最近软基问题的数值物理模型

移民厅大厦 (IMM Building, Lee et al. 1998). NUMERICAL AND PHYSICAL MODELLING OF SOIL PROBLEMS 土壤问题的数值与物理模型研究
by
Fook-Hou Lee National University of Singapore
李福豪 新加坡国立大学 平面图
剖面图
移民厅大厦 (IMM Building, Lee et al. 1998).
移民厅大厦 (IMM Building, Lee et al. 1998).
支撑剖面 显示
土层剖面显示
1

移民厅大厦 (IMM Building, Lee et al. 1998).
移民厅大厦 (IMM Building, Lee et al. 1998).
有限元分析的领域 有限元分析的剖面显示
移民厅大厦 (IMM Building, Lee et al. 1998).
移民厅大 厦 (IMM Building, Lee et al. 1998).
跨 距 中 点 的 挠 曲 形 状
角 落 点 的 挠 曲 形 状
变形后的有限元网格
2

局部的三维性 (Local 3dimensionality) 工桩与被挡土的交互
作用
局部的三维性 (Local 3dimensionality) 卒桩与被挡土的交互
作用
工桩 – 一般是用工 字桩。 工桩之间- 用挡土 木板盖住。 有分载反扛的效 应。
工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
平面图
3

工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
0 10 20 30
metres
EGL = 123.75m
EL = 106.35m
支撑平面图
工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
H610×305×149kg/ 2H610×305×149kg/m Depth 3.5m 6.5m
工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
有限元模型
支撑剖面图
2H610×305×149kg/m
10.5m
2H610×305×149kg/m
14.5m H610×305×149kg/m @ 1.5m c/c
2H610×305×149kg/m 2H610×305×149kg/m
18.5m 21.5m
25.0m
42.0m
4

工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
施工 过程
工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
用剑桥 模型算 出的成 果
工桩与被挡土的交互作用 – 现场 数据确认：东北线，实龙岗站
用双曲 线的剑 桥模型 算出的 成果。 显示出 小应变 非线性 质。
小应变非线性质
在屈服面内的应力-应变关系是非线性的。剪切模-应变关系可以用”S”-型 的曲线来形容。 双曲线的剑桥模型：在在屈服面内的应力-应变关系用双曲线来形容。
5

施工过程的三维性 (Construction sequence 3-dimensionality).
Most commonly tunnelling method used in Singapore is the earth pressure balance (EPB) method, which uses a rotating cutter head to cut the soil whilst applying pressure on the tunnel face to maintain stability and prevent collapse of the face.
Soil flows through openings in cutter head into the shield to be disposed off.
盾构隧道开挖过程
真实的隧道开挖过程(Real Tunnelling Operations)
(1) 开挖隧道工具作 面，同时把隧道钻机 往前推。 (2) 按装隧道衬片和水 泥灌浆。 (3) 隧道钻机切削面的 直径比钻机身的直径 大。所以隧道的直径 也比钻机身的直径 大。隧道的墙壁会向 内移→土体积损失： 尾部损失(tail loss)。
Piston Jacks Concrete Liner
Method of Modelling Tunnelling 模拟开挖隧道的方法
三维有限元 和现场测量 研究。 三维有限元 模拟工作面 压力，开 挖，排水边 界的改变， 尾部损失, 安 装隧道衬片 和水泥灌浆.
EPB machine
Face Loss
Shield Loss
Tail Loss
Piston Jacks
EPB machine
Concrete Liner
Face Loss
Shield Loss
Tail Loss
6

施工过程的三维分析： 现场数据确认-东北线
施工过程的三维分析： 现场数据确认-东北线
Tunnel driving direction North Bound South Bound
L5
L4
L3
L2
L1
施工过程的三维分析： 现场数据确认-东北线
Methods of Modelling Tunnelling 模拟开挖隧道的方法
工作面上压力使到 隧道前土面可能会 升起。 工作面后方还未安 装隧道衬片和水泥 灌浆的隧道墙壁是 排水边界→钻机土 层水位降低→土面 沉降。按装隧道衬 片和水泥灌浆之 后，隧道墙壁变成 不排水边界→水位 恢复原状→土面升 高。
0.20
Activation of Pore Pressure Fixity De-activation of Pore Pressure Fixity
0.00
C704 Data Empirical (Normal Cumulative Distribution)
Normalised Y Settlement (S/Smax)
-0.20
-0.40
-0.60
-0.80
-1.00
excav-A4.grf
-1.20 -20 -15 -10 -5 0 5 10 15 20
Z/D
7

