Modeling Cumulative Damage to Flow Surfaces and Assessing its Impact on Wall Turbulence

Licari, Anthony M.

Permalink

https://hdl.handle.net/2142/18274 Copy

Description

Title

Modeling Cumulative Damage to Flow Surfaces and Assessing its Impact on Wall Turbulence

Author(s)

Licari, Anthony M.

Issue Date

2011-01-14T22:43:33Z

Director of Research (if dissertation) or Advisor (if thesis)

Christensen, Kenneth T.

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• turbulence
• channel flow
• Particle image velocimetry (PIV)
• surface roughness
• cumulative damage

Abstract

The present effort investigated the influence of cumulative damage of an initially smooth surface on the statistical characteristics of wall turbulence. Singular value decomposition was employed to decompose a highly-irregular surface topography replicated from a turbine blade damaged by spallation into topographical modes embodying successively smaller roughness length scales. These modes were then used to develop multiple reduced-order models of the original roughness that, at least in spirit, successively captured the evolution of the surface roughness from initial smaller-scale defects that would likely form just after deployment to the eventual cumulative formation of larger-scale roughness features. Particle-image velocimetry measurements of flow at a friction Reynolds number around 1825 over various topographical models were gathered in turbulent channel flow. The cumulative impact of these reduced-order models on wall turbulence was then assessed by comparing mean velocities, Reynolds normal stresses, and Reynolds shear stresses. It was observed that the turbulence statistics of the flow increase in magnitude as larger-scale features are introduced into the surface, leading up to the full, original surface. In addition, even weak surface damage that a practical flow surface might endure within the first fraction of its deployment lifetime can significantly enhance turbulence and therefore progressively degrade system performance. Thus, proper modeling of flow-system performance must account for the dynamic nature of flow surfaces, particularly those exposed to severe operating conditions.

Graduation Semester

2010-12

Permalink

http://hdl.handle.net/2142/18274

Copyright and License Information

Copyright 2010 Anthony M. Licari

Owning Collections