Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Bending fatigue properties and prediction of asphalt mixtures with ultra-large aggregates
Abstract The laboratory fatigue test of asphalt mixture is generally carried out at a specific temperature, but previous studies have shown that the temperature range of flexible base is large, and the temperature has a significant effect on the performance of asphalt mixture. Therefore, the bending strength test was carried out to study the bending properties of ultra-large particle-size asphalt mixture (LSAM-50) at temperatures ranging from − 15°C to 75 °C, and a prediction model of bending parameters considering temperature was established. Then, the effects of temperature (−15°C to 35 °C) and stress level (0.3 to 0.9) on the fatigue life of LSAM-50 were studied by four-point bending fatigue test, and the experimental results were treated with Weibull distribution. Finally, four machine learning methods, including multiple linear regression method, sparrow search algorithm optimization back propagation (BP) neural networks, genetic algorithm optimization BP and enhanced whale optimization algorithm optimization BP, were used to achieve regression prediction of LSAM-50 equivalent fatigue life under 5% failure probability. The results show that the failure bending tensile strain, bending tensile strength, failure bending tensile modulus and temperature of LSAM-50 conform to the Boltzmann function. The strain and temperature are "S" type, while the strength, modulus and temperature are "inverted S". The average error of the prediction model is about 4.7%. The fatigue life decreases with the increase of temperature (or stress level). The input variables of the prediction model include temperature, stress level, stress, ratio of bending tensile strength to modulus, and the determination coefficient R 2 is greater than 0.9. The results of this study can provide reference for the design of asphalt mixture.
Highlights The bending properties of LSAM-50 over a wide temperature range were studied. The temperature-dependent model of LSAM-50 bending property was established. The effects of temperature and stress level on the fatigue of LSAM-50 were analyzed. The machine learning prediction model of LSAM-50 fatigue life was established.
Bending fatigue properties and prediction of asphalt mixtures with ultra-large aggregates
Abstract The laboratory fatigue test of asphalt mixture is generally carried out at a specific temperature, but previous studies have shown that the temperature range of flexible base is large, and the temperature has a significant effect on the performance of asphalt mixture. Therefore, the bending strength test was carried out to study the bending properties of ultra-large particle-size asphalt mixture (LSAM-50) at temperatures ranging from − 15°C to 75 °C, and a prediction model of bending parameters considering temperature was established. Then, the effects of temperature (−15°C to 35 °C) and stress level (0.3 to 0.9) on the fatigue life of LSAM-50 were studied by four-point bending fatigue test, and the experimental results were treated with Weibull distribution. Finally, four machine learning methods, including multiple linear regression method, sparrow search algorithm optimization back propagation (BP) neural networks, genetic algorithm optimization BP and enhanced whale optimization algorithm optimization BP, were used to achieve regression prediction of LSAM-50 equivalent fatigue life under 5% failure probability. The results show that the failure bending tensile strain, bending tensile strength, failure bending tensile modulus and temperature of LSAM-50 conform to the Boltzmann function. The strain and temperature are "S" type, while the strength, modulus and temperature are "inverted S". The average error of the prediction model is about 4.7%. The fatigue life decreases with the increase of temperature (or stress level). The input variables of the prediction model include temperature, stress level, stress, ratio of bending tensile strength to modulus, and the determination coefficient R 2 is greater than 0.9. The results of this study can provide reference for the design of asphalt mixture.
Highlights The bending properties of LSAM-50 over a wide temperature range were studied. The temperature-dependent model of LSAM-50 bending property was established. The effects of temperature and stress level on the fatigue of LSAM-50 were analyzed. The machine learning prediction model of LSAM-50 fatigue life was established.
Bending fatigue properties and prediction of asphalt mixtures with ultra-large aggregates
Tian, Tian (Autor:in) / Jiang, Yingjun (Autor:in) / Li, Sheng (Autor:in) / Nie, Chenliang (Autor:in) / Yi, Yong (Autor:in) / Zhang, Yu (Autor:in) / Deng, Changqing (Autor:in)
07.01.2024
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Fatigue Properties of Asphalt Mixtures Using Granite Aggregates
| Springer Verlag | 2024
The prediction of fatigue life of asphalt mixtures using four-point bending tests
| British Library Conference Proceedings | 2010
Waste Aggregates in Asphalt Mixtures
| Springer Verlag | 2023