Premium
Characterization and gain identification of time varying flow processes
Author(s) -
Mellichamp D. A.,
Coughanowr D. R.,
Koppel L. B.
Publication year - 1966
Publication title -
aiche journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.958
H-Index - 167
eISSN - 1547-5905
pISSN - 0001-1541
DOI - 10.1002/aic.690120116
Subject(s) - identifier , perturbation (astronomy) , control theory (sociology) , process (computing) , identification (biology) , computer science , physics , artificial intelligence , botany , control (management) , quantum mechanics , biology , programming language , operating system
A method for continuously estimating the gain of a flow process by sinusoidal perturbation is presented. The resulting output perturbation is correlated with a second sinusoid to generate periodically an estimate of the process gain. A method of implementing such an identifier on a small analog computer is described. The experimental testing of this identifier computer with both a real process (a p H regulating system) and with an analog computer simulation of the process is described. The results of identification tests with a nonstationary system are presented. From these results it is concluded that the identifier estimates the process gain satisfactorily, introducing a delay (equal to one‐half the period of identification) and making an effective sampling or clamping of the gain estimate (over each period of identification).