Difference between revisions of "Emgr"

Revision as of 11:54, 6 June 2013

Synopsis

emgr - Empirical Gramian Framework. Empirical gramians can be computed for linear and nonlinear control systems for purposes of model order reduction, uncertainty quantification or system identification. Model reduction using empirical gramians can be applied to the state space, to the parameter space or to both through combined reduction. The emgr framework is a compact open source toolbox for gramian-based model reduction and compatible with OCTAVE and MATLAB.

Features

Application matrix for the empirical gramian framework

emgr encompasses six types of gramians:

Empirical Controllability Gramian
Empirical Observability Gramian
Empirical Cross Gramian
Empirical Sensitivity Gramian
Empirical Identifiability Gramian
Empirical Joint Gramian

applicable to:

Linear + Nonlinear Systems
First + Second Order Systems
Parametric Systems

and with sample implementations for:

Balanced Truncation + Direct Trunction
Parameter Identification + Sensitivity Analysis
Parameter Reduction
Combined State and Parameter Reduction
(Bayesian) Inverse Problem Reduction

References

C. Himpe, M. Ohlberger. "A Unified Software Framework for Empirical Gramians". arXiv E-print, arXiv:1301.6879 [math.OC], 2012.

Links

http://gramian.de

Contact

Christian Himpe

@@ Line 7: / Line 7: @@
 [http://gramian.de emgr] - '''Em'''pirical '''Gr'''amian Framework.
-Empirical gramians can be computed for linear and nonlinear control systems for purposes of model order reduction or system identification.
+Empirical gramians can be computed for linear and nonlinear control systems for purposes of model order reduction, uncertainty quantification or system identification.
 Model reduction using empirical gramians can be applied to the state space, to the parameter space or to both through combined reduction.
 The '''emgr''' framework is a compact open source toolbox for gramian-based model reduction and compatible with [http://www.gnu.org/software/octave/ OCTAVE] and [http://www.mathworks.de/products/matlab/ MATLAB].