In this paper, a novel and simple learning control strategy based on using
a bounded nonlinear controller for process systems with hard input constrai
nts is proposed. To enable the bounded nonlinear controller to learn to con
trol a changing plant by merely observing the process output errors, a simp
le learning algorithm for parameter updating is derived based on the Lyapun
ov stability theorem. The learning scheme is easy to implement, and does no
t require any a priori process knowledge except the system output response
direction. For demonstrating the effectiveness and applicability of the lea
rning control strategy, the control of a once-through boiler, as well as an
open-loop unstable continuously stirred tank reactor (CSTR), were investig
ated. Furthermore, extensive comparisons of the proposed scheme with the co
nventional PI controller and with some existing model-free intelligent cont
rollers were also performed. Due to significant features of simple structur
e, efficient algorithm and good performance, the proposed learning control
strategy appears to be a promising and practical approach to the intelligen
t control of process systems subject to hard input constraints. (C) 1999 El
sevier Science Ltd. All rights reserved.