Data-driven balancing of linear dynamical systems
Preprint
Abstract
We present a novel reformulation of balanced truncation, a classical model reduction method. The principal innovation that we introduce comes through the use of system response data that has been either measured or computed, without reference to any prescribed realization of the original model. Data are represented by sampled values of the transfer function {or the impulse response} corresponding to the original model. We discuss parallels that our approach bears with the Loewner framework, another popular data-driven model reduction method. We illustrate our approach numerically in both continuous-time and discrete-time cases.Keywords
All Related Versions
This publication has 40 references indexed in Scilit:
- Comparing (Empirical-Gramian-Based) Model Order Reduction Algorithms2021
- A Link Between Gramian-Based Model Order Reduction and Moment Matching2021
- Data‐driven model order reduction of quadratic‐bilinear systemsNumerical Linear Algebra with Applications, 2018
- Balanced truncation for linear switched systemsAdvances in Computational Mathematics, 2018
- Model Order Reduction for Linear and Nonlinear Systems: A System-Theoretic PerspectiveArchives of Computational Methods in Engineering, 2014
- Balancing Transformations for Infinite-Dimensional Systems with Nuclear Hankel OperatorIntegral Equations and Operator Theory, 2013
- Numerical solution of large and sparse continuous time algebraic matrix Riccati and Lyapunov equations: a state of the art surveyGAMM-Mitteilungen, 2013
- Reduced-order models for control of fluids using the eigensystem realization algorithmTheoretical and Computational Fluid Dynamics, 2010
- Perturbation bounds for the Cholesky andQR factorizationsBIT Numerical Mathematics, 1991
- Exponentially convergent Fourier-Chebshev quadrature schemes on bounded and infinite intervalsJournal of Scientific Computing, 1987