Dynamic neural network model for identifying cumulative responses of soybean plant growth based on nitrogen fertilizer compositions

In this study, comparison between dynamic neural network and mathematical model was investigated for identifying cumulative responses of soybean plant growth. In this model, cumulative responses of stem diameter and plant height were used as an output factors and fertilizer compositions using nitrogent (N) were used as an input factors. Dynamic neural network was applied to explore and examine the number of data pattern on cumulative responses of soybean plant growth for acceptable identification, through simulation of a given model. The identification cumulative responses of soybean plant growth using dynamic neural network was resulting more performance compared with least square of a mathematical method. Dynamic neural network using time delay in back- propagation algorithm generated best performance in parameter number (n) =1, learning rate (lr) of 10, momentum constant (m) of 0.9 and the limit of error (err) 0.1 for identifying cumulative responses of stem diameter and plant height of soybean plant growth. The techniques obtained here can be applicable to a wide variety of identification problems in plant cultivation systems.