Data-driven synchronization-avoiding algorithms in the explicit distributed structural analysis of soft tissue
چکیده
We propose a data-driven framework to increase the computational efficiency of explicit finite element method in structural analysis soft tissue. An encoder–decoder long short-term memory deep neural network is trained based on data produced by an explicit, distributed solver. leverage this predict synchronized displacements at shared nodes, minimizing amount communication between processors. perform extensive numerical experiments quantify accuracy and stability proposed synchronization-avoiding algorithm.