Research on mechanical properties and mix proportion design of solid waste-based cemented paste backfill

Abstract

Solid waste filling is an important development direction for filling mining. This research proposes the preparation of solid waste-based cemented paste backfill (SWCPB) to fill the underground mined area by using waste rock and copper slag as the mixed aggregate and granulated blast furnace slag, steel slag, and desulfurization gypsum as raw materials of solid waste binder to reduce the environmental hazards caused by the industrial solid wastes. Based on the response surface method, experiments are designed to analyze the effect patterns of different factors on the mechanical properties of SWCPB. The formation mechanism of SWCPB mechanical properties is explored based on various microscopic analysis techniques. The mix proportions of SWCPB are designed based on machine learning. The results show that the strength of SWCPB increases and then decreases with the increase of the proportion of copper slag, and gradually increases with the increase of the steel slag dosage and binder-aggregate ratio. The alkaline environment provided by the steel slag causes the glass body in granulated blast furnace slag to depolymerize and react with desulfurization gypsum to produce hydration products, which are the main source of mechanical properties formation in SWCPB. The machine learning model constructed in this research paper has high prediction accuracy for the strength of backfill, which can achieve high-precision optimization of backfill mix proportion. The research results could guide the recycling of solid waste and the efficient use of filling mining technology.

Keywords

• Mechanical property
• Mix proportion
• Solid waste
• Cemented paste backfill
• Machine learning