Errors-in-variables system identification using structural equation modeling

• David Kreiberg ^[1] ; Torsten Söderström ^[1] ; Fan Yang-Wallentin ^[1]
  1. [1] Uppsala University,Sweden
• Localización: Automatica: A journal of IFAC the International Federation of Automatic Control, ISSN 0005-1098, Vol. 66, 2016, págs. 218-230
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Errors-in-variables (EIV) identification refers to the problem of consistently estimating linear dynamic systems whose output and input variables are affected by additive noise. Various solutions have been presented for identifying such systems. In this study, EIV identification using Structural Equation Modeling (SEM) is considered. Two schemes for how EIV Single-Input Single-Output (SISO) systems can be formulated as SEMs are presented. The proposed formulations allow for quick implementation using standard SEM software. By simulation examples, it is shown that compared to existing procedures, here represented by the covariance matching (CM) approach, SEM-based estimation provide parameter estimates of similar quality.