DEVELOPMENT OF A FIXED-GPC CONTROLLER FOR A FOOD EXTRUDER BASED ON PRODUCT QUALITY ATTRIBUTES Part I: System Identification

Institution of Chemical Engineers 2005 - all rights reserved Stochastic models based on product quality attributes (PQA) variables, collet colour-b (CB) and collet moisture content (CM) were developed and validated for a food extrusion process. The experimental data were used to describe the structure (delay, order) for each form and to determine the coefficients for all models. Both multiple-input-single-output (MISO) and multi-input-multi-output (MIMO) modes were studied. Models were tested using alternate experimental data to confirm the validity of the models. Simplified (2nd order) ARX structures were found to adequately describe the process. X models required more parameters, which although unsuitable for on-line adaptation are acceptable for fixed controllers.

DEVELOPMENT OF A FIXED-GPC CONTROLLER FOR A FOOD EXTRUDER BASED ON PRODUCT QUALITY ATTRIBUTES Part I: System Identification Academic Article