Skip to content
Publicly Available Published by De Gruyter August 31, 2013

Prediction of the polarity parameter π for the radical derived from monomer

  • Xinliang Yu EMAIL logo , Wenhao Yu , Bing Yi EMAIL logo and Xueye Wang
From the journal e-Polymers

Abstract

An artificial neural network (ANN) model was successfully developed for the modelling and prediction of the polarity parameter π used in the revised patterns scheme for the prediction of monomers reactivity ratios in radical polymerizations. Four quantum chemical descriptors based on density functional theory (DFT) calculations were used to develop the ANN model. The optimal condition of the neural network was obtained by adjusting various parameters by trial-and-error. Simulated with the final optimum BP neural network 4-4-1, the results show that the predicted parameter π values are in good agreement with the experimental ones, with the root mean square (rms) errors being 0.053 (R=0.960) for the training set and 0.070 (R=0.942) for the test set. The ANN model has better statistic quality than the MLR model, which indicates there are nonlinear relationships between these quantum chemical descriptors and the parameter π.

Published Online: 2013-8-31
Published in Print: 2009-12-1

© 2013 by Walter de Gruyter GmbH & Co.

Downloaded on 28.3.2024 from https://www.degruyter.com/document/doi/10.1515/epoly.2009.9.1.1562/html
Scroll to top button