Multiple shooting-local linearization method for the identification of dynamical systems

DOI10.1016/J.CNSNS.2016.01.022zbMATH Open1473.65086arXiv1507.00975OpenAlexW815236037MaRDI QIDQ2198902FDOQ2198902

Authors: Felix Carbonell, Juan Carlos Jimenez, Yasser Iturria-Medina 

Publication date: 15 September 2020

Published in: Communications in Nonlinear Science and Numerical Simulation (Search for Journal in Brave)

Abstract: The combination of the multiple shooting strategy with the generalized Gauss-Newton algorithm turns out in a recognized method for estimating parameters in ordinary differential equations (ODEs) from noisy discrete observations. A key issue for an efficient implementation of this method is the accurate integration of the ODE and the evaluation of the derivatives involved in the optimization algorithm. In this paper, we study the feasibility of the Local Linearization (LL) approach for the simultaneous numerical integration of the ODE and the evaluation of such derivatives. This integration approach results in a stable method for the accurate approximation of the derivatives with no more computational cost than the that involved in the integration of the ODE. The numerical simulations show that the proposed Multiple Shooting-Local Linearization method recovers the true parameters value under different scenarios of noisy data.

Full work available at URL: https://arxiv.org/abs/1507.00975

Recommendations

• Practical local identifiability of parameters of linear dynamic system
• Identification of linear systems with multiplicative noise from multiple trajectory data
• Multi-innovation identification methods for linear-parameter systems
• Nonlinear system identification of multi-degree-of-freedom systems
• scientific article; zbMATH DE number 4114530
• Parameter and structure identification of linear multivariable systems
• scientific article; zbMATH DE number 1803056
• scientific article; zbMATH DE number 3916157
• Identification of linear dynamic systems

zbMATH Keywords

parameter estimationchaotic dynamicslocal linear approximationnonlinear equationsmultiple shootingline search algorithmgeneralized Gauss-Newton

Mathematics Subject Classification ID

Strange attractors, chaotic dynamics of systems with hyperbolic behavior (37D45) Numerical solution of inverse problems involving ordinary differential equations (65L09)

Cites Work

• Expokit
• On the Numerical Solution of Constrained Least-Squares Problems
• Title not available (Why is that?)
• Computing integrals involving the matrix exponential
• Parameter Estimation for Differential Equations: a Generalized Smoothing Approach
• Line search algorithms with guaranteed sufficient decrease
• A Nonmonotone Line Search Technique and Its Application to Unconstrained Optimization
• Parameter estimation of ODE's via nonparametric estimators
• A Spline Least Squares Method for Numerical Parameter Estimation in Differential Equations
• A Survey of the Maximum Principles for Optimal Control Problems with State Constraints
• Ordinary differential equations with applications
• Title not available (Why is that?)
• NONLINEAR DYNAMICAL SYSTEM IDENTIFICATION FROM UNCERTAIN AND INDIRECT MEASUREMENTS
• An efficient line search for nonlinear least squares
• Computing multiple integrals involving matrix exponentials
• Dynamic properties of the local linearization method for initial value problems.
• Rate of convergence of local linearization schemes for initial-value problems
• Locally linearized Runge Kutta method of Dormand and Prince
• Construction and study of local linearization adaptive codes for ordinary differential equations
• A higher order local linearization method for solving ordinary differential equations
• Title not available (Why is that?)
• Dynamical State and Parameter Estimation
• A modified Newton method for the solution of ill-conditioned systems of nonlinear equations with application to multiple shooting
• Estimation of parameters in incomplete data models defined by dynamical systems
• Title not available (Why is that?)
• A numerical method for the computation of the Lyapunov exponents of nonlinear ordinary differential equations
• The local linearization method for numerical integration of random differential equations
• Least squares parameter estimation in chaotic differential equations
• Numerical discretization-based estimation methods for ordinary differential equation models via penalized spline smoothing with applications in biomedical research
• Linear Least Squares with Bounds and Linear Constraints
• Local Linearization Method for Numerical Integration of Delay Differential Equations
• Robust estimation for ordinary differential equation models

Cited In (7)

• Parameter estimation and control by penalized multiple shooting
• A partitioned quasi-likelihood for distributed statistical inference
• On the smoothness of nonlinear system identification
• Direct multiple shooting and generalized Gauss-Newton method for parameter estimation problems in ODE models
• Likelihood-based generalization of Markov parameter estimation and multiple shooting objectives in system identification
• A modified multiple shooting algorithm for parameter estimation in ODEs using adjoint sensitivity analysis
• Efficient numerical pricing of American options based on multiple shooting method: a PDE approach

This page was built for publication: Multiple shooting-local linearization method for the identification of dynamical systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2198902)