An optimal IV technique for identifying continuous-time transfer function model of multiple input systems

Hugues Garnier (1) , Marion Gilson (1) , Peter C. Young (2) , Eric Huselstein (1)

1 CRAN - Centre de Recherche en Automatique de Nancy
2 CRES - Centre for Research on Environmental Systems and Statistics

An instrumental variable method for continuous-time model identification is proposed for multiple input single output systems where the characteristic polynomials of the transfer functions associated with each input are not constrained to be identical. An associated model order determination procedure is shown to be reasonably successful. Monte Carlo simulation analysis are used to demonstrate the properties and general robustness of the model order selection and parameter estimation schemes. The results obtained to model a winding process and an industrial binary distillation column illustrate the practical applicability of the proposed identification scheme.

Mots clés: distillation column; continuous-time systems; identification; instrumental variable; linear differential equation; multiple input systems; parameter estimation; sampled data; winding process

Domaines: Automatique / Robotique

Hugues Garnier, Marion Gilson, Peter C. Young, Eric Huselstein. An optimal IV technique for identifying continuous-time transfer function model of multiple input systems. Control Engineering Practice, 2007, 15 (4), pp.471-486. ⟨10.1016/j.conengprac.2006.09.004⟩. ⟨hal-00151323⟩