{"id":"arxiv/2603.17423","url":"https://arxiv.org/abs/2603.17423","tags":["math.DS","cs.LG","nlin.CD"],"score":0,"title":"Data-driven model order reduction for structures with piecewise linear nonlinearity using dynamic mode decomposition","domain":"math.DS","source":"arxiv","authors":["Akira Saito","Masato Tanaka"],"reviews":0,"subject":"math.DS","summary":"Piecewise-linear nonlinear systems appear in many engineering disciplines. Prediction of the dynamic behavior of such systems is of great importance from practical and theoretical viewpoint. In this paper, a data-driven model order reduction method for piecewise-linear systems is proposed, which is based on dynamic mode decomposition (DMD). The overview of the concept of DMD is provided, and its application to model order reduction for nonlinear systems based on Galerkin projection is explained. The proposed approach uses impulse responses of the system to obtain snapshots of the state…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-18","reactions":{}}