Complex processes are often modelled using
input-output data from experimental tests. Regression
and neural network modelling techniques address this
problem to some extent and are being increasingly used
to develop optimisation or model-based control
algorithms. Unfortunately, the latter methods provide
no physical insight into the underlying structural
relationships inherent within the data. Genetic
Programming (GP) is currently finding application in
the modelling of processes from experimental data. The
nature of GP-based modelling is that solutions are
`evolved' from a set of potential solutions in an
environment which mimics Darwinian `survival of the
fittest'. GP performs symbolic regression, determining
both the structure and the complexity of the model
during its evolution. In this contribution two examples
are used to demonstrate the utility of the GP technique
as a process modelling tool. It is concluded that GP
techniques may have further applications in the
modelling and identification of complex processes from
experimental input-output data.
%0 Conference Paper
%1 mckay:1996:iipGP
%A McKay, B.
%A Willis, M. J.
%A Hiden, H. G.
%A Montague, G. A.
%A Barton, G. W.
%B Identification in Engineering Systems
%C Swansea, UK
%D 1996
%K algorithms, genetic programming
%T Identification of Industrial Processes using Genetic
Programming
%V 1
%X Complex processes are often modelled using
input-output data from experimental tests. Regression
and neural network modelling techniques address this
problem to some extent and are being increasingly used
to develop optimisation or model-based control
algorithms. Unfortunately, the latter methods provide
no physical insight into the underlying structural
relationships inherent within the data. Genetic
Programming (GP) is currently finding application in
the modelling of processes from experimental data. The
nature of GP-based modelling is that solutions are
`evolved' from a set of potential solutions in an
environment which mimics Darwinian `survival of the
fittest'. GP performs symbolic regression, determining
both the structure and the complexity of the model
during its evolution. In this contribution two examples
are used to demonstrate the utility of the GP technique
as a process modelling tool. It is concluded that GP
techniques may have further applications in the
modelling and identification of complex processes from
experimental input-output data.
@inproceedings{mckay:1996:iipGP,
abstract = {Complex processes are often modelled using
input-output data from experimental tests. Regression
and neural network modelling techniques address this
problem to some extent and are being increasingly used
to develop optimisation or model-based control
algorithms. Unfortunately, the latter methods provide
no physical insight into the underlying structural
relationships inherent within the data. Genetic
Programming (GP) is currently finding application in
the modelling of processes from experimental data. The
nature of GP-based modelling is that solutions are
`evolved' from a set of potential solutions in an
environment which mimics Darwinian `survival of the
fittest'. GP performs symbolic regression, determining
both the structure and the complexity of the model
during its evolution. In this contribution two examples
are used to demonstrate the utility of the GP technique
as a process modelling tool. It is concluded that GP
techniques may have further applications in the
modelling and identification of complex processes from
experimental input-output data.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Swansea, UK},
author = {McKay, B. and Willis, M. J. and Hiden, H. G. and Montague, G. A. and Barton, G. W.},
biburl = {https://www.bibsonomy.org/bibtex/21ccb2a78c7e9539e0b86a56123d8c956/brazovayeye},
booktitle = {Identification in Engineering Systems},
broken = {http://lorien.ncl.ac.uk/sorg/paper4.ps},
interhash = {8d73a923ee0fddb6419b8a5d5adb99ee},
intrahash = {1ccb2a78c7e9539e0b86a56123d8c956},
keywords = {algorithms, genetic programming},
month = {March},
notes = {MSWord postscript not compatible with unix
cited by \cite{yeun_2004_tec}},
size = {10 pages},
timestamp = {2008-06-19T17:46:38.000+0200},
title = {Identification of Industrial Processes using Genetic
Programming},
volume = 1,
year = 1996
}