The finite element method-based storage battery thermal management analysis and optimization method comprises the following steps of: acquiring parameters of a plurality of batteries and a battery pack respectively and establishing three-dimensional models of the plurality of batteries and the battery pack; performing the meshing of a definite element on the three-dimensional models of the plurality of batteries and then performing thermal field analysis after the meshing to acquire temperature field analytical results of the plurality of batteries; performing fluid mechanics meshing on the three-dimensional models of fluid in the battery pack and then after the meshing, using temperature field analytical results of the plurality of batteries as constraint conditions of the plurality of batteries in the fluid grid models in the battery pack; performing fluid mechanics calculation on the constrained grid models to acquire flow field analytical results of the storage batteries; judging whether the flow field analytical results accord with preset conditions or not; and if the flow field analytical results do not accord with the preset conditions, optimizing the design scheme of the battery pack and establishing the models again. The finite element method-based storage battery thermal management analysis and optimization method is not limited to mathematic analysis capacity, has higher adaptability and solving capacity, does not need entity models, is economic and quick in the analytical process, and has higher degree of freedom and flexibility.

The invention relates to a multi-support shafting finite element method with bearing stiffness coupling nonlinearity considered. The method comprises the steps of (1) establishing a local coordinate system of each bearing and obtaining a six-freedom-degree bearing stiffness matrix of each bearing under the corresponding local coordinate systems through calculating by means of the difference method with a corresponding bearing load calculation formula; (2) dividing a shaft into different shaft sections according to the difference of diameters, establishing nodes at the positions of bearing section starting positions, the bearing installation position and load action positions, then establishing finite element models of the shaft between the nodes, and obtaining a stiffness matrix of the whole shaft; (3) establishing a global coordinate system of a shafting, establishing another node at the bearing installation position to be used for simulating a fixed end, integrating the bearing stiffness matrix onto the installation node and the corresponding fixed end node on the shaft to obtain the overall stiffness matrix of the shafting, and establishing the stiffness equation of the shafting; (4) restraining the freedom degrees in six directions of the fixed end node, restraining the freedom degree, of rotating around the axis, of the node at the position of origin of the global coordinate system of the shafting, reducing the stiffness equation of the shafting according to restraining conditions, and solving the reduced stiffness equation of the shafting with the Newton-Raphson method in an iterative mode to obtain the displacement and load of each node.

The invention discloses a transmission gear fatigue life assessment method based on defect modeling. The method includes the following steps that (1), crack occurrence rules and reasonable crack forms are analyzed, and the method for introducing cracks into a gear model is researched; (2), material fatigue characteristic parameters are obtained through test data fitting or a numerical value estimation method; (3), a stress spectrum under the condition of working situations of a gear containing the cracks is calculated by using a finite element method, and the stress spectrum is corrected according to obtaining conditions of the material fatigue characteristic parameters; (4), fatigue life under the condition of working situations of a gear containing the cracks is calculated by using the reasonable fatigue damage accumulation theory. The fatigue life assessment method based on defect modeling is provided for the transmission gear, and an effective detection basis is provided for the remanufacturing process of a gear recycled part.

The invention belongs to the technical field of mechanical analysis and numerical solution of a three-dimensional structure and relates to a contact analysis method in three-dimensional mechanical finite element model analysis. The method comprises the steps of firstly modeling a target electron device structure; introducing a displacement boundary condition or a stress boundary condition to establish a corresponding geometric structure model; generating a contact relation between parts according to a mutual relation between the parts; then subdividing the established geometric structure model by use of a tetrahedral mesh; generating a joint surface mesh according to a surface mesh in a simulated area contact surface; and finally establishing a finite element general intrinsic equation of a target electron device in consideration of a contact problem by use of a finite element method, solving the equation to obtain a characteristic value and a characteristic vector, and post-processing to obtain a vibration modal frequency and a vibration model. Therefore, the contact analysis in modal analysis is realized and a high-precision numerical calculation result is obtained.

