Intelligent modelling of MIMO nonlinear dynamic process plants for predictive control purposes

Abstract

In this research, the input/output data of a MIMO nonlinear system are used to create intelligent models for nonlinear systems. Multi layer perceptrons and neuro-fuzzy networks are utilized for the intelligent models. To make these models suitable for the predictive control, a variety of subtle points should be considered. Recurrent models and subtractive clustering are used in this research, and a pre-processing is applied to the columns of the raw data. Then the prepared data are used to train models. A reliable checking process is also studied. A Catalytic Continuous Stirred Tank Reactor is used as a case study. A computer model is used to gather the input data rather than a real one. Finally, the simulation is successfully performed to indicate the capabilities of the intelligent modeling method as well as the importance of the design considerations offered in this paper.