Predictive Engineering and Computational Sciences

The PECOS center pursues predictive computational simulations of complex physical systems, especially those involving fluid mechanics. To be predictive, simulations need to be based on well validated mathematical models, have well controlled numerical errors, and include quantified uncertainties. To enable predictive simulations, center researchers develop and implement mathematical models of physical phenomena and systems, formulate representations of uncertainties in such models, validate them and then use them to study the modeled phenomena and systems. Predictive modeling and simulation has been pursued for a variety of complex systems. Of particular in interest is turbulent fluid flows, for which models ranging in fidelity from the direct simulation of the Navier-Stokes equations to the Reynolds Averaged Navier-Stokes equations have been employed. Also of great current interest is predictive simulation of inductively coupled plasma torches.