摘要
针对贵州省高速公路中的大量类土质边坡评价问题,提出了基于朴素贝叶斯方法的类土质边坡稳定性评价方法。首先,通过对贵州省948个既有公路边坡进行分析,选定坡高、坡率、坡后角、黏聚力、内摩擦角和支护措施6个参数建立评价体系;然后,利用FLAC^(3D)软件建立4800条工况数据库作为训练集;基于朴素贝叶斯分类模型的类土质边坡稳定评价方法进行了评价,并采用蒙特卡洛交叉验证法验证;最后,利用MATLAB软件设计评价程序,并在厦蓉高速某边坡评价中试用。结果表明:朴素贝叶斯分类模型的准确率为91.68%,评价效果良好;实例分析结果显示稳定概率为59.42%,与数值模拟结果一致,验证了该方法的正确性。
Aiming at stability evaluation problems of a large number of soil-like slopes in Guizhou expressway construction,a soil-like slope stability evaluation method based on Naive Bayes was proposed.Firstly,based on the analysis of 948existing expressway slopes in Guizhou Province,the evaluation system was established by selecting six parameters including slope height,slope ratio,back slope angle,cohesion,internal friction angle and supporting measures.Then,FLAC^(3D) was used to establish 4800working condition database as training set.The stability evaluation method of soil-like slope based on Naive Bayes classification model was used for evaluation,and Monte Carlo cross validation method was used for verification.Finally,the evaluation program was designed by MATLAB software and used in the evaluation of a slope of Xia Rong Expressway.The results show that the accuracy of naive Baye classification model is 91.68%,and the evaluation effect is good.The results of case analysis show that the stability probability is 59.42%,which is consistent with the numerical simulation results,which verifies the correctness of the method.
作者
龚美
朱生龙
李文娟
王大兵
周苏华
GONG Mei;ZHU Shenglong;LI Wenjuan;WANG Dabing;ZHOU Suhua(Guizhou Traffic Construction Quality Supervision,Guiyang,Guizhou 550008,China;College of Civil Engineering,Central South University,Changsha,Hunan 410075,China;College of Civil Engineering,Hunan University,Changsha,Hunan 410082,China)
出处
《矿业研究与开发》
CAS
北大核心
2021年第12期41-47,共7页
Mining Research and Development
基金
国家自然科学基金项目(51708199)
贵州省科技支撑计划项目(2020-4Y047,2018-123-040)
贵州省交通运输厅科技项目(2017-143-054).
