209 results about "Extended 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.

The invention discloses a transformer electromagnetic vibration noise calculating method based on the finite element method. The transformer electromagnetic vibration noise calculating method comprises the steps of 1, establishing a structured grid model and an acoustics grid model according to a three-dimensional geometrical model of a transformer; 2, conducting transient electromagnetic simulation on the structured grid model to obtain transient electromagnetic force waves borne by each phase winding and an iron core; 3, conducting FFT on the transient electromagnetic force waves borne by each phase winding and the iron core to obtain the frequency domain distribution of electromagnetic force borne by each phase winding and the iron core; 4, conducting harmonic response analysis on the structured grid model to obtain the vibration displacement data of nodes on the surfaces of the windings and the iron core; 5, conducting sound field calculation on the acoustics grid model to obtain the radiation noise distribution of the transformer. According to the transformer electromagnetic vibration noise calculating method, all noise influence factors are considered, and calculation results are accurate and reliable. The transformer electromagnetic vibration noise calculating method can be used for winding short circuit vibration deformation analysis and the like and can also provide method and basis for motor noise analysis.

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.

A computer-implemented method includes defining respective positions of a first set of nodes and a second set of nodes in an enrichment region, and performing a coupled pore fluid diffusion and stress analysis on the enrichment region at the first set of nodes. It is then determined whether the second set of nodes is activated—representing a fracture—as a result of the analysis, and the results are visually output to a user.

The invention relates to damage prediction for composite materials and is aimed at providing a finite element method for predicting progressive failure of composite laminates under low-speed impact. The method comprises following processes: a low-speed impact finite element model containing a hammer, composite laminates and support plates is established; a composite material elastoplasticity damage constitutive model is established; the elastoplasticity damage constitutive model is realized on the basis of ABAQUS-VUMAT user dynamic material subprogram module and by means of the backward Euler algorithm; the low-speed impact is calculated to further obtain impact force, displacement, speed and acceleration. According to the method, the ABAQUS-VUMAT user subprogram is adopted to numerically realize the elastoplasticity damage constitutive model combining plasticity and damage, wherein the model simultaneously considers the influence of plasticity and material performance degeneration; the method can accurately predict progressive damage failure of composite materials having plasticity characteristics under low-speed impact.

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 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 metal fatigue crack full-life estimation method based on a damage mechanics non-probabilistic interval analysis model. The method comprises the following steps: firstly, selecting a damage evolution equation from a damage mechanics model to establish a finite element analysis formation which structurally comprises damage information, and inquiring a fatigue test manual to fit to obtain a parameter in the damage evolution equation; then, combining with a damage mechanics finite element method with an interval finite element method to regard an initial damage degree and a damage parameter as interval uncertainty variables to represent fatigue life dispersivity; further, establishing the finite element analysis model of the structure, giving an initial critical element damage degree increment, constantly carrying out iteration increase on the damage degrees of all elements, judging that the element is damaged through the damage degree of each element, and endowing with intensity and rigidity attributes; and finally, when crack expansion achieves critical crack length, judging that the structure is damaged, and calculating to obtain a fatigue life range via a damage evolution equation variant and an interval vertex propagation analysis method. By use of the metal fatigue crack full-life estimation method, the life of the fatigue crack can be more elaborately estimated.

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 relates to the field of rock fracture prediction, in particular to a shale hydraulic fracture propagation prediction method. The method comprises the steps that 1, the normal direction,tangential stress and effect stress of an inclined fracture under the action of the external stress and water pressure are calculated; 2, a strain energy density function is obtained according to thetype of the fracture; 3, a strain energy density factor is obtained according to the strain energy density function; 4, the propagation direction and propagation angle of the fracture are judged according to a strain energy density criterion; 5, influences of stratification, natural fractures and the like on the propagation direction of hydraulic fractures are obtained through numerical simulation, so that fracture propagation under the effect of shale hydraulic pressure is predicted. The prediction method is based on fracture mechanics, a hydraulic condition factor is introduced, and the influences of different stratification directions on the propagation direction of the shale hydraulic fractures are obtained by researching the relation between the fracture propagation direction and themagnitude of the hydraulic pressure and utilizing an extended finite element method, so that accurate prediction of hydraulic fracture propagation is achieved, and higher prediction accuracy is obtained.

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 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 discloses a gear drive reliability assessment method based on a Kriging model and is applied to the field of reliability assessment. According to the gear drive reliability assessment method, the Kriging model is established firstly according to an initial sample, then whether variable coefficients meet the requirements or not is judged according to the established Kriging model, new variable coefficients are recalculated in a sample point adding mode. The stress calculation quantity based on a finite element method in a traditional method is greatly decreased. The gear drive reliability assessment method adopts a Monte Carlo method to calculate the structural reliability, and meanwhile combines with a learning function to obtain a high-precision agent model with fewer sample points, and accordingly testing times are decreased.

The invention relates to a simulation method for gas static pressure main shaft dynamic predication, and the method can be used for solving the problems of modeling difficulty, and difficulty in predicating the dynamic characteristics of the main shaft at the design stage due to low precision of a traditional method in the prior art. The method comprises the following steps of: partitioning a gas static pressure main shaft bearing gas film into finite element grids by utilizing a finite element method and a gas static pressure theory, and calculating to obtain the pressure distribution data We of a gas static pressure main shaft; establishing a finite element model, corresponding to the finite element grids, of a main shaft body; converting the pressure distribution data We of the finite element main shaft into equivalent spring stiffness values and assigning the equivalent spring stiffness values to nodes corresponding to the finite element grids of the gas film; and establishing a finite element model of the gas static pressure main shaft, and calculating the dynamic characteristics of the gas static pressure main shaft according to the established finite element model. The method is mainly used for predicating the dynamic characteristics of the gas static pressure main shaft.

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 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.