Sorensen, Danny C.
Professor
Department of Computational and Applied Mathematics
Rice University
United States of America
Biography
Professor Sorensen works in the general area of numerical analysis. His current research involves the design, analysis, and computer implementation of algorithms for solving fundamental problems in linear algebra. The primary focus of this research has been on very large-scale eigenvalue problems. Extremely large problems often require the use of parallel algorithms and software. Professor Sorensen and his colleagues have developed a software package, ARPACK, for solving large-scale symmetric, nonsymmetric standard, or generalized eigenvalue problems. The software is currently being used to solve nonsymmetric problems with 16 million degrees of freedom. Currently, Professor Sorensen is working on dimension reduction techniques for dynamical systems. The projection techniques used for large eigenvalue computations are being adapted to provide model order reduction for very large linear time invariant control systems.
Research Interest
Krylov Methods for Large Eigenvalue problems and Model Reduction for Dynamical Systems.
Publications
-
Sorenson DC, Mehrmann V (2003) Dimension reduction of large-scale systems. Proceedings of a workshop held in Oberwolfach, Germany.
-
Antil H, Sorenson DC (2010) Domain decomposition and model reduction for the numerical solution of PDE constrained optimization problems with localized optimization variables. Comput Visualiz Scie.
