人工神经网络的盐害侵蚀混凝土强度预测

Strength prediction of concrete corroded by salt based on artificial neural network

为了分析和预测盐害侵蚀对混凝土强度的影响程度,利用BP人工神经网络进行分析,通过计算方法的优化和样本的训练,对隐含层和各隐含单元多次试取,最优选取trainlm训练函数,建立了盐害预测的人工神经网络系统.解析结果表明,混凝土试件抗压强度预测值和试验实测值的相对误差较小,建立的人工神经网络模型具有较高的预测精度.

To predict the strength of concrete corroded by salt, the BP-artificial neural network is used to analyze salt corrosion damage. A BP-artificial network model is established by optimizing the learning arithmetic, training the net with specimens and choosing the trainlm function as the optimal function. The analysis results show that the relative error between the predicted result and the measured result is slight for concrete specimens, which indicates that the established artificial network model has high prediction precision.