Principal component analysis for greenhouse modelling

Authors:
Nathalie Pessel;Jean-Francois Balmat
Affiliations:
Laboratoire Systèmes Information Signal, Equipe COSI, Université du Sud-Toulon-Var, La Garde Cedex, France;Laboratoire Systèmes Information Signal, Equipe COSI, Université du Sud-Toulon-Var, La Garde Cedex, France
Venue:
WSEAS TRANSACTIONS on SYSTEMS
Year:
2008

Citing 3
Cited 3

Neural Networks for Modelling and Control of Dynamic Systems: A Practitioner's Handbook

Neural Networks for Modelling and Control of Dynamic Systems: A Practitioner's Handbook
Greenhouse modeling using neural networks

AIKED'07 Proceedings of the 6th Conference on 6th WSEAS Int. Conf. on Artificial Intelligence, Knowledge Engineering and Data Bases - Volume 6
Data mining for decision support in multiple-model system identification

SMO'06 Proceedings of the 6th WSEAS International Conference on Simulation, Modelling and Optimization

Computational methods used in the study of grafting 2-chloroethyl phosphonic acid on titania

WSEAS TRANSACTIONS on SYSTEMS
Systems dynamics of future urbanization and energy-related CO2 emissions in china

WSEAS TRANSACTIONS on SYSTEMS
A cumulant-based method for gait identification using accelerometer data with principal component analysis and support vector machine

WSEAS Transactions on Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a Principal Component Analysis (PCA) study for neuronal modelling of a complex system. The PCA transforms a set of correlated variables into a smaller set of uncorrelated variables without lose the original information. Thanks to this first stage it is possible to design a simplified structure of the model. The right choice of the architecture is crucial for the application of neural nets in process identification. The proposed study allows to validate the association of the PCA with neuronal model for a real multivariable process modelling: an experimental greenhouse. The object is to estimate the internal climate (temperature and hygrometry) by reducing the number of the input variables. Thus, we compare two different structures of neural networks. Several tests and results are presented and discussed.