نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه مهندسی عمران، دانشکده فنی و مهندسی، دانشگاه قم
2 دانشیار، دانشکده فنی و مهندسی، دانشگاه قم
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
The non-linear behavior of reinforced concrete columns under cyclic loads reduces the possibility of using classical analysis methods to check their deformation capacity. Mathematical-analytical models based on basic concepts and principles of structural mechanics or empirical-statistical models based on nonlinear regression generally lack sufficient accuracy and speed in answering such problems. The main goal of this research is to investigate the possibility of using soft computing in investigating the lateral deformation capacity of reinforced concrete columns using experimental data. For this purpose, four models are presented by using an adaptive neural fuzzy inference system and artificial neural networks. More than 100 samples of reinforced concrete columns from the PEER database were used for testing the models. The main geometric and mechanical variables affecting the lateral deformation of the column are defined as model inputs. The lateral displacements corresponding to the crushing of the concrete cover of the column and the 20% decrease in the lateral resistance of the column have been used as bending failures and output of the models. Two models are coded with the Adaptive Neuro-Fuzzy Inference System (ANFIS) and two models are modeled with the Multi-Layer Perceptron (MLP) method. Comparing the error values of ANFIS models compared to MLP models shows the appropriate ability of ANFIS in predicting the behavior of reinforced concrete columns under cyclic lateral load.
کلیدواژهها [English]
K.-Y., et al., Deformation capacity of FRP retrofitted reinforced concrete columns with corroded reinforcing bars. Engineering Structures, 2022. 254: p. 113834.