混凝土裂缝存在会对大坝安全构成潜在威胁.提出基于循环神经网络模型的混凝土坝裂缝张开预测方法.根据实际工程背景,引入递归神经网络(Recurrent Neural Network,RNN)进行大坝裂缝建模,并利用门控循环单元(Gated Recurrent Unit,GRU)和长短期记忆(Long-Short Term Memory,LSTM)对混凝土大坝的裂缝宽度进行建模和测试.结果表明,提出的基于RNN的方法能够有效地预测混凝土坝的裂缝变化.
The prediction of concrete dam crack widening based on recurrent neural networks model
Concrete cracks often pose a potential threat to the safety of dams.The study proposes a concrete dam crack opening prediction method based on recurrent neural networks.Based on the actual engineering background,a Recurrent Neural Network(RNN)is introduced for dam crack modeling,and the crack width of concrete dams is modeled and tested using Gated Recurrent Units(GRU)and Long Short Term Memory(LSTM).The experimental results show that the proposed RNN based method can effectively predict the crack changes in concrete dams.
万方数据
维普
