Nonlinear system identification: Comparison between PRBS and Random Gaussian perturbation on steam distillation pilot plant

Abstract

This paper is proposed to model the steam temperature on steam distillation pilot plant using system identification. Random Gaussian Signal (RGS) and Pseudo Random Binary Sequence (PRBS) have been implemented to this system to perturb the input of the process. The objective of using different perturbation signal is to study their capability to excite the nonlinearity behavior of system dynamic. The linear and nonlinear Auto Regressive with Exogenous Input (ARX) model structures is used to estimate and validate the temperature output model of steam distillation pilot plant. Both models will be compared to study the performance and flexibility. The validation test is performed by using auto-correlation function (ACF), cross-correlation function (CCF) and model fit.