CN103345580A - Parallel CFD method based on lattice Boltzmann method - Google Patents
Parallel CFD method based on lattice Boltzmann method Download PDFInfo
- Publication number
- CN103345580A CN103345580A CN2013102745743A CN201310274574A CN103345580A CN 103345580 A CN103345580 A CN 103345580A CN 2013102745743 A CN2013102745743 A CN 2013102745743A CN 201310274574 A CN201310274574 A CN 201310274574A CN 103345580 A CN103345580 A CN 103345580A
- Authority
- CN
- China
- Prior art keywords
- parallel
- flow field
- max
- computing node
- lattice boltzmann
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a parallel CFD method based on the lattice Boltzmann method. The parallel CFD method based on the lattice Boltzmann method can conduct computation and analysis of hydromechanics in parallel for various complicated objects and is characterized in that a parallel distributed mesh generation method which is suitable for the lattice Boltzmann method is introduced, the time of mesh generation is greatly shortened, an algorithm which reduces the communication amount of the lattice Boltzmann method in the process of parallel iterative computation is established according to a distributed computer architecture, the iterative speed is improved, the expandability of the algorithm is improved, and the lattice Boltzmann method is made to be applicable to large-scale computation. A large number of numerical experiments show that the parallel computation method has good expandability and is more applicable to an ultra-large-scale computation system.
Description
Technical field
The present invention relates to Fluid Mechanics Computation and computer realm, proposed a kind of parallel C FD method based on lattice Boltzmann method.
Background technology
Through the development of decades, Fluid Mechanics Computation CFD is in Aero-Space, large-scale energy source device (as nuclear power station), and new traffic tool, numerous engineering department such as environmental protection and field have all obtained using widely.At present, Fluid Mechanics Computation oneself become the modern computing science and the most effectively one of promote.Make us advising the purpose progress although CFD has obtained, because the complicacy of fluid motion and the restriction of computer resource, still there are many limitation in CFD.Calculating, prediction and control to some problem are not sure yet.For example, owing to lack the restriction of effective computation model and computing machine condition, may reach 100% sometimes to the error of the calculating of spacecraft heat problem, many disastrous weather can't compare accurate forecast.Finding the solution these complicated flow field problems at unit often needs long computing time, some extensive problem even can't find the solution, and this has limited the application of CFD in scientific research and engineering problem to a certain extent.This shows, must adopt computation model and advanced Computer Architecture efficiently in order to find the solution complicated CFD problem, so, the development of engineering and problem in science need be to researching and proposing of CFD more urgent requirement, fluid calculates becomes a challenging field.
Effective utilization of massively parallel computer needs new parallel algorithm.Even because utilize classic method to find the solution some engineering and problem in science at supercomputer, as direct Numerical of turbulent flow etc., still need the time of growing very much.Traditional method of finding the solution the Navier-Stokes equation need be found the solution large-scale Algebraic Equation set, and the explicit solution concurrency of Navier-Stokes equation is better, but numerical stability is poor, and speed of convergence is slow; Implicit expression solution better numerical value stability, fast convergence rate, but its parallel extensibility is relatively poor.
Summary of the invention
The objective of the invention is to, in order to address the above problem, a kind of parallel C FD method based on lattice Boltzmann method is provided, provide a kind of novel complete parallel highly effective algorithm that time, the full lattice Boltzmann method that disperses in space and parallel computation adapt, this method is moving theoretical based on molecule, has physical background clearly.This method is discrete method in macroscopic view, is continuation method at microcosmic, thereby is called as the sight analogy method that is situated between.This method has many special advantages, as counting yield height, advantage such as boundary condition treatment is simple, program is easy to implement.This method is a kind of discrete statistical model, the information transmission is local in the computation process, therefore the concurrent operation program degree of association is less, need the quantity of information transmitted less between each computing node, be particularly suitable for carrying out parallel computation at the large-scale computer with distributed architecture, can under the smaller situation of expense, carry out fairly large numerical simulation, the particularly arrival in cloud epoch, resource obtains to be more prone to, and the easier expansion of more massive parallel computation comes.
