基于RF-NSGA-Ⅱ的盾构施工地表沉降安全控制多目标优化  

Multi-objective optimization of surface settlement safety control during shield construction based on RF-NSGA-Ⅱ

作　　者：吴贤国[1] 冯宗宝 刘俊[1] 王雷[1] 陈虹宇[2] 李昕懿 WU Xianguo;FENG Zongbao;LIU Jun;WANG Lei;CHEN Hongyu;LI Xinyi(School of Civil and Hydraulic Engineering,Huazhong University of Science and Technology,Wuhan Hubei 430074,China;School of Civil Engineering and Environment,Nanyang Technological University,Singapore 639798)

机构地区：[1]华中科技大学土木与水利工程学院,湖北武汉430074 [2]南洋理工大学土木工程与环境学院,新加坡639798

出　　处：《中国安全科学学报》2022年第8期45-51,共7页China Safety Science Journal

基　　金：国家自然科学基金资助(51778262,71571078,51308240)。

摘　　要：为有效调整盾构施工参数,指导盾构安全高效施工,建立随机森林(RF)与非支配排序遗传算法(NSGA-Ⅱ)相结合的多目标优化模型,以主要盾构参数为研究对象,以地表沉降和刀盘磨损为控制目标,优化控制分析施工参数;选取控制地表沉降的9个盾构参数作为RF预测模型的输入指标,得到盾构施工参数与地表沉降的非线性关系,作为NSGA-Ⅱ适应度函数,选择刀盘磨损作为第2个优化目标,设定施工参数约束范围进行多目标优化;以岩溶地区某地轨道交通项目为例进行验证。结果表明:采用RF算法模拟训练工程实测数据,得到的地表沉降预测模型的精度较高;基于RF-NSGA-Ⅱ模型智能算法对地表沉降和刀盘磨损的优化效果显著,该模型能够得出相应岩溶地区盾构施工参数控制范围。In order to effectively adjust construction parameters of shields, and to achieve safe construction, a multi-objective optimization model based on RF and NSGA-Ⅱ algorithm was established, in which control analysis of construction parameters were optimized with main shield parameters as research object and ground settlement as control target. Then, nine shield parameters controlling surface settlement were selected as input indexes of RF prediction model, and nonlinear relationship between the parameters and settlement was obtained as NSGA-Ⅱ fitness function. Then, cutter wear was selected as the second optimization objective, and constraint range of construction parameters was set for multi-objective optimization. Finally, with a rail transit project in karst areas as an example, verification was conducted. The results show that using RF algorithm in training and simulating measured data of the project will result in high prediction accuracy. And the proposed model based on RF-NSGA-Ⅱ features significant multi-objective optimization effect of ground settlement and cutter head wear, and it can obtain control range of shield construction parameters in karst areas.

关 键 词：随机森林(RF) 非支配排序遗传算法(NSGA-Ⅱ) 盾构施工参数 地表沉降 刀盘磨损 多目标优化 

分 类 号：X948[环境科学与工程—安全科学]