The invention provides a statistical energy analysis parameter acquisition method based on a finite element method and a power input method. A finite element model of a structure is established according to a geometric model of the structure, boundary conditions are applied, the finite element model of the structure is subjected to mode analysis, a mode of a system and a rigidity matrix and a mass matrix of a subsystem are obtained, vibration energy of the subsystem and input power of the system are obtained through a power flow model, and statistical energy analysis parameters are obtained through calculation through the power input method. The finite element method and the power input method are combined to obtain the statistical energy analysis parameters of the structure, the calculated rigidity matrix, mass matrix, mode shape and inherent frequency are obtained through the finite element method, the input power and the vibration energy are obtained through calculation through the power flow model, and the input power and the vibration energy are substituted into the power input method for calculation to obtain the statistical energy analysis parameters. According to the method, statistical energy analysis can be carried out aiming at the problem of strong coupling between structures, the limitation about weak coupling assumption between the structures in existing statistical energy analysis is solved, and the method can be popularized to complex structures.

System, method and software product for numerically simulating structural behaviors of an engineering product in compressible and near-incomprssible region is disclosed. Meshfree enriched finite element method (ME-FEM) is used for such numerical simulation. ME-FEM requires an engineering product be represented by a FEM model comprising a plurality of finite elements. Finite elements used in the ME-FEM are generally low-order finite elements. Each of the finite elements in the FEM model is enriched by at least one meshfree enriched (ME) node located within the element's domain. Each ME node has additional degrees-of-freedom for the element it belongs independent from those of the corner nodes. A displacement based first-order convex meshfree approximation is applied to the ME node. The convex meshfree approximation has Knonecker-delta property at the element's boundary. The gradient matrix of ME-FEM element satisfies integration constraint. ME-FEM interpolation is an element-wise meshfree interpolation that is discrete divergence-free at the incompressible limit.

A finite element method for on-line calculation of slab band temperature during hot rolling belongs to the rolling technical field, and comprises the following steps that: (1) the establishment of a finite element model: unit division of the cross section of a rolling piece is carried out to set up a finite element analysis model; unit node numbering is carried out to calculate nodal coordinate; (2) the determination of parameters during calculation: the data needed during calculation includes initial information, rolling parameters, thermal physical property parameters of material, unit division information, control parameters and heat exchange coefficient; (3) the establishment of finite element solving equations set: through combining spatial domain finite element discretization and time domain finite difference, the solving linear equations set of temperature field finite element is set up; (4) pass rolling starting temperature is calculated; (5) on-line application is carried out. The invention has the advantages that: the invention can obtain high temperature calculation precision and the detailed information of slab band temperature distribution during entire hot rolling, thereby providing set parameters and optimized parameters for the rolling process; meanwhile, with strong practicality, the invention reduces calculation time and improves calculation efficiency; moreover, the invention is suitable for online application and optimization.

A user uses part data forming means 101 and module data forming means 102 to input the actual dimension, physical constants and mesh dividing number for fundamental shapes which are registered in advance, thereby forming parts, and then indicates the relative position between the parts to form the entire shape of an assembly of plural parts without paying attention to coincidence or non-coincidence of nodal points. Data converting means 104 divides the shape of each part thus assembled according to the indicated mesh divisional number to generate element data and nodal point data. Further, it generates a constraint equation for connecting nodal points which are non-connected between neighboring parts, and forms an analysis model 401. A finite element method analyzer 105 uses approximate calculation means 106 to approximate a non-connected nodal point displacement from a nodal point displacement of neighboring structural elements on the basis of the constraint equation.

The invention discloses a sensitivity analysis based dynamic construction control method for one-time tension of stay cables. The method comprises the following steps: a cable-stayed bridge completion time model is established with finite element software, the model is calculated with a finite element method, and tension import value of each stay cable is obtained; uncontrollable external factors are subjected to sensitivity analysis according to the established cable-stayed bridge completion time model; the sensitivity of each stay cable is determined through analysis according to the established cable-stayed bridge completion time model; multiple tension sequence schemes and corresponding tension force are predetermined according to stay cable sensitivity analysis results, the different tension sequence schemes are calculated with the finite element method, and the optimal tension sequence scheme is selected; construction is performed according to the optimal tension sequence scheme obtained through calculation, the tension force of each stray cable and a result about sensitivity analysis of the uncontrollable external factors. With adoption of the sensitivity analysis based dynamic construction control method for one-time tension of the stay cables, the stay cable construction process is simplified.

