Deep subspace encoders for nonlinear system identification

DOI10.1016/J.AUTOMATICA.2023.111210MaRDI QIDQ6136161zbMATH OpenFDO

Authors Gerben I. Beintema, Maarten Schoukens, Roland Tóth

Publication date 28 August 2023

Published in Automatica (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2210.14816

zbMATH Keywords

deep learningsystem identificationsubspace identificationnonlinear state-space modeling

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) System identification (93B30) Nonlinear systems in control theory (93C10)

Abstract: Using Artificial Neural Networks (ANN) for nonlinear system identification has proven to be a promising approach, but despite of all recent research efforts, many practical and theoretical problems still remain open. Specifically, noise handling and models, issues of consistency and reliable estimation under minimisation of the prediction error are the most severe problems. The latter comes with numerous practical challenges such as explosion of the computational cost in terms of the number of data samples and the occurrence of instabilities during optimization. In this paper, we aim to overcome these issues by proposing a method which uses a truncated prediction loss and a subspace encoder for state estimation. The truncated prediction loss is computed by selecting multiple truncated subsections from the time series and computing the average prediction loss. To obtain a computationally efficient estimation method that minimizes the truncated prediction loss, a subspace encoder represented by an artificial neural network is introduced. This encoder aims to approximate the state reconstructability map of the estimated model to provide an initial state for each truncated subsection given past inputs and outputs. By theoretical analysis, we show that, under mild conditions, the proposed method is locally consistent, increases optimization stability, and achieves increased data efficiency by allowing for overlap between the subsections. Lastly, we provide practical insights and user guidelines employing a numerical example and state-of-the-art benchmark results.

Recommendations

• System identification through Lipschitz regularized deep neural networks
• Randomized algorithms for nonlinear system identification with deep learning modification
• Subspace identification for non-linear systems with measured-input non-linearities
• Subspace learning of neural networks
• Recursive nonlinear-system identification using latent variables
• \(\mathcal L_{2}-\mathcal L_{\infty }\) nonlinear system identification via recurrent neural networks
• Sparse Bayesian deep learning for dynamic system identification
• Sparse deep Gaussian process approximation and application of dynamic system identification
• Non-intrusive model reduction of large-scale, nonlinear dynamical systems using deep learning

Cites work

• scientific article; zbMATH DE number 194951 (Why is no real title available?)
• Automating Data-Driven Modelling of Dynamical Systems
• Block-oriented nonlinear system identification
• Continuous-time system identification with neural networks: model structures and fitting criteria
• Convergence analysis of parametric identification methods
• Fading memory and the problem of approximating nonlinear operators with Volterra series
• Identification of Wiener-Hammerstein systems by a nonparametric separation of the best linear approximation
• Identification of block-oriented nonlinear systems starting from linear approximations: a survey
• Identification of nonlinear systems using polynomial nonlinear state space models
• Immersion and immersion by nonsingular feedback of a discrete-time nonlinear system into a linear system
• Learning nonlinear state-space models using autoencoders
• Model structure simplification of Nonlinear Systems via immersion
• Modeling and identification of linear parameter-varying systems
• Nonlinear System Identification: A User-Oriented Road Map
• Nonlinear control systems.
• Nonlinear control systems: An introduction
• Nonlinear system identification using neural state space models, applicable to robust control design
• Nonlinear system identification. NARMAX methods in the time, frequency, and spatio-temporal domains
• On the smoothness of nonlinear system identification
• Open Stochastic Systems
• Regularized nonparametric Volterra kernel estimation
• Subspace methods for system identification.
• System identification of nonlinear state-space models
• Unbiased minimum variance estimation for systems with unknown exogenous inputs
• Wiener-Hammerstein system identification - an evolutionary approach

Cited in

(2)

• Meta-state-space learning: an identification approach for stochastic dynamical systems
• Exploiting the capacity of deep networks only at the training stage for nonlinear black-box system identification

This page was built for publication: Deep subspace encoders for nonlinear system identification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136161)