Kenneth E. Jansen Associate Professor  7223 Center for Industrial Innovation (CII)  Tel: (518) 276-6755; Fax: (518) 276-4886;  E-Mail: kjansen@scorec.rpi.edu [ ME AE & M || Faculty || Online Search ]

After receiving his B.S. in Mechanical Engineering in 1987 from the University of Missouri-Columbia, Kenneth Jansen went on to graduate school at Stanford University where he earned an M.S. degree in Mechanical Engineering in 1988 and his Ph.D. in Mechanical Engineering with a minor in Aeronautical Engineering in 1993 under an Office of Naval Research Fellowship. He then joined the Center for Turbulence Research, a joint NASA-Stanford program, where he was awarded a three year post-doctoral research fellowship. In August, 1996 he became a member of the Rensselaer faculty. Research Interests and Activities Major Interests: Computational mechanics with emphasis on fluid dynamics. Turbulence theory, simulation, and modeling. Parallel computing.

The motivation of Ken's research is to provide engineers with a better predictive capability for fluid dynamics problems, especially those where turbulence plays an non-negligible role. To this end, his research, at the most applied level, seeks to develop simple models which describe the net effect or average of the turbulence upon the mean flow equations. These models, when combined with a fully unstructured-grid finite element method, allow engineers to model arbitrarily complex flow problems. Unfortunately, these models are not yet able to describe all turbulent flows. Therefore, other forms of simulating turbulence are also pursued. These forms are: 1) Large-Eddy Simulation (LES) where the large scale motions of the turbulence are resolved in the computation leaving only the fine scale motions to be modeled, 2) Direct Numerical Simulation (DNS) where all of the turbulent motions are resolved in the computational model. These alternate forms are useful both to develop a more basic understanding of the theory of turbulence and to help improve the averaged models used by engineers.

Click here to view velocity contours on four planes parallel to the wing and one plane normal to the wing from a LES(35k). To see an animation (colorful one)  of this flow click one or more of the following animations here(8.9MB), here(11.6MB-long time interval), here(15.5MB-shorter time more frequent sampling smoother), here(a single frame image of the animation). Click here to view vorticity isosurfaces from and LES of turbulent flow over a cavity(1.4MB, might want to wait until night or for a quick view try this one which is a much smaller file but is more fuzzy(44k)). If you have a VERY fast connection and want to see a QuickTime animation of the cavity(4.7MB).

An exciting new area of research involves finding ways to locally select models based on the needs of the simulation. For example flow around the airfoil shown above is well predicted by RANS, except when the flow nears separation. Therefore, it would be a dramatic improvement to modeling if most all of the domain could be simulated by RANS and an LES was used only in the separated region where the RANS model was inadequate. Such hybrid models are very difficult to develop so as a first step we are currently studying a simplified version of the airfoil problem. Here we show results obtained by taking a mean flow from a RANS calculation of a flat plate boundary layer, and using it as an inflow for an LES of a flat plate boundary layer. The LES of course requires more than a steady mean. We have generalized the technique of Lund, Wu and Squires, to extract only the fluctuations from an interior plane. Here we show results for the mean flow at various stations, the time and spanwise averaged fluctuations (in plus coordinates), and the development of the boundary layer thickness , the displacement thickness , and the momentum thickness . The results show reasonable agreement with the experiments, though as always, there is room for improvement. Efforts are also underway to extend this work to adverse pressure gradient flows like those encountered on the airfoil. The scaling will need to be modified in this case.

For the those not completely sure what turbulence is, click here to view some experimental photographs of a jet undergoing transition to turbulence. 1(9k), 2(10k), 3(7k). Note that in the third picture the jet (shown in blue) is initially laminar (looks like a column) then begins to wiggle as instabilities grow and then finally transitions to fully developed turbulence in the bottom of the frame. Here we find the simultaneous presence of very large scale motions and very fine scale motions which is the hallmark of turbulence (making computations so difficult).

Flow within the human vascular system is usually not turbulent but can still be highly transiatory and quite interesting.  Here is a link to the web page describing some of our recent work in this area. The animations were made by one of my   undergraduate research assistants in the summer of 2000.

I am also interested in developing animations of some of my computations which help to illustrate fundamental concepts in fluid dynamics. This flow over a cylinder at Re_D=100 illustrates how periodic shedding of vortices behind a cylinder can be visualized using non-stationary streamlines, contours of vorticity, and finally a time varying pressure field which, when integrated over the surface of the cylinder, results in an oscillating lift and drag force on the cylinder. By visualizing the formation and release of the vortices it is clear that a vortex, shed off of the top, creates a low pressure in its core which makes a lower net pressure on the top of the cylinder resulting in a net lifting force on the cylinder (conversely negative lift when shed off of the bottom). The drag force curve appears at twice the frequency since the x direction force could care less whether the vortex is shed off of the top or the bottom.

Teaching Activities

 (MANE 4070) Aerodynamics 1

(MEAE-37-6270) Computational Fluid Dynamics

Download the stabilized FEM code, examples and manuals as a tar file (on a unix  machine type tar -xvf Distrib.tar and it should create a Distrib directory with the Source, a Toy problem, and pre and post processing tools....more on this later).

Download the incompressible version

Selected Publications

  A generalized-alpha method for integrating the filtered Navier-Stokes equations with a stabilized finite element method, K.E. Jansen, C. Whiting and G.M. Hulbert, Computer Methods in Applied Mechanics and Engineering, in press (2000).

Hierarchical basis in stabilized finite element methods for compressible flows, C.H. Whiting, K.E. Jansen and S. Dey, Computer Methods in Applied Mechanics and Engineering, accepted (2000).

