硫酸盐侵蚀混凝土中硫酸根离子分布的神经网络预测分析

Neural Network Prediction of Sulfate-ion Distribution in Concrete under Sulfate Corrosion

基于BP神经网络算法,以硫酸盐侵蚀混凝土试验数据为训练样本,建立了考虑工作环境条件、硫酸盐溶液浓度、循环次数和侵蚀时间等影响因素的硫酸盐侵蚀混凝土中硫酸根离子沿侵蚀深度分布的内推与外推预测模型.经验证,该模型具有良好的预测效果,为混凝土结构的耐久性研究提供了新的途径.

Based on BP neural network methodology, utilizing the experimental data of concrete under sulfate corrosion as the training samples, the models are proposed for inside and outside predicting the distribution of sulfate radical along corrosion depth in concrete under sulfate corrosion, which considering the influence factors of the working condition, the concentration of sulfate radical in solution, the repeat times and corrosion time. It is verified that the proposed model is of well forecasting ability, and a new method is provided for durability study of concrete structures.