论文部分内容阅读
为了精确的预测混凝土的碳化深度,利用径向基网络良好的非线性逼近计算能力,建立了具有3个隐含层的混凝土碳化深度神经网络,通过对该神经网络的训练预测既有钢筋混凝土结构的碳化深度,结果表明:径向基网络的预测值完全满足精度要求,该方法可以应用于既有结构的混凝土碳化深度预测。
In order to accurately predict the depth of carbonation of concrete, a nonlinear carbonation depth neural network with three hidden layers is established by using the good nonlinear approximation ability of radial basis network. Through the training of the neural network, the existing reinforced concrete structure The results show that the predicted value of radial basis network can meet the accuracy requirement completely and this method can be applied to predict the depth of concrete carbonization of existing structures.