Degradation prediction of recycled aggregate concrete under sulphate wetting–drying cycles using BP neural network

Abstract

Recycling construction and demolition wastes (C&DW) and industrial by-products as construction materials is an effective way to solve the serious environmental problems caused by these wastes. In this paper, the performance degradation of recycled aggregate concrete (RAC) containing fly ash (FA) under sulphate wetting–drying cycles was investigated by considering the recycled coarse aggregate (RCA) incorporation ratio, water-to-binder (W/B) ratio, and FA incorporation ratio. The mass loss rate (Km) and corrosion resistance coefficient of compressive strength (Kf) were considered for the analysis. The results show that as the number of sulphate wetting–drying cycle increased, the increase in RCA incorporation ratio, W/B ratio and FA incorporation ratio led to a gradual decrease in the Km of RAC. A substantially linear relationship could be observed between Kf and the three influencing factors. In contract, fly ash incorporation ratio had a positive effect on Kf. Finally, based on the back propagation (BP) neural network, a novel model considering the three influencing factors is developed to predict the sulphate resistance of RCA under wetting–drying cycles. In conclusion, RAC containing fly ash can both promote the C&DW recycling and give excellent sulphate resistance.