Michael Martell - Research

I am a PhD candidate in Mechanical Engineering at the University of Massachusetts Amherst, working in the area of Computational Fluid Dynamics (CFD), specifically turbulence modeling. I work with Prof. Blair Perot. We are currently implementing a new turbulence model, the Oriented-Eddy Collision Turbulence Model, into an open-source collection of CFD libraries called OpenFOAM. For more information about OpenFOAM, see this website. Prior to this, I worked with Prof. Jonathan Rothstein investigating turbulent drag reduction using superhydrophobic surfaces (SHS).

My interests also include open-source software (such as Linux and OpenFOAM), high-performace computing, and efficient use of message-passing schemes such as MPI. In that vein, I administer a parallel supercomputer cluster here at UMass. The cluster is a shared resource, used by the engineering faculty and graduate students from Chemical, Civil, Mechanical, and Industrial engineering. See here for more details.

When I'm not in front of a computer, I'm usually behind a camera, often half way up a mountain. I greatly enjoy photography (see my other site) and hiking. If you were curious, I am standing behind an F-15 Eagle in the photo above.

Research

The Oriented Eddy Collision Turbulence Model

The Oriented Eddy Collision Model (OEC) demonstrates that it is feasible to model turbulence as a collection of interacting (colliding) particles that have an inherent orientation. Each particle (usually referred to as an "eddy") can be represented as a deformable, planar elliptical disk. The transport equations for these eddies can be derived such that all linearized turbulence (turbulence in the rapid distortion theory limit) is captured exactly. Simple models for the eddy collisions capture the non-linear behavior such as the energy cascade present in the flow as well as return to isotropy. In its current state, OEC is physically accurate, fairly simple (having few constants, most of which are set by theory), extensible (to complex flow situations), and computationally tractable. OEC has been implemeted in OpenFOAM (see above), is fully parallelized, and is being validated for a variety of flow situations. See my first paper on the subject here, where the model is explained and tested for some canonical flows.

To read more about the model, its history, and related publications, see my Ph.D Proposal here [PDF]. I've also put together a poster. See a draft here (large PDF).

Turbulent Drag Reduction with Superhydrophobic Surfaces

Superhydrophobic surfaces combine hydrophobic surface chemistry with topological microfeatures. These surfaces have been shown to provide drag reduction in laminar (see Jia Ou's work on the subject here and here [PDFs]) and turbulent flows (for more information about the expirimental side of this research, performed by Robert Daniello, look here [PDF]). Using an in-house parallel DNS code written by myself and Prof. Blair Perot, we performed direct numerical simulations to investigate the drag reducing performance of superhydrophobic surfaces in turbulent channel flow. Slip velocities, wall shear stresses, and Reynolds stresses were considered for a variety of superhydrophobic surface microfeature geometry configurations (including flow-aligned ridges, angled ridges, transverse ridges, as well as flow-aligned posts and angled posts) at friction Reynolds numbers of Re = 180, Re = 395 and Re = 590. We showed that superhydrophobic surfaces are capable of reducing drag in turbulent flow situations by manipulating the laminar sublayer. For the largest micro-feature spacing, an average slip velocity over 80% of the bulk velocity was observed, and the wall shear stress reduction was found to be greater than 50%. Simulation results suggested that the mean velocity profile near the superhydrophobic wall scales with the wall shear stress and the log layer is still present, but both are offset by a slip velocity that is primarily dependent on the microfeature spacing. See selected publications for more information about my simulations. My thesis can be found here [PDF].

Selected Publications / CV

Michael B Martell and J. Blair Perot, "The Oriented-Eddy Collision Turbulence Model," accepted for publication in Flow, Turbulence, and Combustion (January 2012). Preprint available here.

Thomas Elboth, Bjørn Anders Pettersson Reif, Øyvind Andreassen and Michael B. Martell, "Flow Noise Reduction from Superhydrophobic Surfaces," accepted to GEOPHYSICS (September 2011). Preprint available here.

Michael B. Martell, Jonathan P. Rothstein and J. Blair Perot, "An analysis of superhydrophobic turbulent drag reduction mechanisims using direct numerical simulation," Phys. Fluids 22, 065102 (2010). Download here [PDF].

Michael B. Martell, J. Blair Perot and Jonathan P. Rothstein, "Direct numerical simulations of turbulent flows over superhydrophobic surfaces," Journal of Fluid Mechanics 620, 31-41 (2009). Download here [PDF].

Current Curriculum Vitae - updated 23 January 2012 [PDF]

NOTE TO LINUX USERS: Preprint and CV PDFs were generated using DVItoPDF. Some PDF viewers may render them improperly (including in-browser viewers such as those found in Google Chrome). I suggest getting Acrobat Reader for Linux (its free).

Images

Below is a small collection of images from my work on turbulent flow over superhydrophobic surfaces. Stream traces reveal the swirling eddies in the flow as they interact with the superhydrophobic surfaces. This work made it on to the cover of Physics Today in 2009.

Some preliminary images of turbulent flow over an oblate spheroid using the Oriented Eddy Collision model.

A mesh I recently made for an Akula-class submarine, and "Cyclops", the 608-core Linux computer cluster I helped to build and administer here at UMass. This cluster was used to generate the pictures of the oblate spheroid above.

Links

OpenFOAM - the focus of my current work

NNFD website - information about SHS turbulent drag reduction

LinkedIn - my current profile

theory37 - my personal site

slashdot - news for nerds, stuff that matters

hacker news - I'm not a hacker, but I read their news

openSUSE - the OS I use at work

Xubuntu - the OS I use at home