Title:
Data- and Machine Learning-Driven Approaches to Analyses of Complex Reinforced Concrete Structures
Author(s):
Abuoliem
Publication:
Web Session
Volume:
ws_S24_ Abuoliem.pdf
Issue:
Appears on pages(s):
Keywords:
DOI:
Date:
3/23/2024
Abstract:
Accurate analyses of the full spectrum of nonlinear damages of complex reinforced concrete (RC) structures subject to extreme loads have been at the center of infrastructure engineering research. Recently, an active movement has emerged to combine the finite element method (FEM) with the new technologies of machine learning (ML) and data science. The fusion of FEM-data-ML has a promising potential to offer a new angle to understand the complex damage behaviors of RC structures. Also, the fusion is computationally light, easy to evolve, and less dependent upon expensive experiments and upon model calibrations by human interventions. This presentation summarizes recent advances in FEM-data-ML fusion for analyses of complex RC structures. It covers statistical and ML-oriented methods, spans multiple length scales from meters to millimeters, and covers how ML is embedded into FEM for data-driven analyses of complex RC structures.