鉴于新拌混凝土流变参数与硫酸盐侵蚀劣化指标之间存在的非线性映射关系,基于新拌混凝土流变参数,利用BP神经网络建立了硫酸盐干湿循环侵蚀后混凝土抗压强度损失率、质量损失率及相对动弹性模量的预测模型,预测了混凝土受硫酸盐侵蚀程度。结果表明,BP神经网络模型预测的硫酸盐侵蚀结果与实测结果吻合较好,且预测误差较小,说明BP神经网络模型对混凝土受硫酸盐侵蚀后的劣化程度具有良好的预测效果。 In view of the nonlinear mapping relationship between the rheological parameters of the fresh concrete and the degradation index of the sulfate erosion, the compressive strength loss rate of the concrete after the sulfate wetting and wetting cycle was established by BP neural network based on the rheological parameters of the fresh concrete. Mass loss rate and the relative dynamic elastic modulus of the prediction model to predict the extent of concrete sulfate attack. The results show that the BP neural network model predicts that the sulfate erosion results are in good agreement with the measured results and the prediction errors are small, which shows that the BP neural network model has a good predictive effect on the deterioration of concrete after sulfate attack.