%K Identification (control systems); Pilot plants; Predictive control systems; State space methods, Bumpless transfer; Dynamic characteristics; Local state space; Manipulated variables; Multiple MPC; Multiple-modeling; Nonlinear; Nonlinear process, Model predictive control
%X This paper presents a practical approach of Multiple Model Predictive Control (MMPC) to deal with the nonlinearity of a process plant. The regulation of the nonlinear system over a wide range of operation is handled by a number of linear local MPCs, each is designed based on a local State Space model which describes the dynamic characteristics of the system in a certain level of operation. At each certain time of operation, only one local MPC is active while others are in standby mode and switching among the local MPCs is taken place when the system changes its working condition. In order to produce a 'bumpless transfer' during the switching, a Feedback of External Manipulated Variable (FEMV) is provided to the all local MPC controllers. The proposed method is applied to real time Gaseous Pilot Plant and results from the test show that the proposed MMPC improves the performance of nonlinear control system. © 2014 Swinburne University of Technology, Australia.
%O cited By 3; Conference of 6th International Conference on Modelling, Identification and Control, ICMIC 2014 ; Conference Date: 3 December 2014 Through 5 December 2014; Conference Code:110290
%L scholars6046
%J Proceedings of 2014 International Conference on Modelling, Identification and Control, ICMIC 2014
%D 2015
%R 10.1109/ICMIC.2014.7020728
%T A practical approach of control of real time nonlinear process plant using Multiple Model Predictive Control
%I Institute of Electrical and Electronics Engineers Inc.
%A T.H. Nguyen
%A I. Ismail
%A N.B. Saad
%A A. Faisal
%A M. Irfan
%P 59-64