Abstract: The artificial neural network (ANN)model for polyphenol extraction from apple pomace was studied and ANN model was optimized by using genetic algorithm (GA) in this paper. Results showed that the structure of network is of four neurons of input layer with one neuron of output layer, one hidden layer and nine neurons. The trained network has a high generalization and the correlation coefficient between simulating outputs of the BP network and the experiments result is 0.985. The optimumperformance, 16.9% greater than the greatest value in experiments, is obtained when alcohol concentration is 62% while ratio of alcohol solution volume to apple pomace weight is 14:1, temperature was 69.7 ℃, and treatment time 5.9 hours. The optimial process parameters can be obtained with ANN model to describe the relationship between process parameters and target and GA so as to optimize process parameters.

Key words: artificial neural network, genetic algorithm, polyphenols from apple pomace, optimization, model