Isaías González Pérez, Antonio José Calderón Godoy
Greenhouses are multivariable and nonlinear systems with high degree of complexity, so it is hard to build models that represent the whole dynamics of the system. This paper presents models of greenhouse climate based on neural networks. The models predict inside air temperature and relative humidity in the greenhouse as a function of the variables used as input for the network, as outside temperature, relative humidity, solar radiation, etc., and the actuators state signals, as window opening and others. Data sets used for modelling have been measured with real red pepper plants inside the greenhouse. The developed models are described and the achieved results are reported.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados