Optimization of EC Values of Nutrient Solution for Tomato Fruits Quality in Hydroponics System Using Artificial Neural Network and Genetic Algorithms

Herry Suhardiyanto (1,2,3Department of Agricultural Engineering, Bogor Agricultural University, Indonesia)
Chusnul Arif (1,2,3Department of Agricultural Engineering, Bogor Agricultural University, Indonesia)
Budi I. Setiawan (1,2,3Department of Agricultural Engineering, Bogor Agricultural University, Indonesia)

Journal of Mathematical and Fundamental Sciences
Vol. 41 No. 1 (2009)

Total soluble solids (TSS) and fruit fresh weight are two indicators to show the quality of tomato fruits. To gain high values of TSS and fruit fresh weight, it is important to consider the concentration of nutrient solution, which is commonly represented by Electrical Conductivity (EC) value. Generally, the increasing of EC value not only increases the number of TSS, but also decreases fruit fresh weight. Therefore, it is important to optimize the EC value for both indicators of quality of tomato fruits. The objective of this research is to optimize the EC value of nutrient solution on each generative stage using Artificial Neural Network (ANN) and Genetic Algorithms (GA). ANN was used to identify the relationship between different EC value treatments with TSS value and fruit fresh weight. GA was applied to determine the optimal EC value in generative growth, which is divided into three stages. Results showed that the optimal EC values in the flowering stage, the fruiting stage and the harvesting stage were 1.4 mS/cm, 10.2 mS/cm and 9.7 mS/cm, respectively.