Parallel method suitable for Lagrange particle class numerical simulation

Abstract

The invention is suitable for the technical field of fluid mechanics, and provides a parallel method suitable for Lagrangian type particle class numerical simulation, which comprises the following steps: obtaining particles in a calculation area, dividing the calculation area into a plurality of sub-calculation areas, the number of the particles in each sub-calculation area being the same; the thread corresponding to each sub-calculation area obtains particle information, calculation parameter information and thread attribute information in the current sub-calculation area; setting a connection area of each sub-calculation area; obtaining particle information in a junction area of a sub-calculation area adjacent to each sub-calculation area at the current moment, and updating the particle information in each sub-calculation area at the next moment; when the particles in each sub-calculation area at the next moment cross the boundary of each current sub-calculation area, the particles crossing the boundary are sent to the corresponding sub-calculation areas, and the method improves the calculation efficiency and memory supply quantity of a particle numerical simulation algorithm.

Description

technical field [0001] The invention relates to the technical field of fluid mechanics, in particular to a parallel method suitable for numerical simulation of Lagrangian particles. Background technique [0002] For scientific research in science and engineering, there are mainly scientific research methods such as theory, experiment and numerical simulation. Difficulties encountered in theoretical scientific research methods mainly include difficulties in establishing theoretical models and difficulties in theoretical verification. Difficulties encountered in experimental scientific research methods mainly include difficulty in technical implementation and high cost. Numerical simulation has the advantages of high efficiency, low cost, and comprehensive data acquisition by executing mathematical equations and formulas for specific problems on the computer to obtain calculation results and conduct corresponding analysis. At present, the main numerical simulation methods can...

AI Technical Summary

Problems solved by technology

Taking the smooth particle hydrodynamics method as an example, it needs to find the adjacent particles around each particle in each calculation step, so as to realize the particle pairing calculation, which leads to its low calculation efficiency. Currently, the serial Calculations can only calculate two-dimensional problems or simple three-dimensional problems. For millions of particles, due to single-thread memory limitations, it may even be impossible to calculate or the calculation time is too long to be acceptable.

Recently, a more parallel algorithm based on particle decomposition is used. It is a multi-threaded concurrent programming framework that supports cross-platform shared memory. Threads share memory space, which is greatly restricted by hardware, and it is mainly aimed at program loop codes, so it cannot handle large-scale computing problems. Generally speaking, problems with tens of millions of particles are difficult to deal with.

Images