摘要
为了对边坡安全系数进行预测,选用重度、内聚力、摩擦角、边坡角、边坡高度、孔隙压力比这6个特征,以相同的训练样本分别建立Lasso回归模型、Ridge回归模型、决策树模型、Xgboost模型,并且以相同的测试样本预测边坡的安全系数,通过比较各样本的预测值与真实值的差距、计算预测值与真实值的均方根误差评价各模型的预测性能,并且对各模型进行对比分析。结果表明Xgboost模型较其他模型更为精确,能分析边坡的稳定性,有一定的工程参考和实用价值。
In order to predict the safety factor of slope,this paper chooses six characteristics:gravity,cohesion,friction angle,slope angle,slope height and pore pressure ratio.Lasso regression model,Ridge regression model,decision tree model and Xgboost model are established with the same training sample,and the same test sample is used to predict the slope.The safety factor is used to evaluate the prediction performance of each model by comparing the difference between the predicted value and the real value of each sample,calculating the root mean square error between the predicted value and the real value,and comparing and analyzing each model.The results show that Xgboost model is more accurate than other models and can analyze the stability of slope.It has certain engineering reference and practical value.
作者
王杰
刘毅聪
刘祚秋
Wang Jie;Liu Yicong;Liu Zuoqiu(Department of Applied Mechanics and Engineering,School of Aeronautics and Astronautics,Sun Yat-sen University Guangzhou 510006,China)
出处
《广东土木与建筑》
2020年第7期55-58,63,共5页
Guangdong Architecture Civil Engineering
