RBFN Identification of an Industrial Polymerization Reactor Model

Abstract

Methods developed for radial basis function network (RBFN) identification are applied to a complex multiple-input, multiple-output (MIMO) simulation of a solution copolymerization reactor. For RBFN identification, k-means clustering and stepwise regression analysis are used. The practicality of applying these methods to large industrial identification problems is discussed, considering the restrictions of industrially practical input sequence design. The RBFN model has three inputs and two outputs, and the dimensionality of the identification problem poses some difficulties for nonlinear empirical model identification; specifically, the large amount of data required is a problem for plant testing and may cause computational difficulties for identification algorithms as well.

Data‐Based Modeling and Analysis of Bioprocesses: Some Real Experiences

“…The reactor was described by a complex multi‐input, multi‐output (MIMO) model. Radial basis function networks, based on the stepwise regression technique or the k ‐mean clustering algorithm, were robust and fast, even in the presence of high dimensional input vectors (Bomberger et al, 2001). In the radial basis function network, the centers and widths of the hidden layer have to be estimated from the input data.…”