Biopolycaprolactone molecular weight prediction based on neural network technique in a batch reactor

Nowadays, biopolymer has been actively used in two important areas in our daily activities; packaging and medical devices. One of the important criteria in production of biopolymer is the quality of the polymer. Therefore, a method of controlling biopolymer quality (i.e. molecular weight) is certainly indispensable in this matter. Moreover, biopolymerization is a nonlinear process that requires a powerful tool to tackle the nonlinearity of the process. Neural network is a powerful tool especially in modeling nonlinear and intricate process. Therefore, neural network had been chosen the tool to tackle this task. This work presented a prediction of biopolymer quality using neural networks with bootstrap re-sampling method. Based on the results, neural network provides an alternative path in order to predict biopolymer molecular weight.