超大型管幕—箱涵顶进施工变形的分析与预测研究.pdf

大型“管幕－箱涵”顶进施工变形的分析与预测 同济大学隧道与地下工程研究所 孙钧 上海城建集团院士研究室 虞兴富 昊 李向阳 要：本文简要地介绍了工程概况和管幕－箱涵顶进施工工法（RBJ 法），对钢管幕和钢 筋砼箱涵顶进与土体开挖施工全过程中的施工变形完成了3D-FLAC 和3-D 弹塑性ANSYS 数值分析；进行了工程施工相似材料模拟试验；并对地表隆/沉变形进行了人工神经网络 智能法预测。本文系统细致的 学描述，明晰而定量地判明了工程上方地表施工变形随 施工进程的发展与变化，对确保工程施工安全实施显示了相当助益。 关键词：管幕，箱涵，顶进，施工变形，数值分析，智能预测 Abstract: By utilizing the Shanghai Mid-ring Line-Beihong Road underground intersection traffic tunneling Proj ect (constructed by using the pipe-roofing and large section box-culvert j acking trenchless excavation techlonogy) as an engineering background of this article s research, the authors here give a systematic study on the numerical simulation by using respectively the AN SYS FLAC 3-D computer programming, of ground movement in the whole construction process of above-mentioned proj ect. Secondly, by adopting the elastic foundation beam method, lab .-model experiment, and actual test data analogical techniques in situ separately, the authors give in advance a comprehensive study on characteristics of the ground movement variations developed in construction of the large section box j acking with pipe-roofing trenchless techniques. Finally, as adopting the intelligent soft science method, the authors make a so-called “multi-stepped rolling” prediction study by using Artificial Neural Networks (ANN) on surface ground settlement increasing with time around the proj ect. The contribution results of the research work has been provided to the design and construction companies of this proj ect for a valuable working reference. Key words: urban shallow-buried tunneling, pipe-roofing box-culvert j acking trenchless excavation, ground deformation, AN SYS FLAC computer program, numerical simulation, elastic-foundation beam, lab .-model experiment, multi-stepped rolling intelligent prediction, Artificial Neural Networks (ANN) .