For achieving the above object, design of the present invention is: at first according to the parameter request of data and the actual computation of veil lattice, the exploitation distributed earth carries out the cartesian grid subdivision method, accelerates mesh generation; Simultaneously in conjunction with certain data structure, the initialization of acceleration grid, the parallel lattice Boltzmann method that has an enhanced scalability of the less traffic according to lattice Boltzmann method exploitation is carried out iterative computation then, output flow field and object plane information are carried out visual analyzing afterwards after satisfying the accuracy requirement of calculating.
According to the foregoing invention design, at the needs of actual computation problem, Flow Field Calculation scope, object plane information, the present invention adopts following technical proposals:
A, according to the computation requirement of practical problems, read in the parameters calculated file;
B, according to the path of the veil lattice stl file in the Parameter File, reading face grid file;
C, according to the data of the flow field scope in the Parameter File and veil lattice, utilize the distributed cartesian grid subdivision method of exploitation, concurrently the distributed subdivision that carries out cartesian grid;
D, according to the calculation requirement in the Parameter File, the cartesian grid of distributed parallel ground after with subdivision carries out initialization;
The parallel lattice Boltzmann method of the enhanced scalability of E, utilization exploitation is carried out parallel iteration and is calculated the accuracy requirement in satisfying Parameter File;
Flow field information and object plane information in F, the output result of calculation;
G, Visualization of calculation analysis.
Parallel C FD method based on lattice Boltzmann method of the present invention compared with prior art, has following outstanding substantive distinguishing features and remarkable advantage:
1. this method has been set up a kind of subdivision method of distributed grid fast, has accelerated the speed of mesh generation.Because grid is the distributed subdivision that carries out, carrying out between each node of subdivision is not need to carry out communication, is complete parallel schema, and the speed-up ratio of near-linear is arranged.
Since behind the mesh generation be distributed storage on each computing node, so can distributedly carry out the grid computing information initializing, do not need to communicate, accelerated initialization time, also have simultaneously the speed-up ratio of near-linear.
3. this method is simultaneously according to lattice Boltzmann method, provided a kind of high extendible parallel lattice Boltzmann method that has, carried out iterative computation post, shortened the parallel computation time of original method, this method also is with good expansibility simultaneously.
Description of drawings
Fig. 1 is the process flow diagram of the parallel C FD method based on lattice Boltzmann method of the present invention.
Fig. 2 be among Fig. 1 among the step C distributed parallel handle the particular flow sheet of cartesian grid subdivision.
Fig. 3 is that the described flow field of step D information is carried out initialized process flow diagram among Fig. 1.
Fig. 4 is step e and the described process flow diagram of F among Fig. 1.
Fig. 5 is the model velocity structural drawing of D3Q19 in the lattice Boltzmann method.
Fig. 6 is communication and the hairdo of two-dimensional lattice Boltzmann method.
Fig. 7 is the communication sequence formula of two-dimensional lattice Boltzmann method.
Fig. 8 is the test result figure of scale and efficient.
Fig. 9 is the speed-up ratio figure of three kinds of scale test problems.
Figure 10 is the efficiency diagram of three kinds of scale test problems.
Embodiment
Below in conjunction with accompanying drawing the preferred embodiments of the present invention are described in detail.
