Deep neural networks for large deformation of photo-thermo-pH responsive cationic gels

In this work, a model is developed to analyze homogeneous and inhomogeneous large deformation of photo-thermo-pH responsive cationic gels. Constitutive equations are achieved by considering the equilibrium thermodynamics of swelling gels through vari-ational method. Employing this model, coupling effects of light intensity, temperature and pH variations on large deformation of gels are analyzed. The simulation results are com-pared with available experimental data. Then deep neural networks are developed to ap-proximate solutions to equilibrium equations of inhomogeneous swelling of spherical shell structure gels. The volume phase transition temperature of the gels and their dependence on light intensity are also demonstrated. (c) 2021 Elsevier Inc. All rights reserved.

Automated summary

This work develops a model to analyze large deformation of photo-thermo-pH responsive cationic gels, considering the equilibrium thermodynamics of swelling gels to obtain constitutive equations. The coupling effects of light intensity, temperature and pH variations on gel deformation are analyzed, with simulation results compared to experimental data. Deep neural networks are also used to approximate solutions to equilibrium equations of inhomogeneous swelling of spherical shell structure gels, demonstrating the volume phase transition temperature and its dependence on light intensity.