The control of a pH process using neural networks is examined. The neu
ral network as a universal approximator is used to good effect in this
nonlinear problem, as is shown in the simulation results. In the mode
lling task, the dynamics of the process was carefully examined to dete
rmine a suitable structure for the net. In particular, a multilayer ne
t consisting of two single hidden layers was constructed to reflect th
e Wiener model of the pH process. This led to much simpler training co
mpared to similar modelling attempts by other researchers. For the con
trol task, two schemes were simulated. In one approach, a net was used
to deal with the static nonlinearity to achieve control over a wide w
orking range. The dynamic controller used was the PID, with its parame
ters tuned on a relay auto-tuner. This control design was compared wit
h the strong acid equivalent method. In the second approach, a direct
model reference adaptive neural network control scheme was proposed. T
he training procedure uses the more efficient least squares algorithm
developed by Loh and Fong.