Data-Guided Model Reduction for Multilinear Time Invariant Systems.
Year: 2019
Type: preprint
Abstract: In this paper, we explore the role of tensor algebra in the balanced truncation (BT) based model reduction/identification for high-dimensional multilinear/linear time invariant systems. Tensor decompositions facilitate discovery of hidden patterns/redundancies in multiway data by capturing higher-order interactions and couplings to obtain highly compressed yet accurate representations, enabling memory and computational advantages. In particular, we employ tensor train decomposition (TTD), which compared to other tensor decompositions, provides a good compromise between numerical stability and level of compression, and has an associated algebra which facilitates computations. Using TTD, we propose a new BT approach which we refer to as higher-order balanced truncation, and consider different data-guided variations including higher-order empirical gramians, higher-order balanced proper orthogonal decomposition and higher-order eigensystem realization algorithm. We perform computational and memory complexity analysis for these different flavors of TTD based BT methods, and compare with the corresponding standard BT methods, to develop insights into where the proposed framework may be beneficial. We provide numerical results on simulated and experimental datasets showing the efficacy of the proposed framework.
Source: arXiv (Cornell University)
Language: English
Subfield: Computational Mathematics
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2992432622