Embodiment one:
Referring to Fig. 1, this is based on the parallel C FD method of lattice Boltzmann method, provide a kind of novel complete parallel highly effective algorithm that time, the full lattice Boltzmann method that disperses in space and parallel computation adapt, it is characterized in that: at given real fluid computational problem, can adopt this method efficiently to simulate, the concrete operations step is as follows:
A, according to the needs of actual computation problem, given calculating parameter also saves as ini format parameter file, reads this Parameter File; Concrete steps comprise:
A1, host process are read in Parameter File;
A2, host process send to the Parameter File content of reading in all the other each processes;
B, according to the veil lattice stl file storage location that provides in the steps A, read in this document; Concrete steps comprise:
B1, host process are read in veil lattice stl file;
B2, host process send to the veil lattice data of reading in all the other each processes;
C, according to the content among steps A and the step C, distributed parallel carries out mesh generation;
D, according to the calculation requirement in the Parameter File, each computing node carries out initialization with its cartesian grid of being responsible for;
The parallel lattice Boltzmann method of the enhanced scalability of E, utilization exploitation is carried out parallel iteration calculating, till the accuracy requirement in satisfying Parameter File;
F, output result calculated; Concrete steps are as follows:
F1, host computer node output object plane result and flow field result, result's output format is the Tecplot/VTK file layout;
F2, all the other computing nodes are input to the back as a result that host node is exported with object plane and Flow Field Calculation result in order, form complete object plane and Flow Field Calculation result;
G, object plane and Flow Field Calculation result's is visual: mainly adopt Tecplot business software or ParaView open source software.
Embodiment two:
In the present embodiment, this is based on the experiment of the parallel C FD method of the lattice Boltzmann method magic square 5000A in Shanghai supercomputer center, magic square 5000A is based on the large computer system of cluster concept design, and its overall calculation capability theory peak value is 200T flops (1Tflops is per second 1012 Floating-point Computation).
At the needs of actual computation problem, Flow Field Calculation scope, object plane information, the parallel C FD method based on lattice Boltzmann method of the present invention as Fig. 1-shown in Figure 7, may further comprise the steps:
H, according to the needs of actual computation problem, given calculating parameter also saves as ini format parameter file, reads this Parameter File.Concrete steps comprise:
A1, host process are read in Parameter File;
A2, host process send to the Parameter File content of reading in all the other each processes.
I, according to the veil lattice stl file storage location that provides in the steps A, read in this document.Concrete steps comprise:
B1, host process are read in veil lattice stl file;
B2, host process send to the veil lattice data of reading in all the other each processes.
J, according to the content among steps A and the step C, distributed parallel carries out mesh generation.Concrete steps comprise:
C1, according to the information of the calculating parameter in the Parameter File and veil lattice, the mesh generation zone that each computing node is responsible for is divided.Concrete steps comprise:
C11, according to computing node
nWith the flow field regions size, with flow field regions (
x,
y,
zThree directions) evenly be divided into
nEqual portions, each computing node portion wherein, namely the responsible scope of each process is:
x∈[
min_x i ,?
max_x i ]、
y∈[
min_y i ,?
max_y i ]、
z∈[
min_z i ,?
max_z i ];
In the formula
Min_x i Represent flow field regions
xThe minimum value of direction,
Max_x i Represent flow field regions
xThe maximal value of direction,
Min_y i Represent flow field regions
yThe minimum value of direction,
Max_y i Represent flow field regions
yThe maximal value of direction,
Min_z i Represent flow field regions
zThe minimum value of direction,
Max_z i Represent flow field regions
zThe maximal value of direction;
C12, according to the size of mesh opening in the Parameter File that reads in the steps A
Size, portion of in each computing node it being responsible for respectively to
x,
y,
zThree directions all increase and reduce one
Size, namely the responsible scope of each process is:
x∈[
min_x i -
size,?
max_x i +
size]、
y∈[
min_y i -
size,?
max_y i +
size]、
z∈[
min_z i -
size,?
max_z i +
size];
C2, each computing node carry out mesh generation to the scope of being responsible for separately, judge the type of each net point simultaneously.Concrete steps are:
C21, generate initial cartesian grid;
C22, judge the situation that each net point and object plane intersect;
In C3, each computing node the net point of solid interior (being that object plane surrounds formation) is deleted.