The invention discloses a Yeh-multi-scale finite element method for simulating water flow Darcy velocity of a porous medium, comprising the steps of: performing variation on the problem which needs resolution by a Galerkin method; subdividing a research area into coarse grid cells and subdividing all the coarse grid cells into fine grid units; resolving a degradation elliptic equation on each coarse grid cell to construct a basis function; resolving variational form by applying the basis function to obtain a total rigidity matrix; obtaining a right-hand term according to the source sink term and the boundary condition of the research area; performing simultaneous operation to obtain a waterhead equation set; resolving the equation set by an effective numerical method to obtain the node waterhead of the research area; and resolving a Darcy equation directly in the research area by combining a Galerkin finite element model of Yeh and applying the constructed basis function and the waterhead value of the research area to obtain continuous Darcy permeating velocity on the coarse-scale node, and linearly expressing the fine-scale Darcy permeating velocity by the basis function. Compared with the prior art, the method has similar precision and higher efficiency.

The invention discloses a 3D free bending forming technology optimization method for complex metal components. The method includes the steps that quantitative relations between geometrical shape parameters of a straight section, a transition section and a bent section and movement velocity u of a spherical bearing within the X / Y plane, pipe Z-axis feed-in velocity v, movement time t and a distance A between the center of a bending mold and the front end of a guiding mechanism are established; a finite element method is adopted for establishing a 3D bending model, and a correction factor k is introduced into the quantitative relations; the quantitative relations with the introduced correction factor k are adopted as bending technology parameters, bent pipe finite element repeated iterative computation is started, computation results are imported into geometry software to be processed, and the dimension difference between the computation results and the established 3D bending model is obtained through comparison; based on an established dimension error criterion, whether the correction factor k is modified, and iteration is performed again is determined; when the error is smaller than a set value, iteration is ended; final technology parameters are transmitted to equipment to execute actual bending forming.

The invention belongs to the technical field of detection and discloses a load measurement-based residual stress detection method. The method comprises the following steps: establishing a computer geometric model of a to-be-detected member, applying restrictions to the model by a finite element method, performing linear elasticity finite element static analysis after a load is applied to a loading point, reading stress value sigma'<b1> of an experimental measuring point, and adjusting the position of the loading point or increasing the load so as to enable the maximum value of the component of the stress value sigma'<b1> to be larger than 0.1MPa if the maximum value of the component of the stress value sigma'<b1> is less than 0.1MPa; applying restrictions and a load same as those of the computer geometric model to the to-be-detected member, and performing a strain experiment to obtain a stress value sigma<c1> of the experimental measuring point; and transforming the sigma'<b1> under local coordinates of the experimental measuring point to obtain sigma<b1>, and subtracting sigma<b1> from sigma<c1> to obtain residual stress value sigma<a1> of the member at the experimental measuring point. According to the method disclosed by the invention, the surface of the member is not damaged in the detection process, and the problem that a member is damaged due to a conventional mechanical residual stress method is solved. In addition, the method is convenient, simple and flexible to apply.

The invention aims at providing a structural vibration analysis method based on a finite element method and a generalized Fourier series method, which comprises the following steps of: dividing a structure region to be subjected to vibration analysis to respectively and correspondingly form a finite element expression region and a generalized Fourier series expression region; dividing the finite element grids of the finite element expression region to form a corresponding quality stiffness matrix, and selecting a corresponding assumption displacement form according to the characteristics of the generalized Fourier series expression region to form a quality stiffness matrix; subsequently, establishing a virtual spring between the two regions, and converting the potential energy of the virtual spring into an overall coupling stiffness matrix by using an energy variation method; then, arranging the formed quality stiffness matrices according to displacement to form an overall structure quality stiffness matrix; and solving linear equations to obtain an unknown coefficient in corresponding node displacement and series expansion. When applied to a large complex structure, the method not only can obtain precision higher than the finite element method, but also can save a large amount of calculation cost.