A Stabilized finite element method for the incompressible Navier-Stokes equations using a hierarchical basis, C.H. Whiting and K.E. Jansen, International Journal of Numerical Methods in Fluids,, in press (2000).

Large Eddy Simulation and the variational multiscale method,.  T.J.R. Hughes, L. Mazzei, and K.E.. Jansen, Computing and Visualization in Science, 3 (2000) 47-59.

A better consistency for low-order stabilized finite element methods, K.E. Jansen, S.S. Collis, C. Whiting and F. Shakib, Computer Methods in Applied Mechanics and Engineering, 174 (1999) 153-170.

A stabilized finite element method for computing turbulence, K.E. Jansen, Computer Methods in Applied Mechanics and Engineering, 174 (1999) 299-317.

Computation of turbulence with stabilized methods, K.E. Jansen, invited paper in 4th Japan-US Symposium on Finite Element Methods in Large-Scale Computational Fluid Dynamics, April 1998.

Large-eddy simulation using unstructured grids, K.E. Jansen, invited paper in Advances in DNS/LES, Greyden Press, (1997) 117--128.

Large-eddy simulations of flow around a NACA 4412 airfoil using unstructured grids, K.E. Jansen, in CTR Annual Research Briefs 1996, Center for Turbulence Research, Stanford University/NASA Ames Research Center, (1996) 225--232.

LES on unstructured deforming meshes: towards reciprocating IC engines, D. Haworth and K.E. Jansen, in Proceedings of the 1996 Summer Program, Center for Turbulence Research, Stanford University/NASA Ames Research Center, (1996) 329--346.

Preliminary large-eddy simulations of flow over an airfoil using unstructured grids, K.E. Jansen, in CTR Annual Research Briefs 1995, Center for Turbulence Research, Stanford University/NASA Ames Research Center, (1995) 61--72.

A family of dynamic models for large-eddy simulation, D. Carati, K.E. Jansen, and T. Lund in CTR Annual Research Briefs 1995, Center for Turbulence Research, Stanford University/NASA Ames Research Center, (1995) 35--40.

A stabilized finite element method for the Reynolds-averaged Navier-Stokes equations, T.J.R. Hughes and K.E. Jansen, Surveys on Mathematics for Industry Vol. 4 (1995) 279--317.

Unstructured grid large eddy simulation of flow over an airfoil, K.E. Jansen, in CTR Annual Research Briefs 1994, Center for Turbulence Research, Stanford University/NASA Ames Research Center, (1994) 161--173.

Stabilized finite element methods in fluids: inspirations, origins, status and recent developments, T.J.R. Hughes, G. Hauke, K.E. Jansen, and Z. Johan, in Recent Developments in Finite Element Analysis, (Eds.: T.J.R. Hughes, E. O\~nate and O.C. Zienkiewicz), CIMNE, Barcelona (1994).

Finite element applications to the Reynolds-averaged Navier-Stokes equations, K.E. Jansen, Advances in Computational Methods in Fluid Dynamics, (ed. K.N. Ghia, U. Ghia, and D. Goldstein), ASME FED-Vol. 196, New York (1994) 107--115.

New perspectives and possibilities within the field of finite element methods, T.J.R. Hughes and K.E. Jansen, Proceedings of the XXII International FEM-Congress, Baden-Baden, Germany (1993).

Unstructured grid large eddy simulation of wall bounded turbulent flows, K.E. Jansen, in CTR Annual Research Briefs 1993, Center for Turbulence Research, Stanford University/NASA Ames Research Center, (1993) 151--156.

Current reflections upon stabilized finite element methods for computational fluid mechanics, T.J.R. Hughes, G. Hauke, K.E. Jansen, and Z. Johan, in Finite Elements in Fluids; New Trends and Applications, (Eds.: K. Morgan, E. O\~nate, J. Periaux, J. Peraire, O.C. Zienkiewicz), SEMNI, Barcelona, Spain (1993) 44--63 .

Implementation of a one-equation turbulence model within a stabilized finite element formulation of a symmetric advective-diffusive system, K.E. Jansen, Z. Johan, T.J.R. Hughes, Computational Methods in Applied Mechanics and Engineering Vol. 105 (1993) 405--433.

Finite element methods in wind engineering, T.J.R. Hughes and K.E. Jansen, Journal of Wind Engineering, Vol. 52 (1992) 32--48.

Fast projection algorithm for unstructured meshes, K.E. Jansen, F. Shakib, and T.J.R. Hughes, in \it Computational Nonlinear Mechanics in Aerospace Engineering, (ed. S.N. Atluri), AIAA, Washington D.C. (1992).

Symmetrization of the turbulent Navier-Stokes equations, K.E. Jansen and T.J.R. Hughes, ONR/DARPA Structural Acoustics Review, The University of Texas, Austin, Texas (1991).
 
 

Acknowledgements

Much of the material presented above (since Febuary 2000) has been based upon work supported by the National Science Foundation under Grant No.9935840.

Any opinions, findings and conclusions or recommendation expessed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation (NSF).
 

Personal

    Kenneth is married to Lupita Montoya who also received her Ph.D. from Stanford in Environmental Engineering. They are expecting their first child on January 22, 2002 and offer the following ultrasound images to friends and family:  joshua-whole.jpg ,  joshua-head.jpg ,  joshua-thumbs.jpg  and  joshua-feet.jpg and finally he was born on January 31, 2002 at 2:33 AM weighing in at 8 lbs 11 oz and 21 inches long (here is the first picuture of the three of us...more to follow)   threeofus.jpg. Working off of dialup from the hospital makes transfering files too slow so all I can give you for now in terms of closups is this beautiful yawn.