ISHS Acta Horticulturae 406: II IFAC/ISHS Workshop :Mathematical & Control Applications in Agriculture & Horticulture
OPTIMAL CONTROL OF PLANT GROWTH IN HYDROPONICS USING NEURAL NETWORKS AND GENETIC ALGORITHMS |
| Authors: | T. Morimoto, Y. Hashimoto |
| Abstract:
In the cultivation of fruit vegetables, well-balanced growth between vegetative growth and reproductive growth is important to obtain the better quality of fruits. We defined the ratio (TLL/SD) of total leaf length (TLL) to stem diameter (SD) as an indicator for the well-balanced growth only in the seedling stage. In hydroponics, nutrient concentration of the solution is one of the most important manipulating factors for adjusting TLL/SD. This paper presents the application of genetic algorithms and neural networks to the optimal control of TLL/SD of tomato plants during the seedling stage in hydroponics. The control input is the nutrient concentration of the solution. We first divided the seedling stage into four stages and then tried to obtain the 4-step setpoints of nutrient concentration which maximize TLL/SD. A three layer neural network was markedly effective for identifying the process model of TLL/SD to nutrient concentration. Furthermore, the genetic algorithm ensured that the optimal value was obtained quickly from the neural network model simulation. The optimal 4-step setpoints of nutrient concentration obtained here were useful for the actual control of plant growth. Thus, this intelligent control technique based on neural networks and genetic algorithms was shown to be quite useful for the control of such complex systems as plant growth systems. | |
Download Adobe Acrobat Reader (free software to read PDF files) | |