The invention discloses a random thermal homogenizing analysis method of a two-phase composite material. The method comprises the following steps: (1) establishing a corresponding microscopic homogenization model of a two-phase composite material composed of a matrix and particles, constructing constitutive equations, and calculating the thermal boundary values of the heterogeneous material; (2) determining and analyzing to obtain one RVE from the two-phase composite material, and determining the effective constitutive relationships among the effective heat conduction coefficient, the volumetric average temperature gradient and the volumetric average heat flux, wherein the volumetric average temperature gradient and the volumetric average heat flux are obtained from RVE; (3) applying boundary conditions on RVE, and at the same time carrying out finite element analysis and calculation to obtain the value of effective heat conduction coefficient of the RVE numerical model; (4) establishing a random homogenizing model to obtain the macroscopic effective quantity of the composite material. In the provided method, a finite element method and Matlab software are used to solve the problem of complicated RVE numerical modeling, the influences of parameter randomness of components of the composite material under three boundary conditions on the macroscopic thermal physical properties are taken into account comprehensively, and thus the provided method has an application value, a certain academic value, and theoretical significance.

The invention provides a soft tissue deformation method based on a finite element model of an octree mesh. The method comprises: drawing a three-dimensional model of soft tissue, constructing a plurality of uniform hexahedral meshes based on the AABB method, and generating an octree mesh based on the mesh generation algorithm of an octree based on the hexahedral mesh, modeling the hexahedron meshby the finite element method, solving the deformation process of the soft tissue, assembling the element stiffness matrixes on adjacent element nodes into the total stiffness matrix of the discrete domain, solving the dynamic equilibrium differential equation for each matrix by time integral under dynamic equilibrium, to obtain the displacement of the nodes with time, and displaying the stress deformation of the soft tissue by the rendering of the displacement of the nodes. The method realistically simulates the process of tension deformation of the soft tissue epidermis of any shape in virtual surgery, has high real-time performance, reduces the calculation amount, and solves the problem that the number of finite element meshes is complicated and the soft tissue deformation process cannotbe simulated in real time.

The invention discloses a fluorescent molecular tomography imaging reconstruction method based on limited projection of double grids. The method includes the following implementation steps that firstly, measurement data is acquired; secondly, the coarse grid is reconstructed, and a feasible region of a fluorescent target is provided for reconstruction of the refined grid; thirdly, the refined grid is reconstructed under the feasible region, and three-dimensional distribution of the fluorescent target is achieved. The fluorescent molecular tomography imaging reconstruction method is based on the optical transmission theory and a finite element method, optical property parameters, anatomical structures and other prior information are utilized, single-point excitation is adopted, limited angle measurement is adopted, a double-grid strategy is utilized, a reconstructing result of the coarse grid provides the feasible region of the fluorescent target for reconstruction of the refined grid, and therefore morbidity of problems is reduced, a reconstruction result of fluorescent molecular tomography imaging is effectively improved, and the method has important application value in the fields of optical tomography three-dimensional reconstruction algorithms and the like.

The invention relates to an electrical tomography image reconstruction method based on a convolutional neural network. The method comprises the following steps: solving a positive problem of electrical tomography by adopting a finite element method; Designing a convolutional neural network structure to enable the convolutional neural network structure to be suitable for an electrical tomography image reconstruction process; Determining a loss function; Updating the network parameters by adopting a small-batch gradient descent strategy, and synthesizing the parameters obtained by each round ofiteration by using a moving average model to determine a final parameter updating value; After the iteration is finished, obtaining a convolutional neural network of which the connection weight and the threshold are determined; When the image is reconstructed, taking the actually measured boundary measurement value as a trained convolutional neural network input layer neuron, wherein the output ofthe output layer neuron is the value of each pixel point in the image.

