Structural optimization of PNIPAM-derived thermoresponsive polymers: a computational approach employing artificial neural networks and genetic algorithms

Abstract

In this study, artificial neural networks (ANNs) and genetic algorithms (GAs) are employed together to design optimized polymeric structures with superior cloud points. The database from a previous study of polymer synthesis with thermoresponsive polymers was used to create ANN-based models, which enabled the formulation and solution of the inverse problem using the GA. The regressors, with an average RMSE of less than 0.7 ºC, were used in the polymer evolution process over 20 generations. Mutation and selection operations led to the creation of 10 novel hybrid macromolecules with an average cloud point of 80 ºC. Furthermore, the special roles of some chemical groups are recognized and favor the structural mapping of PNIPAM-based materials. The computational approach presented here demonstrates that it is a promising tool in the development of new materials.