In this article, a new simplistic way of predictive modeling of proces
s variables in nonlinear dynamic processes is introduced. This approac
h, which is semi-empirical, is demonstrated on a simulated continuous
stirred tank reactor. Model development uses a first-order-plus-dead-t
ime structure and only two Or three input changes for determining the
coefficients. This approach is evaluated for a variety of situations w
hich include measured output, unmeasured output, extrapolation beyond
the input range, various levels of dead time, various levels of measur
ement error, large dynamics, and various levels of nonlinear behavior.
In the situation of unmeasured output, the proposed approach is very
accurate and in the other cases it is extremely accurate and far super
ior to linear regression and artificial neural network models. (C) 199
8 Elsevier Science Ltd.