Multi-directional model validity tests for non-linear system identification

DOI10.1080/002071700219830MaRDI QIDQ4485348zbMATH OpenOpenAlexFDO

Authors Kezhi Mao, Stephen A. Billings

Publication date 12 June 2000

Published in International Journal of Control (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1080/002071700219830, http://eprints.whiterose.ac.uk/81021/1/acse%20research%20report%20677.pdf

zbMATH Keywords

model validationmulti-directional test algorithmsnonlinear system identificationstatistical validation

Mathematics Subject Classification ID

System identification (93B30) Nonlinear systems in control theory (93C10) Identification in stochastic control theory (93E12)

Review text

Model validation is an important procedure in a system identification study. Tests using the autocorrelation function between the residuals and the inputs have been developed and implemented. But these tests are not sufficient to detect unmodelled nonlinear terms. Nonlinear model validation algorithms have been developed. They are not performable in some situations. This paper studies several examples to describe and explain why we have failures. The causes of the problem are analyzed and two multi-directional test algorithms are developed as solutions. The multi-directional test algorithms are shown to provide more reliable results than previously developed algorithms. Numerical examples are given for demonstration.

Recommendations

• Nonlinear model validation using correlation tests
• Model validation tests for multivariable nonlinear models including neural networks
• Development of omni-directional correlation functions for nonlinear model validation
• MODEL VALIDATION OF SPATIOTEMPORAL SYSTEMS USING CORRELATION FUNCTION TESTS

Cited in

(13)

• Testing multiple equation systems for common nonlinear components
• Validity of the standard cross-correlation test for model structure validation
• Development of omni-directional correlation functions for nonlinear model validation
• Model selection approaches for non-linear system identification: a review
• A correlation test-based validity monitoring procedure for online detecting the quality of nonlinear adaptive noise cancellation
• Model validity tests for non-linear signal processing applications
• A set of novel correlation tests for nonlinear system variables
• MODEL VALIDATION OF SPATIOTEMPORAL SYSTEMS USING CORRELATION FUNCTION TESTS
• Modeling and validation of nonlinear feedback systems
• Nonlinear model validation using correlation tests
• VALIDATING IDENTIFIED NONLINEAR MODELS WITH CHAOTIC DYNAMICS
• Model validation tests for multivariable nonlinear models including neural networks
• On covariance function tests used in system identification

This page was built for publication: Multi-directional model validity tests for non-linear system identification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4485348)