The invention discloses a wind power master control cabinet heat management analysis and optimization method based on a finite element method. The wind power master control cabinet heat management analysis and optimization method includes the steps of from a cabinet-top arrangement diagram, acquiring attributes of materials of a wind power master control cabinet and parameters of a circuit breaker and a relay inside the wind power master control cabinet; performing initial geometric modeling on the circuit breaker and the relay to obtain a three-dimensional model; subjecting the three-dimensional model to finite element meshing to obtain a finite element mesh model; acquiring design parameters of a heater and a fan; establishing a three-dimensional model for the wind power master control cabinet and fluid inside; performing hydro-mechanical meshing to obtain a flow field mesh model; restraining the flow field mesh model; determining physical parameters of the fluid in the flow field; calculating the temperature field and flow field of the flow field mesh model to obtain flow field analysis results; judging whether or not the flow field analysis results for the wind power master control cabinet meet industrial standards; if yes, ending; if not, optimizing a design scheme. The precise calculation results can be obtained with no need of building a solid model.

The present invention discloses a cubic spline multi-scale finite element method for simulating two-dimension flow movement. The method comprises: firstly converting a problem that needs resolution into a variational version; determining a boundary condition of a research area, setting a coarse grid cell dimension h, dissecting the research area to obtain a coarse grid cell, and dissecting the coarse grid cell into a fine grid cell; resolving a degradation elliptic problem on each coarse grid cell to construct a basis function; resolving a total rigidity matrix and a right-hand side simultaneous equations system by using an effective calculation matrix method; obtaining a water head of each node (the coarse dimension) on the research area; obtaining a coarse dimension water head derivative by using the cubic spline technique, and obtaining Darcy flow velocity of the coarse dimension according to Darcy's Law; obtaining a water head of each node (the fine dimension) by using the basis function and an interpolation formula; and obtaining a fine dimension water head derivative by using the cubic spline technique, and obtaining Darcy flow velocity of the fine dimension according to the Darcy's Law. Compared to the prior art, the method disclosed by the present invention has similar precision, but has higher calculation efficiency.

A variable discretization method for general multiphase flow simulation in a producing hydrocarbon reservoir. For subsurface regions for which a regular or Voronoi computational mesh is suitable, a finite difference / finite volume method (“FDM”) is used to discretize numerical solution of the differential equations governing fluid flow (101). For subsurface regions with more complex geometries, a finite element method (“FEM”) is used. The invention combines FDM and FEM in a single computational framework (102). Mathematical coupling at interfaces between different discretization regions is accomplished by decomposing individual phase velocity into an averaged component and a correction term. The averaged velocity component may be determined from pressure and averaged capillary pressure and other properties based on the discretization method employed, while the velocity correction term may be computed using a multipoint flux approximation type method, which may be reduced to two-point flux approximation for simple grid and permeability fields.

The invention discloses a frequency domain load identification method based on Tikhonov regularization. The method aims at solving the problems existing in frequency response function acquisition in practical engineering application of a matrix inversion method and the morbidity problems existing in calculation. A frequency response function is acquired through a finite element method, the morbidity of mechanical equipment is evaluated through the number of matrix conditions of the frequency response function, and loads are identified based on different regularization methods when the morbidity is different. A proper mobility condition number threshold value is determined. The method has the advantages that the problem that the frequency response function of the mechanical equipment is difficult to acquire in practical engineering application of the matrix inversion method and the morbidity problems existing in calculation can be solved, the load identification accuracy in a frequency domain is improved, and high engineering application value is achieved.

