Investigation of the Storage Stability of polymer modified bitumen containing nanoparticles and estimation of the Marshall Stability of asphalt mixtures using artificial neural network

Document Type : Original Article

Authors

1 Faculty of Civil Engineering, Semnan University, Semnan, Iran

2 Islamic Azad University

Abstract

Despite the many advantages of polymer modified bitumen, such as greater resistance to rutting and fatigue, during storage at high temperatures, the separation of the bitumen and polymer phase will occur and due to the complex behavior of asphalt mixtures, predicting the performance of asphalt mixtures is difficult. Therefore, this study aimed to create stability in polymer modified bitumen with nanoparticles and give a model for predicting the Marshall stability of asphalt mixtures using artificial neural networks. 18 types of bitumen samples are prepared with a combination of AC 60/70 bitumen with SBS polymer and aluminum oxide and magnesium oxide nanosized particles and storage stability test has been carried out. Marshall samples were made and the marshall stability of the asphalt mixture samples measured at 60 ° C. In the ANN model, the amount of bitumen, the amount of SBS, the amount of AL2O3 nanoparticles, the MgO nanoparticles, the sample weight, the sample height were the parameters for the input layer where the Marshall stability was the parameter for the output layer. The most appropriate algorithm and the number of neurons in the hidden layer are determined. Based on the experimental results and the neural network model was concluded that ANN can be used as an accurate method to predict the Marshall stability of polymer modified bitumen containing nanoparticles.

Keywords