Growth optimization of plant by means of the hybrid system of genetic algorithm and neural network

A new control technique for growth optimization of plant in hydroponics is proposed. In the method, physiological processes of the plant to environmental factors are firstly identified by using a neural network, and then optimal values of the environmental factors are determined through the prediction of the identified model by using a genetic algorithm. Here, we divided the growth process into 4 stages and tried to obtain optimal 4-step values of the nutrient concentration of the hydroponic solution which maximize the ratio of total leaf length to stem diameter (TLL/SD), which is a good indicator for plant growth, using this method. For the identification, multi-input (nutrient concentration and light intensity) and single-output (TLL/SD)-system was considered. This control technique permitted one to successfully identify the complex system and quickly search the optimal 4-step concentrations. The optimal values obtained here was effective for the actual growth control.