Greenhouse air temperature modelling

• Author(s): Cunha, José Boaventura ; Oliveira, Paulo ; Coelho, J.P.
• Date: 2004
• Persistent ID: http://hdl.handle.net/10198/2781
• Origin: Biblioteca Digital do IPB
• Subject(s): Greenhouse climate models; Neural networks; Parameter estimation; Temperature control
• Document Type: Conference Object
• Language: English

Description

This paper describes two implementation approaches for modelling the air temperature of an automated greenhouse located in the campus of the University of Trás-os- Montes e Alto Douro. Linear models, based in the discretization of the heat transfer physical laws, and non-linear neural networks models are used. These models are describes as functions of the outside climate and control actions performed for heating and cooling. Results are presented to illustrate the performance of each model in the simulation and prediction of the greenhouse air temperature. The data used to compute the simulation models was collected with a PC-based acquisition and control system using a sampling time interval of 1 minute.