Road Materials and Pavement Design ( IF 3.4 ) Pub Date : 2020-04-08 , DOI: 10.1080/14680629.2020.1748693 Hassan Ziari 1 , Amir Amini 2 , Ali Moniri 2 , Mahdi Habibpour 2
In this paper, the effectiveness of the group method of data handling (GMDH) and the adaptive neuro-fuzzy inference system (ANFIS) methods in modelling the fracture parameters of asphalt mixtures were studied. For this aim, the models were investigated on the fracture energy and J-integral results of hot mix asphalt in terms of temperature, RAP content and fibre content. It was found that the fibres have an outstanding effect on the fracture behaviour of asphalt mixtures especially at intermediate and high temperatures and can be considered as an alternative to enhance the fracture resistance of recycled asphalt mixtures. The fracture data of asphalt mixtures can be successfully modelled by the ANFIS method with a high level of correlation. The GMDH was unable to model the J-integral results, however, it had a fair correlation with the results of fracture energy.
中文翻译:
使用 GMDH 和 ANFIS 方法预测纤维增强高 RAP 沥青混合料的抗裂性
在本文中,研究了数据处理组方法(GMDH)和自适应神经模糊推理系统(ANFIS)方法在沥青混合料断裂参数建模中的有效性。为此,研究了热拌沥青在温度、RAP 含量和纤维含量方面的断裂能和 J 积分结果的模型。结果表明,纤维对沥青混合料的断裂行为具有显着影响,尤其是在中高温下,可被视为提高再生沥青混合料抗断裂性的替代方法。沥青混合料的断裂数据可以通过具有高度相关性的 ANFIS 方法成功建模。GMDH 无法对 J 积分结果进行建模,但是,