Work

Physics-Informed and Data-Driven Neural Networks for Modeling and Optimization of Residual Stresses in Dissimilar Material Systems