page loader
  1. Trang chủ
  2. Bài báo khoa học
  3. Chi tiết bài báo
Investigating the Behavior of Steel Flush Endplate Connections at Elevated Temperatures Using FEM and ANN
Tác giả: Viet-Linh Tran
258    0
International Journal of Steel Structures
Quyển: 22     Trang: 1433–1451
Đường link/DOI: https://doi.org/10.1007/s13296-022-00650-x
Năm xuất bản: 8/2022
Tóm tắt
This paper investigates the moment-rotation (M-θ) behavior of flush endplate (FEP) connections at elevated temperatures using the finite element (FE) method and an artificial neural network (ANN). Firstly, a three-dimensional nonlinear FE model of flush endplate connection is carried out and verified with the tests conducted by others using ABAQUS. Then, an extensive database is created by varying several parameters (i.e., the endplate thickness, the bolt row distance, the pitch of bolts, the gage distance, the outer edge distance, the number of bolt rows, the bolt diameter, and the material properties) to get insight into the influences of each parameter on the connection behaviors at elevated temperatures. Additionally, a simple and accurate model with two shape parameters for the M-θ relationship of semi-rigid flush endplate connections at elevated temperatures is proposed based on this database. Accordingly, two shape parameters and the ultimate moment (Mu) of the model are determined using the ANN model. Finally, the performance of the proposed model is verified and has a good agreement with various test data.
Từ khóa
Artificial neural network; Elevated temperatures; Flush end plate connection; Moment-rotation model; Ultimate moment.
Cùng tác giả
Ensemble machine learning-based models for estimating the transfer length of strands in PSC beamsInnovative formulas for reinforcing bar bonding failure stress of tension lap splice using ANN and TLBOPrediction of the ultimate axial load of circular concrete-filled stainless steel tubular columns using machine learning approachesPatch loading resistance prediction of steel plate girders using a deep artificial neural network and an interior-point algorithmNovel hybrid WOA-GBM model for patch loading resistance prediction of longitudinally stiffened steel plate girdersA new framework for damage detection of steel frames using burg autoregressive and stacked autoencoder-based deep neural networkRevealing the nonlinear behavior of steel flush endplate connections using ANN-based hybrid modelsBuckling resistance of axially loaded square concrete-filled double steel tubular columnsRapid prediction of the ultimate moment of flush endplate connections at elevated temperatures through an artificial neural networkComputational analysis of axially loaded thin-walled rectangular concrete-filled stainless steel tubular short columns incorporating local buckling effectsAxial compressive behavior of circular concrete-filled double steel tubular short columnsEvaluation of Seismic Site Amplification Using 1D Site Response Analyses at Ba Dinh Square Area, VietnamMachine Learning Models for Predicting Shear Strength and Identifying Failure Modes of Rectangular RC ColumnsApplication of ANN in predicting ACC of SCFST columnA new empirical formula for prediction of the axial compression capacity of CCFT columnsMoment-rotation-temperature model of semi-rigid cruciform flush endplate connection in firePractical artificial neural network tool for predicting the axial compression capacity of circular concrete-filled steel tube columns with ultra-high-strength concrete