K, according to the calculation requirement in the Parameter File, each computing node carries out initialization with its cartesian grid of being responsible for.Concrete steps comprise:
D1, according to the calculating parameter in the Parameter File, different trellis-types is carried out the assignment of flow field macroscopic information;
D2, calculate the microscopic information amount according to the equilibrium state distribution function of the D3Q19 rate pattern of the macroscopic information of assignment and lattice Boltzmann method, concrete computing formula is seen as follows:
In the formula
Represent
The equilibrium state distribution function value of individual discrete velocity reversal,
Represent density,
Representation speed,
Represent
Individual discrete speed (see figure 5) is:
Weight function
For:
The velocity of sound:
, wherein
cBe grid speed.
The parallel lattice Boltzmann method of the enhanced scalability of L, utilization exploitation is carried out parallel iteration calculating, till the accuracy requirement in satisfying Parameter File.Concrete steps are as follows:
E1, iterations variable Iteration if reach iterations in the Parameter File, then withdraw from iteration, otherwise proceed iteration;
E2, calculate the error of macroscopic quantity, judge according to the accuracy requirement in the Parameter File, if satisfy then withdraw from, otherwise continue;
E21, according to microscopic quantity and following formula, the new macroscopic quantity on the computing grid point:
Take absolute value after original macroscopic quantity is subtracted each other on E22, the new macroscopic quantity that will calculate and the net point and obtain the error of macroscopic quantity;
Whether E23, error in judgement satisfy accuracy requirement, if satisfy then withdraw from, otherwise continue;
The Parallel Collision transition process of E3, lattice Boltzmann method, main computing formula is seen as follows:
In the formula
Represent relaxation factor,
The representative time,
Represent the time interval, this formula is divided into two parts: migration and collision, and concrete steps are as follows:
E31, mobile grid (transition process) in the transition process of the net point that each computing node is responsible for, all need to use the gridding information of neighborhood calculation node, and there are following two kinds of operation stepss in the traffic operation that this just need carry out between computing node:
E311, according to the communication under the two-dimensional case among Fig. 6 and hairdo, can be generalized under the three-dimensional situation easily.Each computing node communicates according to similar burse mode earlier
E312 or according to the communication sequence formula under the two-dimensional case among Fig. 7 also can be generalized under the three-dimensional situation easily.Each computing node communicates according to similar ordered mode earlier
After E32, communication were finished, each computing node can carry out impact operations, and this step do not need to communicate, and was complete parallelization, but collision is only meeting
x∈ [
Min_x i ,
Max_x i ],
y∈ [
Min_y i ,
Max_y i ],
z∈ [
Min_z i ,
Max_z i ] carry out on the net point in the scope
The E1 step is continued to carry out in E33, iteration variable Iteration=Iteration+1.
M, output result calculated.Concrete steps are as follows:
F1, host computer node output object plane result and flow field result, result's output format is the Tecplot/VTK file layout;
F2, all the other computing nodes are input to the back as a result that host node is exported with object plane and Flow Field Calculation result in order, form complete object plane and Flow Field Calculation result;
N, object plane and Flow Field Calculation result's is visual.Main Tecplot business software or the ParaView open source software of adopting.
Following Fig. 8-10 is primarily aimed at the calculating test result of parallel C FD method on the super calculation in Shanghai dawn 5000A based on lattice Boltzmann method of the present invention and describes.
With reference to Fig. 8, provided the parallel C FD method based on lattice Boltzmann method of the present invention at the efficiency diagram under the various computing scale problem.Illustrated that the inventive method can be applicable to the processing of extensive problem, and can keep good efficiency.
Fig. 9 has provided at the parallel speed-up ratio under three kinds of different scales and the comparison diagram of ideal value, and as can be seen from the figure, method of the present invention is with good expansibility.
Figure 10 provided with Fig. 9 in parallel efficiency figure under same three kinds of scales, as can be seen from the figure, method parallel efficiency of the present invention has still been kept the efficient more than 80% on 4000 nuclear magnitudes.
