Prediction of Punching Capacity of Slab‐Column Connections without Transverse Reinforcement Based on a Backpropagation Neural Network

Abstract

Punching shear failure of slab-column connections can cause the progressive collapse of a structure. In this study, a punching test database is first established. Then, based on the Levenberg–Marquardt (LM) algorithm and using the nonlinear function of the backpropagation neural network (BPNN), a prediction model of the punching capacity of slab-column connections without transverse reinforcement is established. Finally, the proposed model is compared with the formulas of the Chinese, American, and European standards using several methods. The statistical eigenvalue method shows that the BPNN model has the highest accuracy and the lowest dispersion. The defect point counting method shows that the BPNN model had the fewest total number of defects and was the safest and most economical. The influencing factor analysis suggests that factors in the BPNN model had the most reasonable influence on the punching bearing capacity of slab-column connections. Finally, the model is verified using a case study and the Matlab program. The results show that the average error of the formulas in the Chinese, American, and European standards are 21.08%, 30.21%, and 11.47%, respectively, higher than that of the BPNN model.

An Automated Essay Scoring Based on Neural Networks to Predict and Classify Competence of Examinees in Community Academy

“…The AES used BPNN to assist in predicting and classifying. The advantages BPNN has been widely applied, such as in the financial sector [15], civil engineering [16], wireless sensor networks [17], electricity [18]. There are several reasons for choosing AES based on BPNN, such as BPNN having accuracy and precision in making predictions [19], [20], [21].…”