摘要
以软土地区某地铁车站深基坑工程中挡墙实测侧移为依据,结合有限单元法和改进的神经网络法开展位移反分析研究,获得了以硬化土模型作为本构模型的土层反演参数,并通过挡墙侧移、坑外地表沉降的计算值与实测值的比较,验证反分析方法的可行性和所得反演参数的可靠性。研究成果有助于今后更有效地利用有限元法等数值方法评估软土地区车站深基坑开挖对基坑自身及周边环境的影响。
Based on the measured lateral displacement of retaining wall due to the excavation of a subway station in soft soil area, a displacement back analysis is carried out by combining the finite element method and the modified neural network method to obtain the inversion parameters of the soil layers which are constituted by the hardening soil model, and the feasibility of the back analysis method and the reliability of the inversion parameters are verified by comparing the lateral displacement of retaining wall and the settlement of ground surface by calculating with those by measuring. The researchach ievement may contribute to the effective utilization of the numerical methods such as the finite element method in evaluating the influence of deep excavation of subway station in soft soil area on the foundation pit itself and the surrounding environment.
作者
吴才德
章玉明
田领川
成怡冲
曾婕
Wu Caide;Zhang Yuming;Tian Lingchuan;Cheng Yichong;Zeng Jie(Zhejiang Huazhan Institute of Engineering Research & Design,Ningbo Zhejiang 315012,China;Ningbo Preparation Office of Municipal Engineering,Ningbo Zhejiang 315012,China)
出处
《科技通报》
北大核心
2017年第1期142-146,共5页
Bulletin of Science and Technology
关键词
位移反分析
神经网络
硬化土模型
深基坑
地铁车站
displacement back analysis
neural network
hardening soil model
deep foundation pit
subway station