In sum, Fig. 8~Figure 10 shows, parallel C FD method based on lattice Boltzmann method of the present invention, can calculate at extensive problem, and the extensibility of this method is good, efficient can maintain higher level, and this method can be brought into play the effectiveness of ultra-large supercomputer well simultaneously, exchanges computing time for computational resource.This parallel C FD method is effectively replenishing of traditional C FD method, and implement comparatively convenient.
This paper sets forth just for helping to understand method of the present invention and core concept in conjunction with Figure of description and specific embodiment.Method of the present invention is not limited to the embodiment described in the embodiment, and those skilled in the art belong to technological innovation scope of the present invention equally according to other embodiment that method of the present invention and thought draw.This description should not be construed as limitation of the present invention.
Claims (5)
1. based on the parallel C FD method of lattice Boltzmann method, provide a kind of novel complete parallel highly effective algorithm that time, the full lattice Boltzmann method that disperses in space and parallel computation adapt, it is characterized in that: at given real fluid computational problem, can adopt this method efficiently to simulate, the concrete operations step is as follows:
According to the needs of actual computation problem, given calculating parameter also saves as ini format parameter file, reads this Parameter File; Concrete steps comprise:
A1, host process are read in Parameter File;
A2, host process send to the Parameter File content of reading in all the other each processes;
Veil lattice stl file storage location according to providing in the steps A reads in this document; Concrete steps comprise:
B1, host process are read in veil lattice stl file;
B2, host process send to the veil lattice data of reading in all the other each processes;
According to the content among steps A and the step C, distributed parallel carries out mesh generation;
According to the calculation requirement in the Parameter File, each computing node carries out initialization with its cartesian grid of being responsible for;
Utilize the parallel lattice Boltzmann method of the enhanced scalability of exploitation to carry out parallel iteration calculating, till the accuracy requirement in satisfying Parameter File;
The output result calculated; Concrete steps are as follows:
F1, host computer node output object plane result and flow field result, result's output format is the Tecplot/VTK file layout;
F2, all the other computing nodes are input to the back as a result that host node is exported with object plane and Flow Field Calculation result in order, form complete object plane and Flow Field Calculation result;
Object plane and Flow Field Calculation result's is visual: mainly adopt Tecplot business software or ParaView open source software.
2. the parallel C FD method based on lattice Boltzmann method according to claim 1, it is characterized in that: distributed must the walking abreast of described step C carried out mesh generation, and concrete steps are as follows:
C1, according to the information of the calculating parameter in the Parameter File and veil lattice, the mesh generation zone that each computing node is responsible for is divided; Concrete steps comprise:
C11, according to computing node
nWith the flow field regions size, with flow field regions
x,
y,
zThree directions evenly are divided into
nEqual portions, each computing node portion wherein, namely the responsible scope of each process is:
x∈[
min_x i ,?
max_x i ]、
y∈[
min_y i ,?
max_y i ]、
z∈[
min_z i ,?
max_z i ];
In the formula
Min_x i Represent flow field regions
xThe minimum value of direction,
Max_x i Represent flow field regions
xThe maximal value of direction,
Min_y i Represent flow field regions
yThe minimum value of direction,
Max_y i Represent flow field regions
yThe maximal value of direction,
Min_z i Represent flow field regions
zThe minimum value of direction,
Max_z i Represent flow field regions
zThe maximal value of direction;
C12, according to the size of mesh opening in the Parameter File that reads in the steps A
Size, in each computing node its portion of being responsible for is increased and reduces one to x, y, three directions of z respectively
Size, namely the responsible scope of each process is:
x∈[
min_x i -
size,?
max_x i +
size]、
y∈[
min_y i -
size,?
max_y i +
size]、
z∈[
min_z i -
size,?
max_z i +
size];
C2, each computing node carry out mesh generation to the scope of being responsible for separately, judge the type of each net point simultaneously; Concrete steps are:
C21, generate initial cartesian grid;
C22, judge the situation that each net point and object plane intersect;
Be that the net point that object plane surround to form is deleted to solid interior in C3, each computing node.
