T. KORAKIANITIS Research in turbomachinery applications	"Korakianitis" is pronounced phonetically
email: tk@mecf.wustl.edu	link: [ 183 kB audio wav]

   The following table of links is a site navigation map  

T. Korakianitis' main page

Research in piston engines	(you are here) Research in turbomachines	Engineering practice	Opinions on education
Internal combustion engines laboratory Emissions Combustion Piston dynamics Valve train dynamics Nutating-disk engine Turbochargers	Airfoil and blade design Stator-rotor interactions Aircraft inlets Combustion Emissions Gas-turbine transients Gas-turbine performance	Machine design topics Thermofluids design Blood flow in human heart Surgical instruments Power generation Aeronautical applications Piston engines	Graduate thermodynamics (ME 512) Graduate piston engine design (ME 578) Graduate turbomachinery design (ME 574) Senior design course (ME 404/404a/404b) Junior thermodynamics (ME 320A) Freshman internal combustion engines (ME 147)

Archival publications

The impetus for our experimental, numerical and analytic work is the practical analysis, design and manufacturing of machinery, machinery components and processes. We use mechanics, dynamics, thermofluid sciences, and other sciences as necessary in the analysis and design of various types of contemporary machinery, including but not limited to turbomachinery and piston engines. Most of our scholarly publications address aspects of steady, unsteady and transient performance of machinery, power generation and energy conversion components, systems and processes.

Other members of the research group, 1996-1999

Dr. Harry Xin (piston-assembly dynamics and lubrication)
Dr. Lonn Grandia (unsteady flow in bi-ventricular assist devices)
Mr. John Ladd (preparing doctoral dissertation in aircraft-inlet turbulence modeling)
Mr. Richard Dyer (preparing doctoral dissertation in numerical models for transient combustion)
Mr. Brandon Wegge (preparing SM thesis in 3D CAD model for turbomachinery airfoils)
Mr. Brian Mann (selecting doctoral topic)
Mr. Jacob Grantstrom (SM thesis in off-design performance of cogeneration power plants)
Mr. Peter Wassingbo (SM thesis in off-design performance of cogeneration power plants)
Mr. Krister Svensson (SM thesis on part-load performance of industrial gas turbines)
Mr. Paolo Tavella (SM thesis on part-load performance of aircraft gas turbines)
Mr. Joachim Landes (measurement of emissions during engine transients)

Unifying themes: steady, periodic, transient and unsteady models of natural processes

We prefer to think of natural phenomena as steady, because it is easier to model them in controlled experiments or analysis. Most models of unsteady or transient natural phenomena make the governing equations hyperbolic. Hyperbolic equations (in most cases) are harder to solve than the parabolic or elliptic equations that (again in most cases) can be used to solve steady processes. As a result, whenever possible, experimental, numerical and analytic processes are modeled as steady.

Given our desire to simplify models, which of the following examples (several of which have applications in internal combustion engines) can be modeled as steady, and which must be modeled as unsteady? Which are periodic and which are transient? While the graphics are loading we offer this hint: the answer is the same in all cases...

The answer is: IT DEPENDS!

For example in the "night-to-day island" image above modeling the growth of a palm-tree leaf for 5 minutes around noon can be modeled as steady. Modeling the same process over 24 hours may be modeled as periodic, while modeling it for 3 months it is transient.

Yet the actual duration of one day changes as the sun emits approximately 390 x 10¹⁸ MW of power, reducing its mass by about 4,300,000 tons per second. However, the last change can be ignored for most (but not all !!) practical applications, as this loss of mass depletes 1/10,000 of the total mass of the sun in approximately 1,500,000 years. This effect can be ignored for most phenomena of practical importance. Ignoring this last effect may lead to erroneous conclusions on research on the origins of life on earth. Several other factors (of much faster time scale) must be considered in models of climate change, or "global warming" as some call climate change recently; otherwise the model will lead to erroneous conclusions. This should not be interpreted as a blessing or condemnation for either side of the discussion on the important issue of climate change, but as a wake-up call to attention on what is modeled and how.

When examined in detail all processes in nature are unsteady and all states are non-equilibrium states. Depending on the purpose of the model, several of these processes can be modeled as steady. Deciding when an unsteady model is necessary is based on examination of suitable non-dimensional parameters. Theoretical models are attempts to emulate nature's response. Numerical and analytic solutions of the theoretical model can only reproduce aspects of nature included in the model. Comparison of analytic solutions (sometimes simplified models) with numerical results gives confidence in the accuracy of the model. Experimental models of the natural process provide data that can improve or validate the theoretical model. However, in all cases the final test is building and testing of a prototype. Steady models of inherently unsteady processes always lead to erroneous results.

The individual research projects in the table of links above outline aspects of steady and unsteady modeling of turbomachinery.

Click the left arrow  to return to T. Korakianitis' main page