In this paper, a damage detection methodology for steel frame structures under fire load using time-history acceleration and machine learning (ML) is proposed. A randomly created dataset by finite element analysis (FEA) is utilized to develop deep neural networks (DNNs). |