Optimization of storage system of fruits using neural networks and genetic algorithms

Physiological processes of fruits are characterized by complexity and uncertainty. The present work is an attempt to apply an intelligent control technique, including neural networks and genetic algorithms, to the optimization of a storage process of fruits. The optimization problem is to determine the 4-step setpoints of relative humidity which minimize the objective function F(h) (h=relative humidity). F(h) is composed of two evaluation factors; the water loss and the degree of disease of fruits. At first, F(h) as affected by relative humidity is identified using neural networks and then the optimal values (4-step setpoints of relative humidity which minimize F(h)) are sought through simulation of the identified model using genetic algorithms. The genetic algorithm with high crossover and mutation rates allowed the optimal value to be quickly sought. The intelligent control technique proposed here was quite useful for the optimization of such complex systems as fruit storage system