Xu, Shuxian
(2023)
Accurate prediction of fluids in motion: Attacking chaos, understanding turbulence and resolving complexity.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Simulating fluid motion accurately and robustly is an enduring problem due to the com- plexity and chaotic nature of turbulent flows, the latter of which is often amplified by mea- surement errors in initial conditions. These challenges are deeply embedded in fluid flow simulations, making it a complex field that requires both efficient computing and accurate predictions. At the core of the unpredictability in fluid flow simulations is the exponential growth of errors until saturation, which stands as a formidable obstacle to the precision of predictions. Recognizing that extending the predictability horizon of simulations is possible through ensemble simulations, the thesis explores computational solutions to overcome the practical limitations tied to these methods.
Building on recent advances in the field, the research introduces improved ensemble algorithms centered around an optimized application of the penalty method. It presents novel solutions to the critical computational problems traditionally associated with the penalty method, namely increased matrix coupling, ill-conditioned systems, and sensitivity to parameter selection. By refining the penalty method, we manage to achieve ensemble simulations capable of handling larger sets with computational complexity comparable to a single realization.
The findings mark a significant stride towards accurate and efficient prediction of fluids in motion. By confronting turbulence and chaos in fluid behaviors head-on and adeptly managing the complexities inherent in fluid dynamics simulations, the research paves the way for enhanced predictive precision in the field. The novel methodologies and insights presented in this thesis will serve as invaluable resources for future investigations and applications in the diverse world of fluid dynamics.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
6 September 2023 |
| Date Type: |
Publication |
| Defense Date: |
27 June 2023 |
| Approval Date: |
6 September 2023 |
| Submission Date: |
11 July 2023 |
| Access Restriction: |
2 year -- Restrict access to University of Pittsburgh for a period of 2 years. |
| Number of Pages: |
131 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Dietrich School of Arts and Sciences > Applied Mathematics |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Navier-Stokes Equation, Penalty method, Sparse Grad-Div, Effective Con- dition Number, Adaptive, Ensemble Calculation |
| Date Deposited: |
07 Sep 2023 01:33 |
| Last Modified: |
07 Sep 2023 01:33 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/45083 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[img]](https://d-scholarship.pitt.edu/style/images/fileicons/application_pdf.png)
![[feed]](/images/feed-icon-32x32.png){kind=icon}