Studies on on-line bioreactor identification. I. Theory.

An integrated approach is presented for the on-line estimation of the state of a biochemical reactor from presently attainable real-time measurements. Elemental and macroscopic balances are used for the determination of the total rate of growth and state-of-the-art estimation techniques are subsequently employed for the elimination of process and measurement noises and the estimation of state variables and unknown culture parameters. The proposed approach is very flexible in that as new sensors become available they can be easily incorporated within the present framework to estimate new variables or improve the accuray of the old ones. The method does not require any model for the growth kinetics and is very successful in accurately estimating the above variables in the presence of intense noise and under both steady-state and transient conditions. State estimates obtained by the presented method can be used for the development of adaptive optimal control schemes as well as for basic studies of the characteristic properties of microbial cultures.