A Comparison Between GA and PSO Algorithms in Training ANN to Predict the Refractive Index of Binary Liquid Solutions

A total number of 1099 data points consisting of alcohol-alcohol, al- cohol-alkane, alkane-alkane, alcohol-amine, and acid-acid binary solu- tions were collected from scientific literature to develop an appropri- ate artificial neural network (ann), model. temperature, molecular weight of the pure components, mole fraction of one component, and the structural groups of the components were used as input param- eters of the network while the refractive index was selected as its output. the ann was optimized once by the genetic algorithm (ga) and once again by the particle swarm optimization algorithm (pso) in order to predict the refractive index of binary solutions. the op- timal topology of the ann-ga and ann-pso consisted of 13 and 16 neurons in the hidden layer, respectively. the results revealed that the ann optimized with pso had a better accuracy (mse=0.003441 for test data) compared to the ann optimized with ga (mse=0.005117 for test data).