摘要
考虑到冻融劈裂强度比(TSR)试验需要消耗大量的时间与物力,并且沥青混合料水稳定性影响因素与评价指标之间很难建立一个十分准确的数学模型,在分析选取影响沥青混合料水稳定性的因素后,通过BP神经网络模型对已有35组试验数据进行训练学习及检验,建立了TSR预测模型。结果表明:这种预测方法的最小相对误差为1.50%,最大相对误差为4.94%,预测精度较好,可用于TSR试验值的预测,为混合料水稳定性的初步判断与评价提供了一定借鉴。
Considering that the test on the freeze-thaw tensile strength ratio（TSR） requires a lot of time and material resources, and it is difficult to establish a very accurate mathematical model between the influencing factors and the evaluation indicators of the moisture susceptibility of asphalt mixture, after analyzing and selecting the factors that affect the moisture susceptibility of asphalt mixture, the BP neural network model was used to train and inspect 35 sets of test data, and the TSR prediction model was established. The results show that the minimum relative error of this prediction method is i. 50% and the maximum is 4. 94%. The great accuracy makes it capable of TSR test value prediction, which provides a reference for the preliminary judgment and evaluation of asphalt mixture＇s moisture susceptibility.
出处
《筑路机械与施工机械化》
北大核心
2017年第10期63-67,共5页
Road Machinery & Construction Mechanization
基金
江西省交通科技项目(2015B0050)
重庆市科学技术委员会社会民生科技创新专项项目(CSTC2016SHMSZX30005)
