Abstract

A compact and efficient model that is capable of approximating both the linear and nonlinear components of the process is in high demand. In this paper, a novel black-box modeling technique <em>viz.</em> Wiener type Laguerre-Wavelet model is proposed. The Laguerre-Wavelet model has the capability to approximate a function with moderate/reasonable number of data with appreciable approximation accuracy. The Laguerre filter is used to approximate the linear dynamic components of the process, whereas wavelet structure is used for the static nonlinear components. The ability of wavelets to approximate any square-integrable function to any arbitrary precision by input-output mappings is utilised for the nonlinear approximation following a modified single scaling method. The performance efficiency of the proposed Wiener type model structure, Laguerre-Wavelet model, is demonstrated using simulation case study on a continuous bioreactor.

Recommended Citation

Aadaleesan, P. and Saha, Prabirkumar (2008) "Nonlinear System Identification Using Laguerre Wavelet Models," Chemical Product and Process Modeling: Vol. 3 : Iss. 2, Article 3.
DOI: 10.2202/1934-2659.1136
Available at: http://www.bepress.com/cppm/vol3/iss2/3