A real-time multiple-model based control and identification of a nonlinear process

This paper presents implementations of Real-Time Control and Identification based on Multiple-Modeling approach for an experimental thermal process. The thermal process is a nonlinear plant; therefore, based on variations of the input and disturbance, four local operating regimes are defined. Then linear local ARMAX models are identified and integrated into a NARMAX model by combining local models via proper validity and interpolation functions. Results of a single model and multiple model approach show superior performance of multiple modeling technique which is also more flexible. Moreover, the Real-Time control of the plant with four locally designed controllers is addressed. The platform used for the Real-Time implementation is Matlab/Simulink/Real-Time-Workshop with Visual C++ and Watcom compilers using a DAQ interface. The Real-Time application of the global controller based on multiple model approach proves excellent performance of this design in comparison to a single PID controller.