The biological treatment process in a wastewater treatment system is a very
complex process. The efficiency of the treatment is usually measured by la
boratory tests, which typically take five days. In this paper, a time-delay
neural network (TDNN) modeling method is proposed for predicting the treat
ment results. As the first step, a sensitivity analysis performed on a mult
ilayer perceptron (MLP) network model is used to reduce the input dimension
s of the model. Then a TDNN model is further used to improve the performanc
e of the original MLP network model. Subsequently, an on-line prediction an
d model-updating strategy is proposed and implemented. Simulations using in
dustrial process data show that the prediction accuracy can be improved by
the on-line model updating. (C) 1998 Elsevier Science Ltd. All rights reser
ved.