A prediction method for salt frost resistance of carbon-glass mixed fiber concrete based on improved BP neural network

Low temperature and freeze-thaw can greatly reduce the service life of concrete, leading to serious safety hazards. Salt frost resistance is one of the important characteristics of concrete. Therefore, in order to more accurately evaluate the salt frost resistance of carbon glass hybrid fiber reinforced concrete. This article proposes a prediction method using measured data based on an improved BP neural network. Firstly, conduct freeze-thaw cycle tests on fiber reinforced concrete with different proportions. Studied the variation law of concrete quality loss rate and dynamic elastic modulus. Then, based on BP neural network and particle swarm optimization algorithm, a salt frost damage prediction model for carbon glass hybrid fiber reinforced concrete was established. Finally, conduct optimization based on the learning samples and compare and analyze the prediction errors of the model. The simulation results demonstrate that the proposed method has good accuracy and stability.