3. the parallel C FD method based on lattice Boltzmann method according to claim 1, it is characterized in that: the cartesian grid among the described step D carries out initialization, and its concrete grammar is:
D1, according to the calculating parameter in the Parameter File, different trellis-types is carried out the assignment of flow field macroscopic information;
D2, calculate the microscopic information amount according to the equilibrium state distribution function of the D3Q19 rate pattern of the macroscopic information of assignment and lattice Boltzmann method, concrete computing formula is seen as follows:
In the formula
Represent
The equilibrium state distribution function value of individual discrete velocity reversal,
Represent density,
Representation speed,
Represent
Individual discrete speed (see figure 5) is:
4. the parallel C FD method based on lattice Boltzmann method according to claim 1, it is characterized in that: in the described step e, utilize lattice Boltzmann method to carry out parallel iteration and calculate, distributed parallel carries out the numerical evaluation of fluid, and concrete steps are as follows:
E1, iterations variable Iteration if reach iterations in the Parameter File, then withdraw from iteration, otherwise proceed iteration;
E2, calculate the error of macroscopic quantity, judge according to the accuracy requirement in the Parameter File, if satisfy then withdraw from, otherwise continue;
E21, according to microscopic quantity and following formula, the new macroscopic quantity on the computing grid point:
Take absolute value after original macroscopic quantity is subtracted each other on E22, the new macroscopic quantity that will calculate and the net point and obtain the error of macroscopic quantity;
Whether E23, error in judgement satisfy accuracy requirement, if satisfy then withdraw from, otherwise continue;
The Parallel Collision transition process of E3, lattice Boltzmann method, main computing formula is seen as follows:
In the formula
Represent relaxation factor,
The representative time,
Represent the time interval, this formula is divided into two parts: migration and collision, and concrete steps are as follows:
E31, mobile grid, transition process in the transition process of the net point that each computing node is responsible for, all needs to use the gridding information of neighborhood calculation node, and the traffic operation that this just need carry out between computing node can adopt sequential and and hairdo communication pattern
After E32, communication were finished, each computing node can carry out impact operations, and this step do not need to communicate, and was complete parallelization, but collision is only meeting
x∈ [
Min_x i ,
Max_x i ],
y∈ [
Min_y i ,
Max_y i ],
z∈ [
Min_z i ,
Max_z i ] carry out on the net point in the scope
The E1 step is continued to carry out in E33, iteration variable Iteration=Iteration+1.
5. the parallel C FD method based on lattice Boltzmann method according to claim 4 is characterized in that: in the described step e 31, sequential and and the hairdo communication pattern, concrete grammar is as follows:
E311, communication and hairdo, each computing node sends to the computing node that needs separately with the second layer grid in the claim 2 simultaneously, and this communication mode needs cushion space, so be applicable to the cluster of middle and small scale;
E312, communication sequence formula, each computing node is in a certain order, and is point-to-point, and data are delivered to corresponding computing node, and this communication mode does not need cushion space, and applicability is good, is applicable to large-scale cluster.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310274574.3A CN103345580B (en) | 2013-07-02 | 2013-07-02 | Based on the parallel CFD method of lattice Boltzmann method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310274574.3A CN103345580B (en) | 2013-07-02 | 2013-07-02 | Based on the parallel CFD method of lattice Boltzmann method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103345580A true CN103345580A (en) | 2013-10-09 |
