A new hybrid genetic method for establishing a neural network model of the boiling sugar crystalization

The backpropagation and simplex are the classical search methods, so both are easily entrapped in a local optimum. To solve this problem, we develop a new hybrid genetic algorithm that is based on the probabilistic simplex and backpropagation. Moreover, we have established a neural network model of the boiling sugar crystallization. The simulation results demonstrate that the proposed method is very effective.