Method for predicting residual flexural capacity of RC beam considering high-temperature bursting of concrete

The invention discloses a method for predicting the residual flexural capacity of an RC beam by considering high-temperature burst of concrete, which comprises the following steps: establishing a refined analysis model of the RC beam by utilizing ABAQUS software and considering burst parameters and a steel bar-concrete bond slip constitutive relationship; taking different burst parameters as variables, constructing a sample database corresponding to the burst parameters and the residual flexural capacity, training by adopting a support vector machine, and establishing a prediction model of the residual flexural capacity of the RC beam after the high temperature; taking a prediction sample composed of different burst parameters as input data, and utilizing the prediction model constructed in the step 2 to obtain a prediction result of the residual flexural capacity of the RC beam after the high temperature; and comparing a prediction result with a simulation result of the residual bending resistance bearing capacity obtained by ABAQUS software, and checking the generalization performance of the prediction model. According to the invention, the feasibility of analyzing the high-temperature residual flexural capacity of the beam by establishing a prediction model through machine learning is verified, and help is provided for fire prevention of buildings.