CN103345580B CN103345580B (en) | 2016-04-27 |
Family
ID=49280375
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310274574.3A Expired - Fee Related CN103345580B (en) | 2013-07-02 | 2013-07-02 | Based on the parallel CFD method of lattice Boltzmann method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103345580B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104462829A (en) * | 2014-12-11 | 2015-03-25 | 中国航天空气动力技术研究院 | Handling method of complex multi-region grids in space marching solution |
CN104573151A (en) * | 2013-10-22 | 2015-04-29 | 北京临近空间飞行器***工程研究所 | Cyclic communication method for acquiring wall grid point data |
CN104573296A (en) * | 2013-10-22 | 2015-04-29 | 北京临近空间飞行器***工程研究所 | Hypersonic-speed flow field initializing method oriented to similar grids |
CN104933225A (en) * | 2015-05-25 | 2015-09-23 | 中国科学院过程工程研究所 | Method for realizing computational fluid dynamics large-scale real-time simulation |
CN105760602A (en) * | 2015-12-30 | 2016-07-13 | 南京航空航天大学 | Total flow field numerical simulation method for finite volume weighted essentially non-oscillatory scheme |
CN107515987A (en) * | 2017-08-25 | 2017-12-26 | 中国地质大学(北京) | The simulation accelerated method of Groundwater Flow based on more relaxation Lattice Boltzmann models |
CN111105341A (en) * | 2019-12-16 | 2020-05-05 | 上海大学 | Framework method for solving computational fluid dynamics with low power consumption and high operational performance |
CN111489447A (en) * | 2020-04-14 | 2020-08-04 | 西北工业大学 | Right-angle grid adaptive modeling method suitable for lattice Boltzmann method |
CN111538487A (en) * | 2020-04-17 | 2020-08-14 | 中国空气动力研究与发展中心计算空气动力研究所 | Distributed parallel grid generation software framework |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102681972A (en) * | 2012-04-28 | 2012-09-19 | 浪潮电子信息产业股份有限公司 | Method for accelerating lattice-Boltzmann by utilizing graphic processing units (GPUs) |
CN102945295A (en) * | 2012-10-15 | 2013-02-27 | 浪潮(北京)电子信息产业有限公司 | Parallel acceleration method and system of lattice Boltzmann method |
CN103064819A (en) * | 2012-10-25 | 2013-04-24 | 浪潮电子信息产业股份有限公司 | Method for utilizing microwave integrated circuit (MIC) to rapidly achieve lattice Boltzmann parallel acceleration |
-
2013
- 2013-07-02 CN CN201310274574.3A patent/CN103345580B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102681972A (en) * | 2012-04-28 | 2012-09-19 | 浪潮电子信息产业股份有限公司 | Method for accelerating lattice-Boltzmann by utilizing graphic processing units (GPUs) |
CN102945295A (en) * | 2012-10-15 | 2013-02-27 | 浪潮(北京)电子信息产业有限公司 | Parallel acceleration method and system of lattice Boltzmann method |
CN103064819A (en) * | 2012-10-25 | 2013-04-24 | 浪潮电子信息产业股份有限公司 | Method for utilizing microwave integrated circuit (MIC) to rapidly achieve lattice Boltzmann parallel acceleration |
Non-Patent Citations (1)
Title |
---|
宋能: "《球面空间网格的分布式并行技术研究与实现》", 《中国优秀硕士学位论文全文数据库-信息科技辑》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104573151A (en) * | 2013-10-22 | 2015-04-29 | 北京临近空间飞行器***工程研究所 | Cyclic communication method for acquiring wall grid point data |
CN104573296A (en) * | 2013-10-22 | 2015-04-29 | 北京临近空间飞行器***工程研究所 | Hypersonic-speed flow field initializing method oriented to similar grids |
CN104573296B (en) * | 2013-10-22 | 2017-09-22 | 北京临近空间飞行器***工程研究所 | A kind of Hypersonic Flow Field initial method towards similar grid |
CN104462829B (en) * | 2014-12-11 | 2019-01-18 | 中国航天空气动力技术研究院 | The processing method of complicated multi-region grid in space propultion solution |
CN104462829A (en) * | 2014-12-11 | 2015-03-25 | 中国航天空气动力技术研究院 | Handling method of complex multi-region grids in space marching solution |
CN104933225A (en) * | 2015-05-25 | 2015-09-23 | 中国科学院过程工程研究所 | Method for realizing computational fluid dynamics large-scale real-time simulation |
