Paper Information

Journal:   IRANIAN JOURNAL OF SOIL AND WATERS SCIENCES   2004 , Volume 18 , Number 2; Page(s) 0 To 0.
 
Paper: 

ESTIMATION OF GREENHOUSE CUCUMBER TRANSPIRATION USING ARTIFICIAL NEURAL NETWORK

 
 
Author(s):  FATHI P., KOUCHAKZADEH M.
 
* 
 
Abstract: 
Exact estimation of plant transpiration is an important factor in optimal management of greenhouse in order to achieve high quality and yield. Weighing lysimeters and water balance are two most accurate methods for measuring transpiration. Beside the high costs and time consuming aspects, the methods require careful experiments to be carried out, which them make them somewhat impractical and uneconomical. Moreover, results of the measurements are only valid for the specified greenhouse and are not transferable. Since many researches have been conducted in order to relate transpiration and climatic parameters in greenhouse, efforts have been made to fit a curve on the experimental data. Usually, this relation has been assumed to be linear, an unrealistic assumption that causes errors in the estimates. The objective of this research was to establish artificial neural network for transpiration estimation of greenhouse cucumber. In this paper after defining the artificial neural network specifications, a multilayer perception network with error back-propagation training method, was developed to fit a nonlinear relation to plant transpiration and climatic parameters within greenhouse and this physiologic characteristic of the plant (transpiration) was estimated. Results showed good agreement between the estimated values by the artificial neural network and the measured data and the errors were less than the errors of the field methods.
 
Keyword(s): ARTIFICIAL NEURAL NETWORK, GREENHOUSE CUCUMBER, INTELLIGENT ESTIMATION, RANSPIRATION
 
References: 
  • ندارد
 
  Yearly Visit 51
 
Latest on Blog
Enter SID Blog