Experimental and ANN analysis of temperature effects on the permanent deformation properties of demolition wastes

Abstract

Abstract The aim of this research is to study the effect of temperature and stress levels on the permanent strain of blends of two types of recycled waste materials, namely recycled concrete aggregate (RCA) and reclaimed asphalt pavement (RAP). RAP was mixed with RCA in different percentages of up to 80% by dry mass. A three-stage repeated load triaxial (RLT) testing procedure was employed for evaluation of permanent strain of RCA/RAP blends in different stress paths and temperatures. The tests were conducted at different temperatures of 5 °C, 20 °C (room temperature), and 50 °C, to assess the behavior of blends in a wide range of temperatures. The addition of RAP was found to result in an increase of permanent strain of the blends. As the RAP content increased in the blends, the accumulation of permanent strain increased at 50 °C. The specimens with up to 40% RAP exhibited stable behavior at 50 °C in all stages of the test. With increasing RAP content, the permanent strain behavior of the specimens was found to improve for the same blends at 5 °C. Based on the results of the experiments, an artificial neural network (ANN) model was developed for prediction of permanent strain of blends and investigating the impact of the test variables. Effect of RAP content, temperature, deviator stress, and number of cycles was evaluated. Although the RAP content and temperature were found to have a significant effect on the permanent strain, the number of cycles in the RLT test was found to be the more predominant influencing factor.