Title :
An Experimental Investigation of Process Identification by Competitive Evolution
Author_Institution :
Cornell Aeronautical Laboratory, Buffalo, N.Y.
fDate :
6/1/1967 12:00:00 AM
Abstract :
The feasibility of using evolutionary techniques to construct a model for a given process was investigated. Programs were written to synthesize six competing models and to adjust, combine, and eliminate these according to their performance with respect to the actual process. Results show that although the use of evolutionary techniques is promising for identifying processing having nonanalytic properties, it is also, at the same time, very costly with respect to computer time.
Keywords :
Application software; Biology computing; Decision making; Equations; Evolution (biology); Least squares approximation; Performance analysis; Probability; State estimation; Transfer functions;
Journal_Title :
Systems Science and Cybernetics, IEEE Transactions on
DOI :
10.1109/TSSC.1967.300102