Predicting the Compressive Strength of Cemented Tailings Filling Body Based on GA-BP

Abstract

Taking the filling tailings of Gangli iron mine as the experimental slurry, this paper use the orthogonal design method to design the experimental scheme to study the influencing rules of cement-sand ratio, mass concentration and curing period to the compressive strength. Use the variance analysis to test the results and the adaptive neuro-fuzzy inference system to establish a prediction model of compressive strength. The results showed that the cement-sand ratio was the most important influence factors on compressive strength and those factors have similar influence. The GA-BP model prediction has high accuracy rate of 95.1% and provides an effective model and method for prediction of compressive strength of cemented tailings backfill.