不同层位沥青路面温度预估模型

Prediction model of temperature in different layers of asphalt pavement

为了明确山东地区沥青路面内部温度分布规律,以指导沥青路面设计,开展以路表温度及气候条件预测沥青路面内部温度的研究,以山东滨德(滨州—德州)高速公路为依托,通过对试验路段不同天气状况下沥青路面不同深度处温度和该区气象数据测量。采用统计学方法分析不同气象条件下温度数据,并建立适用于沥青面层和柔性沥青基层的2种温度预估模型,并将该模型、BELLS3温度模型预估结果与实测温度进行对比,验证提出的模型的精度。研究结果表明:沥青路面的温度分布与外界气温关系密切,且距路表面越近,其敏感性越高;LSPM(大粒径透水性沥青混合料)柔性基层主要接受来自面层的热量传递,使得不同深度沥青路面结构的温度时间分布规律存在显著差异,面层温度时间分布变化具有明显周期性,呈近似正弦函数曲线;LSPM柔性基层温度随时间变化略平缓;在±1℃误差范围内,提出的模型温度预估结果准确性达到64.1%,而BELLS3温度模型仅为54.8%,提出的模型具有较高的预测精度和实用性。

In order to define the internal temperature distribution law of asphalt pavement in Shandong district, this paper carried out the study by surface temperature of asphalt pavement and climate condition to predict the internal temperature of asphalt pavement, so as to further guide the asphalt pavement design. Taken Binzhou to Dezhou Freeway in Shandong as research object, the pavement temperature at different depths of asphalt pavement and the hourly climate data were gathered during different weather conditions. Temperature data of different weather conditions were analyzed by statistics method, and two kinds of temperature prediction models suitable for asphalt pavement and flexible base asphalt pavement were established. Prediction results of this model and BELLS3 temperature model were compared with the measured temperature to verify the accuracy of the proposed model. The results show that the asphalt temperature has a close relationship with air temperature, and it is more sensitive when the depth is close to pavement surface. The temperature distribution of different depths in asphalt pavement structure present significant differences, due to LSPM （large stone porous asphalt mixes） flexible base which mainly received heat from the surface of asphalt pavement. The temperature distribution of asphalt pavement appears obvious periodicity and approximate sine curve. The temperature distribution of LSPM flexible base is slightly flat with the change of time. Within the error range of ±1℃, the temperature prediction accuracy of the proposed model is 64.1 %, while the accuracy of BELLS3 temperature model is only 54. 8%, which indicates that the proposed temperature prediction model has higher prediction accuracy and practicability. 1 tab, 10 figs, 20 refs.