     
|
|
|
I continued to organized the CCM colloquium series and Tom Russell handled the Denver portion of the CU-Denver/CU-Boulder Joint Seminars in Computational Math. The list of the talks can be found at the respective web pages, or in Section 9 of the Annual Report. We were fortunate to have had several visitors from overseas and out of state last academic year, namely: Paola Pietra from the IAN/CNR-Italy; Patrick Le Tallec and Herve Guillard from INRIA-France; Martin Stynes from University of Cork-Ireland; Vladimir Janovsky and Karel Segeth from the Charles University-Czech Republic; Houde Han from Tsinghua University-China; Azzeddine Soulaimani from Universite du Quebec-Canada; Fred Furtado and Patrick O'Leary from the University of Wyoming; Ivo Babuska from University of Texas at Austin; Deborah Sulsky from the University of New Mexico. Andrew Knyazev will be the new coordinator for the CCM colloquium series, starting Fall 98.
The Center for Computational Mathematics Series of Reports has now over 120 reports, and the collection can be found in hard-copy format in suite 647 of the CU-Denver Building. Some of the reports are available on line and further instructions on how to get copies can be found in its web page. Lynn Bennethum is the new Coordinator of the CCM Reports.
A highlight of the year was the Finite Element Circus. The circus was attended by prominent Applied Mathematicians and was held at the CU-Denver building. For those of you who do not know what finite elements are, here is a quick explanation: Various problems in engineering and physics are described by ``Physical Laws'' (such as momentum equilibrium, energy conservation) known to us for several centuries now (from Newton to Einstein). These physical laws in turn are represented by differential equations, which, in many applications, are complex to solve explicitly. Therefore computational methods are necessary to approximate the solutions of these differential equations. The finite element method is a numerical technique that systematically attack various differential equations and offer a road map to solve them accurately, even if the physical laws are posed in difficult geometries (such as the air flow around a car of a shape different than a box - I have never seen a car in the shape of a box, therefore techniques that deal with domains of non-trivial shapes can be handled by this technique). Roughly speaking, the method basically subdivide the domain of the physical problem into triangles or quadrilaterals in two dimensions and into tetrahedrals and hexahedrals in three dimensions and apply the equations using simple polynomial functions defined in these small regions. The applications of the finite element method are numerous to list here, but here are just a few examples: study of structural deformation (important for civil engineering and aerodynamics problems: computation of stresses that can yield rupture of a body, key to make design decisions in several bodies); aerodynamics (flows around objects: airplanes and cars are the most obvious examples. Both Boeing and car manufactures use the finite element method to make decision on the shape of their products); acoustics (study of noises or wave propagation: for example, how to interpret the signals of a sonar impeding on a surface under water: is it a fish or an enemy submarine?); porous media (oil and pollution problems); etc...
During this year we also initiated a CLAS Network pilot project with a server bought by CLAS and housed in the CCM lab: clas2000. The system is now operational and is being used by the Extended Studies Office as well as the Math Department. It has an application package, Applix, that runs on a LINUX operational system and is a suitable substitute to MS-Office suite. We plan to expand its usage to other units of CLAS.
For news and updates on the system throughout the year, please check our CCM homepage at http://www-math.cudenver.edu/ccm/ and the CCM Lab homepage at http://www-math.cudenver.edu/ccm/ccm-lab.html.