The invention discloses an efficient multi-scale finite element method for simulating two-dimension water flow movement in porous media. The method comprises the steps that a problem needing to be solved is converted into a variation form; boundary conditions of a research area are determined, coarse grid unit dimension is set, the research area is divided, and coarse grid units are obtained; middle grid unit dimension is set, and each coarse grid unit is divided into middle grid units; fine grid unit dimension is set, and each middle grid unit is divided into fine grid units; the degenerate ellipse type problem on the coarse grid units is converted into subproblems of the number of the middle grid units through the area decomposition technique, and values of a multi-scale primary function on all nodes of the middle grid units are obtained by solving the subproblems; a total stiffness matrix can be obtained through the variation form, and a system of simultaneous equations of a water head total stiffness matrix and a right end term is solved through an effective calculation method; water heads of all nodes on the research area are obtained. Compared with a traditional finite element method and a multi-scale finite element method, the calculation efficiency is higher.

The invention relates to a finite element analysis method for simulating ride comfort of an entire vehicle under pulse input. The finite element analysis method comprises the following steps: (1) establishing an automobile finite element model by using computer software at a design and development stage of an automobile; (2) defining a transient excitation buffer strip or triangular bump size model; (3) calculating loading parameters; (4) enabling front and rear wheels to passthrough a buffer strip or triangular bump at a certain speed while simulating left and right wheels of the automobile; (5) calculating transient response of the automobile at certain timeunder aninput determinate signal. The research and development period is shortened; the research and development cost is saved; the operation is convenient; only 2-3 hours are required for calculating the ride comfort of the entire vehicle through the software; the ride comfort of the automobile can be very easily improved through a finite element method.

The invention relates to an aircraft aeroelasticity inertia sensor layout method. According to the method, the aircraft structure dynamic response is solved by employing the finite element method, a finite element model is corrected via a fluid-solid interaction method, and a fluid-solid coupling analysis of all the flight states and disturbance factors is conducted; the relatively precise structure dynamic response is calculated, and the response of each node in the finite element model is obtained; standardization processing of a data set is performed, and an elasticity wing structure dynamic response data set is generated; and data clustering by employing a clustering method based on distance measurement in pattern recognition is finally performed, and the final positions of the sensors are optimized and obtained. According to the method, the maximization of measuring information of sensors is regarded as the overall goal, optimization balance between the quantity of the sensors and the optimal layout positions is performed, and valuable aeroelasticity characteristic values can be acquired at the arranging position of each sensor.

The invention provides a finite element analysis method for calculating the deflection of the assembly axis of a rotor considering the contact elastic deformation, wherein the relationship between theend face runout and the inclination amount is calculated by the finite element analysis method under the condition of considering the contact deformation of the mating surface. Considering such an ideal shell stacking model, the contact part of two shells is tangent to each other when they are initially modeled and assembled, and the two 3D curves are fitted according to the measured end-jump data, which can fully reflect the end-jump topography of the joint surface. The shell stacking model proposed by the invention is based on the finite element analysis method, considers the elastic deformation in the assembly process, and more accurately predicts the axis deflection than the algorithm. Compared with the general finite element analysis model, the shell model proposed in this inventionhas higher computational efficiency, and the finite element model can be optimized continuously to reduce the time of analysis and to give consideration to both analysis efficiency and accuracy.

The invention provides a three-dimensional frequency domain controllable source numerical simulation method. The method comprises the steps of S100: defining a calculation region in a resistivity anomaly body and an adjacent wrapping layer, and dividing the calculation region into multiple regular unit bodies; S200: calculating an initial electric field under an artificial source excitation condition, and by adopting a weighted residual method, establishing a finite element equation set of edge midpoints of the unit bodies in the calculation region; S300: defining an observation point as a boundary node by using a Green function, and enabling a secondary electric field of an internal node to express a secondary electric field of the boundary node; S400: substituting a Green function expression into a mixed equation set to obtain a first equation set only about the secondary electric field of the internal node; and S500: solving the first equation set to obtain secondary electric fieldvalues of nodes comprised in the unit bodies. According to the method, a vector finite element method and a volume integral equation method are applied; the calculation region is reduced; the calculation amount is reduced; the calculation precision is improved; formed equations are sparse in coefficient matrix and good in condition number; and a calculation result is accurate.