Abstract：The slurry pressure in cutting face of shield is the most important factor for maintaining the stability of the excavation face. In view of the unclear relationship between the slurry pressure in cutting face and shield tunneling parameters at present, a random forest prediction model is ptoposed based on Python. Based on the Qinghuayuan tunnel of JingZhang high-speed railway, shield geometry parameters, construction parameters and characteristic of slurry were considered. Random forest prediction model of slurry pressure in SPB shield was proposed of which the hyperparameters were adjusted through grid search method and 5-fold cross-validation method. The research results show that the random forest model has high accuracy and generalization for the prediction of slurry pressure in the shield of the composite stratum. The performance scores in the training and test sets are 0.963 and 0.946, respectively. The predictive ability of the model was also verified by evaluation parameters such as RMSE and R2. Finally, through importance analysis, the parameters that have a greater impact on the model prediction are air pressure, burial depth ratio, thrust and torque. This research can provide an effective method for the prediction of the slurry pressure in the excavation face of the shield tunnel under the same conditions which has important practical significance for the predictive control of the excavation face of mud-water shield.
罗维平 袁大军 金大龙 陆平. 基于随机森林的复合地层盾构#br#
切口泥水压力预测与分析[J]. 土木工程学报, 2020, 53(S1): 43-49.
Luo Weiping Yuan Dajun Jin Dalong Lu ping. Prediction and analysis of slurry pressure at the shield cut in #br#
composite strata based on random forest. 土木工程学报, 2020, 53(S1): 43-49.