CN104933225B (en) * | 2015-05-25 | 2018-01-09 | 中国科学院过程工程研究所 | The method for realizing the extensive real-time Simulation of Fluid Mechanics Computation |
CN105760602A (en) * | 2015-12-30 | 2016-07-13 | 南京航空航天大学 | Total flow field numerical simulation method for finite volume weighted essentially non-oscillatory scheme |
CN107515987A (en) * | 2017-08-25 | 2017-12-26 | 中国地质大学(北京) | The simulation accelerated method of Groundwater Flow based on more relaxation Lattice Boltzmann models |
CN111105341A (en) * | 2019-12-16 | 2020-05-05 | 上海大学 | Framework method for solving computational fluid dynamics with low power consumption and high operational performance |
CN111489447A (en) * | 2020-04-14 | 2020-08-04 | 西北工业大学 | Right-angle grid adaptive modeling method suitable for lattice Boltzmann method |
CN111489447B (en) * | 2020-04-14 | 2022-04-29 | 西北工业大学 | Right-angle grid adaptive modeling method suitable for lattice Boltzmann method |
CN111538487A (en) * | 2020-04-17 | 2020-08-14 | 中国空气动力研究与发展中心计算空气动力研究所 | Distributed parallel grid generation software framework |
Also Published As
Publication number | Publication date |
---|---|
CN103345580B (en) | 2016-04-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103345580B (en) | Based on the parallel CFD method of lattice Boltzmann method | |
US10007742B2 (en) | Particle flow simulation system and method | |
Kampolis et al. | CFD-based analysis and two-level aerodynamic optimization on graphics processing units | |
CN103970960B (en) | The element-free Galerkin structural topological optimization method accelerated parallel based on GPU | |
Griebel et al. | A multi-GPU accelerated solver for the three-dimensional two-phase incompressible Navier-Stokes equations | |
JP6605610B2 (en) | Semiconductor device | |
Zabelok et al. | Adaptive kinetic-fluid solvers for heterogeneous computing architectures | |
CN105302974A (en) | Real-time cutting simulation method of flexible object on the basis of finite element and time-variant modal analysis | |
CN103617150A (en) | GPU (graphic processing unit) based parallel power flow calculation system and method for large-scale power system | |
Groth et al. | Parallel implicit adaptive mesh refinement scheme for body-fitted multi-block mesh | |
Käppeli et al. | FISH: a three-dimensional parallel magnetohydrodynamics code for astrophysical applications | |
CN107609141A (en) | It is a kind of that quick modelling method of probabilistic is carried out to extensive renewable energy source data | |
CN107403466A (en) | Ultra-large unstrctured grid generation method based on overall situation encryption | |
CN106446432A (en) | Optimal transportation meshless method for solving large deformation of material | |
Su et al. | Large-scale simulations on multiple Graphics Processing Units (GPUs) for the direct simulation Monte Carlo method | |
Gan et al. | A highly-efficient and green data flow engine for solving euler atmospheric equations | |
Cai et al. | A high performance crashworthiness simulation system based on GPU | |
CN112632874A (en) | Optimization method and system for numerical simulation of helicopter flow field | |
CN102902590A (en) | Parallel digital terrain analysis-oriented massive DEM (Digital Elevation Model) deploying and scheduling method | |
Kono et al. | Evaluations of OpenCL-written tsunami simulation on FPGA and comparison with GPU implementation | |
CN104050148A (en) | FFT accelerator | |
CN106846457B (en) | Octree parallel construction method for CT slice data visual reconstruction | |
Marshall et al. | Performance evaluation and enhancements of a flood simulator application for heterogeneous hpc environments | |
Trenti et al. | Gravitational N-body simulations | |
Playne et al. | Benchmarking multi-GPU communication using the shallow water equations |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160427 Termination date: 20190702 |
|
CF01 | Termination of patent right due to non-payment of annual fee |