Compressive strength prediction of solidified laterite based on BP neural network
In order to explore the improvement effect of different dosages of metakaolin and lime mixed with residual soil of basaltic weathered soil,this experiment selected kaolin dosages of 0％,2％,4％,6％,and 8％,and lime dosages of 0％,2.5％,5.0％,7.5％,and 10.0％,simultaneously mixed with basalt residual laterite to prepare 25 groups of differently solidified red soils.An orthogonal test on the unconfined compressive strength of the soils was conducted for 28 days,and a neural network prediction model was established by MATLAB to predict the 28-day compressive strength of solidified laterite.The results show that the maximum prediction error of the model is 4.56％,with a coefficient of determination of 0.997.Furthermore,this method offers simplicity,higher efficiency,and greater accuracy over conventional regression analysis techniques,which enables the prediction of the compressive strength of solidified laterite with different consolidation materials and dosages,thereby improving experimental efficiency.
万方数据
维普
