Prediction of solubility of carbon dioxide in different polymers using support vector machine algorithm

This paper concerns with implementation of support vector machine algorithm for developing improved models capable of predicting the solubility of CO2 in five different polymers namely polystyrene (PS), poly vinyl acetate (PVAC), polypropylene (PP), poly butylene succinate-co-adipate (PBSA) and poly butylene succinate (PBS). Validity of the presented models has been evaluated by utilizing several statistical parameters. The predictions of the developed models for polymers of PS, PVAC, PP, PBSA, PBS are in excellent agreement with corresponding experimental data with the average absolute relative deviation percent (%AARD) equal to %0.151, %0.500, %1.381, %0.158, %0.239 and R2 values of greater than 0.999. Furthermore, the estimation capability of the proposed models has been compared to a well-known equation of state (EOS) as well as artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models. According to the results of comparative studies, it was found that the developed models are more robust, reliable and efficient than other existing techniques for improved analysis and design of polymer processing technology.