Modelling and control of a continuous frying process: A simulation study. Part I: Dynamic analysis and system identification
Academic Article
Overview
Identity
Additional Document Info
View All
Overview
abstract
The dynamics of a continuous frying process making a snack product were modelled and a controller developed to automatically control product quality attributes on-line. The process was modelled using X (exogenous input), ARX (autoregressive with exogenous input), and ARMAX (autoregressive moving average with exogenous input) model structures. For control purposes, a Generalized Predictive Controller (GPC) was used. The dynamics of the process were analysed. Colour and oil content showed the largest and fastest responses to input changes and colour and moisture content were highly correlated. Due to the high interaction among control loops, a multivariable control scheme was suggested. The X, ARX, and ARMAX models with the lowest Aikaike's Information Criteria (AIC) values were chosen as the best ones to predict the process dynamics. The process was modelled as a MIMO (multiple-input multiple-output) and as a MISO (multiple-input single-output). The inputs were the oil temperature, C, the submerger conveyor speed, %, and the takeout conveyor speed, %. The outputs, or product quality attributes, were the colour b, the oil content, % wb, and the moisture content, % wb.
