Symbolic machine learning improved MCFT model for punching shear resistance of FRP-reinforced concrete slabs

Fiber reinforced polymer (FRP)-reinforced concrete slabs, an extension of reinforced concrete (RC) slabs leveraged for resisting environment corrosion, are susceptible to punching shear failure due to the lower elasticity modulus of FRP reinforcement. To estimate the punching shear resistance accurately, there are two types of models (e.g., white box and black-box models) proposed based on theoretical derivations and machine learning methods. However, these two types of models are considered as independent of each other. In this study, a hybrid model (e.g., grey-box model) derived from modified compression field theory (MCFT) is proposed by this paper, in which the performance is improved by a machine-learning-aided approach (genetic programming). In order to exploit the performance of machine learning, a database containing 154 experimental data is established and used for fitting the correction equations. Iterating the population containing 300 tree-based individuals in 300 times, a correction equation with simple format is obtained, which performs well in performance improvement of the basic model derived from MCFT. Herein, the influential factors involved in the correction equation comply with the sorting in order of the importance quantified by extreme gradient boosting (XGBoost) and shapley additive explanation (SHAP). Combining the correction equation with the basic model derived from MCFT, a symbolic regression MCFT (SR-MCFT) model is established, which performs better prediction performance than other five empirical models.