The accuracy aspects of identification (with respect to both variance
and bias of estimates) and the role of filtering in closed-loop identi
fication is discussed in this paper. It is shown that the key differen
ce between closed-loop and open-loop identification is the existence o
f the sensitivity function. A closed-loop identification algorithm whi
ch asymptotically yields the same expressions as open-loop identificat
ion, in both variance and bias errors, is proposed. The proposed algor
ithm is evaluated by simulated examples as well as experiments perform
ed on a computer-interfaced pilot-scale process. (C) 1997 Elsevier Sci
ence Ltd.