東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Multi-scale simulation of micro/nano...

The Johns Hopkins University.

FindBook

Google Book

Amazon

博客來

Multi-scale simulation of micro/nano flows and granular flows.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Multi-scale simulation of micro/nano flows and granular flows./
作者:	Liu, Jin.
面頁冊數:	146 p.
附註:	Adviser: Shiyi Chen.
Contained By:	Dissertation Abstracts International69-12B.
標題:	Engineering, Mechanical. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3339874
ISBN:	9780549938477

Multi-scale simulation of micro/nano flows and granular flows.
Liu, Jin.

Multi-scale simulation of micro/nano flows and granular flows. - 146 p.

Adviser: Shiyi Chen.

Thesis (Ph.D.)--The Johns Hopkins University, 2009.

In this thesis we first develop a hybrid atomistic-continuum scheme for simulating micro- and nano-flows with heat transfer. The approach is based on spatial "domain decomposition" in which molecular dynamics (MD) is used in regions where atomistic details are important, while classical continuum fluid dynamics is used in the remaining bulk regions. The two descriptions are matched in a coupling region where we ensure continuity of mass, momentum, energy and their fluxes. We demonstrate our scheme in simulating 1-D and 2-D, steady and transient heat transfers in channel flows. Good agreement between hybrid results and analytical or pure MD results is achieved. We then develop a multi-scale time algorithm to further accelerate hybrid simulations. The time algorithm is based on the temporal separation between the continuum time step and MD time step, and the assumption that the molecular dynamics in one continuum time step can be treated as quasi-steady state. We demonstrate our time algorithm by simulating isothermal channel flow driven by a sinusoidally moving wall. The results with our time algorithm converge to the original hybrid simulation results when the temporal separation is large enough, meanwhile the computational time is tremendously reduced.

ISBN: 9780549938477Subjects--Topical Terms:

783786
Engineering, Mechanical.

Multi-scale simulation of micro/nano flows and granular flows.
LDR:04180nam 2200301 a 45 001 862433
005 20100720
008 100720s2009 ||||||||||||||||| ||eng d
020 $a 9780549938477
035 $a (UMI)AAI3339874
035 $a AAI3339874
040 $a UMI $c UMI
100 1 $a Liu, Jin. $3 1030247
245 1 0 $a Multi-scale simulation of micro/nano flows and granular flows.
300 $a 146 p.
500 $a Adviser: Shiyi Chen.
500 $a Source: Dissertation Abstracts International, Volume: 69-12, Section: B, page: 7769.
502 $a Thesis (Ph.D.)--The Johns Hopkins University, 2009.
520 $a In this thesis we first develop a hybrid atomistic-continuum scheme for simulating micro- and nano-flows with heat transfer. The approach is based on spatial "domain decomposition" in which molecular dynamics (MD) is used in regions where atomistic details are important, while classical continuum fluid dynamics is used in the remaining bulk regions. The two descriptions are matched in a coupling region where we ensure continuity of mass, momentum, energy and their fluxes. We demonstrate our scheme in simulating 1-D and 2-D, steady and transient heat transfers in channel flows. Good agreement between hybrid results and analytical or pure MD results is achieved. We then develop a multi-scale time algorithm to further accelerate hybrid simulations. The time algorithm is based on the temporal separation between the continuum time step and MD time step, and the assumption that the molecular dynamics in one continuum time step can be treated as quasi-steady state. We demonstrate our time algorithm by simulating isothermal channel flow driven by a sinusoidally moving wall. The results with our time algorithm converge to the original hybrid simulation results when the temporal separation is large enough, meanwhile the computational time is tremendously reduced.
520 $a We develop a highly accurate and efficient molecular approach to simulate micro and nano electrokinetic flows. We calculate the long range Coulombic interactions using the Particle-Particle Particle-Mesh (P3M) approach. The Poisson equation for the electrostatic potential is solved in physical space using an iterative multi-grid technique. We first demonstrate our approach by simulating electro-osmotic channel flow with nano roughness on the walls. By comparing with traditional pressure driven flows, our results indicate that in electro-osmotic flow the roughness affects the flow through altering the charge density distribution in the electrical double layer (EDL). We then apply our approach for simulating electrowetting on dielectric (EWOD) and for the first time we test Lippmann's prediction of the relation between contact angle and applied voltage using atomistic simulations. We also suggest a possible hybrid scheme for electrokinetic flows based on "domain decomposition" and "constrained dynamics".
520 $a From recent theoretical work on models for binary particulate systems, an interspecies stress term was derived [Zhang, Ma, and Rauenzahn, Phys. Rev. Lett. 97, 048301 (2006)] in a two-equation formulation. This term is usually neglected in other models. However this term plays the key role in recovering the averaged equations for disperse two-phase flow from binary particulate systems. As a starting point, we study the behavior of stress terms in this two-equation formulation through numerical simulations of the simplest possible granular system. We find the interspecies stress is of the same order of magnitude as other stresses for dense systems. For systems with relative motion between two species, we find that both intra and interspecies stresses are quadratically proportionally to the relative velocity. The results are important in the attempt to seek a unified mathematical description that is valid for problems ranging from molecular scale mixing to disperse multiphase flows.
590 $a School code: 0098.
650 4 $a Engineering, Mechanical. $3 783786
650 4 $a Physics, Molecular. $3 1018648
690 $a 0548
690 $a 0609
710 2 $a The Johns Hopkins University. $3 1017431
773 0 $t Dissertation Abstracts International $g 69-12B.
790 $a 0098
790 1 0 $a Chen, Shiyi, $e advisor
791 $a Ph.D.
792 $a 2009
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3339874