Abstract
Concrete is widely used in civil engineering, and fiber materials are usually added to building concrete to enhance its durability. In order to study the freeze-thaw damage performance of different fiber reinforced concrete mixtures and accurately predict the degree of freeze-thaw damage of fiber reinforced concrete with different ratios under salt freezing conditions, this paper proposes an improved BP neural network-based method for predicting freeze-thaw damage in building concrete. Firstly, freeze-thaw cycle tests were conducted on fiber reinforced concrete with different proportions to study the variation patterns of concrete quality loss rate and dynamic elastic modulus; Then, based on BP neural network and particle swarm optimization algorithm, a concrete damage prediction model is established; Finally, conduct optimization based on the learning samples and compare and analyze the prediction errors of the model. The method proposed in this article has been verified to have good accuracy and stability.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call