31results about How to "Reduce computing memory" patented technology

The invention provides a fast numerical simulation method of two-dimensional magnetotelluric field, which combines the finite element method and Fourier transform, and converts the two-dimensional partial differential problem into different wave numbers by performing Fourier transform along the horizontal direction One-dimensional ordinary differential problems that are independent of each other, with high parallelism, effectively reduce the calculation amount and improve the calculation efficiency; provide conditions for the refined numerical simulation of the magnetotelluric method under large-scale conditions; after verification, adopt the present invention The calculation result of the method is within the range of the error bars of the international public data included in the literature of Zhdanov et al.

The present invention provides a method and system for extracting azimuth gathers based on offset vector slices, comprising the following steps: dividing seismic data into several offset vector slices; calculating the ground azimuth of each offset vector slice; Perform three-dimensional Kirchhoff integral pre-stack time migration on the acquired seismic data to calculate the subsurface scattering azimuth and reflection angle; retain the subsurface scattering azimuth consistent with the ground azimuth angle; select several The x-axis, the x-axis is used to represent the angular change of the emission angle, the normal direction of the plane is the z-axis, and the z-axis is used to represent the change of the formation depth, and the obtained reflection angle is marked in the x-z coordinate system to obtain the reflection angle distribution map ; Obtain the azimuth angle gather according to the reflection angle distribution map. It effectively reduces the variation range of the scattering angle γ and the scattering azimuth angle, can greatly save memory and hard disk resources, and shorten the calculation cycle.

The invention relates to a method and system for self-adaptive subdivision of a polygonal unstructured grid of an integrated circuit layout. The method includes sequentially inserting eccentric points into the existing mesh according to the lost and eroded polygon edges to recover the lost and eliminated eroded polygon edges, determining the triangles to be subdivided according to the set ratio / minimum angle threshold, in order Insert the center of the eccentric circle of the triangle to be subdivided into the existing mesh to eliminate the triangle to be subdivided. The above restores the lost and eliminated the eroded polygon edges and eliminates the triangle to be subdivided. Interleaved until the integrated circuit layout polygonal unstructured network There are no missing and eroded polygon edges and triangles to be subdivided in the grid, and finally form a new unstructured grid of VLSI layout to solve the redundancy caused by newly inserted nodes in the prior art While solving the problem of excessively dense local grids, it improves grid division accuracy and numerical calculation efficiency, and reduces calculation memory.

A three-dimensional quasi-transport acceleration method for the uniform geometric variational nodal method, which eliminates the axial and radial cross-derivative terms in the second-order even-parity neutron transport equation through quasi-transport approximation; considering the In the actual pressurized water reactor, the axial inhomogeneity is relatively weak, and the present invention adopts diffusion approximation to the angular distribution of the odd neutron angular flux density on the axial surface in the Ritz discrete process; , taking advantage of the symmetry of the angular space, reduces the number of angular basis functions for the odd neutron angular flux density on the radial surface. The invention can greatly improve the calculation efficiency and reduce the calculation memory without significantly affecting the calculation accuracy, so as to be used for accurate and efficient neutronics simulation in nuclear reactor design.

The application provides a method for coordinating distributed power flow optimization of the main distribution network and related equipment, wherein the method of the application includes: according to the topological relationship of the distribution network system model, the distribution network system model is divided into a main network model and a distribution network model Model; according to the preset power flow optimization target threshold, the main distribution coordination power flow optimization model is established, and the main network ADMM iterative optimization model and the distribution network ADMM iterative optimization model are respectively established through the alternate direction multiplier calculation method according to the main distribution coordination power flow optimization model ; The main network ADMM iterative optimization model and the distribution network ADMM iterative optimization model are iteratively calculated to obtain the current main network optimization parameters and distribution network optimization parameters. This application is based on the target-synchronized alternating direction multiplier method of the master-coordinated distributed power flow optimization method, which does not require a unified upper-level optimization center, and solves the calculation and communication bottleneck problems of the existing centralized optimization degree method with a distributed calculation mode , which improves the power flow optimization efficiency of the main distribution coordination.

The invention relates to a large-scale integrated circuit layout unstructured grid eccentric midpoint generation method and system. According to the large-scale integrated circuit layout unstructuredgrid eccentric midpoint generation method and system, lost edges are recovered by an eccentric midpoint recovery method, and eroded edges are eliminated by an eccentric midpoint elimination method. Inthe recovery process of the lost edge, the elimination of the eroded edge is nested, and in the elimination process of the eroded edge, the recovery of the lost edge is nested, so that it is can be ensured that all the lost edges and the eroded edges are eliminated. Moreover, an included angle formed by eccentric midpoints generated for the lost edges and the eroded edges is just a preset triangle minimum angle threshold value for defining the quality of the triangle, so that grid nodes are reduced while the quality of the generated grid meets the requirements, and the calculation memory is reduced while the speed of numerical calculation is increased.

The invention relates to a method and device for compressing nuclear magnetic resonance echo data, wherein the compression method comprises: acquiring nuclear magnetic resonance echo data to be analyzed; using the nuclear magnetic resonance echo data to obtain principal components of the nuclear magnetic resonance echo data; A compression matrix is ​​constructed by using the principal components of the nuclear magnetic resonance echo data; and the nuclear magnetic resonance echo data is compressed by using the compression matrix. This technical solution can effectively eliminate the information that has little contribution to the characterization of formation characteristics or even redundant information in the original echo data, can still ensure the characteristics of the original data under the condition of high compression ratio, and can effectively prevent the loss of accuracy. Improve inversion speed and reduce computational memory.