Combining off-white and sparse black models in multi-step physics-based systems identification

DOI10.1016/J.AUTOMATICA.2025.112409MaRDI QIDQ6947625zbMATH OpenFDO

Authors Cesare Donati, Martina Mammarella, Fabrizio Dabbene, Carlo Novara, Constantino M. Lagoa

Publication date 5 August 2025

Published in Automatica (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/J.AUTOMATICA.2025.112409

zbMATH Keywords

gray-box modelingnonlinear system identificationparametric optimizationtime-invariant systems

Mathematics Subject Classification ID

System identification (93B30) Nonlinear systems in control theory (93C10)

Review text

The main result of this paper is the theoretical analysis of the identification parameter error bounds. The physical interpretability is incorporated using an off-white model described by the specific nonlinear equations derived from physical principles. The proposed framework is validated using the cascade tank system identification benchmark and one example is based on synthetic data.

Cites work

• A flexible state-space model for learning nonlinear dynamical systems
• A simple test to check the optimality of a sparse signal approximation
• Kernel-based methods for Volterra series identification
• Nonlinear black-box modeling in system identification: A unified overview
• Nonlinear System Identification: A User-Oriented Road Map
• Sparse Identification of Nonlinear Functions and Parametric Set Membership Optimality Analysis
• Sparse plus low rank network identification: a nonparametric approach

This page was built for publication: Combining off-white and sparse black models in multi-step physics-based systems identification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6947625)