Residual strength of concrete subjected to fatigue based on machine learning technique

Abstract

Abstract This study utilizes the machine learning technique to solve the complex fatigue problem of concrete materials. To this end, several learning algorithms were addressed including the random forest (RF), support vector machine (SVM), and artificial neural networks (ANNs) models. Extensive experimental data were collected from literature to train the machine learning models for estimating the maximum number of cycles at failure (i.e., the so‐called fatigue life). A machine learning model providing the best correlation was chosen through verifications. On this basis, a strength degradation model of concrete under fatigue loading was finally proposed to evaluate the residual strength of concrete after fatigue damage, which is a key factor in determining the remaining service life of concrete structures.