System Identification for State Feedback Integral Observer Control of Polymer Plastic Extrusion

Abstract In this article, the system dynamics of a realistic plastic extrusion process is identified by least squares approximation associated with loss and covariance function criteria applied to a set of experimentally measured data of true single-screw extrusion processes. The system identification yields a third-order mathematical model of the process with a single-input/single-output nature. This modeling approach leads to a reliable and effective system model which facilitates the control design. In designing the control algorithm, the derived dynamic model is formulated in the state-space form to obtain the required Riccati equation. Then, an integral observer control methodology is performed. To reduce the computation time, a program which simulates deterministic model and controller design is proposed. In addition, a steady-state matrix based on the algebraic Riccati equations is utilized to design the controller gain matrices. Finally, the observer design methodology employing the observer chara...