ABAQUS 分析打沙桩过程 有限元分析
以上的分析是属于工作应力情况下的分 析。 非常大应变的分析（？）
Distortion even with very fine mesh
地基加固 (Ground Improvement) – 砂桩 （砂柱）
挤密砂桩装置过程
土地改良 (Ground Improvement) – 砂桩 （砂柱）
（1）把充满砂的钢管（直径~1m）插进软 土层里。 （2）把钢管拉出~3m，同时让砂从下端流 出。 （3）把钢管重复插进~2m→砂桩垂直压 缩，向侧面扩张→圆孔扩张效应。 （4）重复（2）和（3）。
8

Methodologies
Substructuring ABAQUS/Standard Re-zoning ABAQUS/Explicit Analysis with contact algorithm and adaptive meshing （显式解法）
Problems Encountered
Severe distortion of elements with coarse mesh （单元 的变形）
ABAQUS 分析打沙桩过程
非常大应变与混和的问题：有限元 有可能不能用。
9

离心模型的原理
用高-g的效果在小模型内生产出大原型的应力。
原型
模型
SCALING RELATIONS OF CENTRIFUGE MODELLING
缩放比例的问题
Particle size effects. 粒度效应
范围 体积 密度 质量 加速度 速度 位移 应变 能量密度 能量
Dynamic effects. 动力效应 Consolidation effects. 固结效应 Soil viscosity effects. 泥土的粘度效应 These are well-known and will not be discussed in this presentation.
10

离心测试
A. Numerical Validation （数字确认）: 用离心实验来确认分析 或程序. Examples: VELACS (Arulanandan & Scott, 1993). 地震分析程序
CENTRIFUGE TESTING
cyclic settlement of offshore spudcan footings (Ng, 1999). 探油台基脚的循环沉降
Mechanistic studies (机理的研究）: to elucidate ground behaviour and mechanisms. Examples:
FAILURE MECHANISMS OF TUNNELS (遂道的破坏机理）
11

EARTHQUAKE BEHAVIOUR OF SAND EMBANKMENTS （沙堤和沙岛的地震反应）
SPUDCAN FOOTINGS探油台基脚
NUS GEOTECHNICAL CENTRIFUGE
12

CYCLIC LOADING ON SPUDCAN FOOTINGS
PORE PRESSURE CONTOURS BENEATH CONICAL FOOTINGS
y
(0,0)
x
PPT1 PPT2
PPT1 PPT2
PPT1 PPT2
PPT1 PPT2
PPT3 PPT4
PPT3 PPT4
PPT3 PPT4
PPT3 PPT4
PPT5
PPT5
PPT5
PPT5
1st cycle
5th cycle
20th cycle
100th cycle
CENTRIFUGE TESTING C. Parametric Studies: to assess current design procedures and develop new design procedures ( 评定通用的设计方法和发展新设计 方法). Examples:
EXPLOSIVE CRATERING
Development of guidelines for explosive cratering (Holsapple & Schmidt, 1979, 1980).暴炸坑。 Bearing capacity and settlement of offshore spudcan footings under combined loadings (Tan, 1990; Murff et al., 1991, Ng, 1999). 探油台基脚的承载量与沉降。
13

CRATER VOLUME AGAINST CHARGE WEIGHT
14

SETTLEMENT OF SPUDCAN FOOTINGS
2.6
NG18 NG14
Settlement at V=50 MN (m)
2.4
NG04
NG05
NG16
2.2
NG13 NG19
2.0
Types of sand beds Dry sand Water sat. sand
NG12 NG03
NG09 NG10
NG11
1.8
Viscosity-scaled oil sat. Distorted-scaled oil sat.
1.6 40 50 60 70 80 90 100
Corrected RD (%)
CENTRIFUGE MODELLING B. Site-specific Modelling （具体地点的模拟）: modelling of a specific site or problem. Generally the most challenging class of centrifuge studies because of need to ensure adequate representation of: topography （地形）and soil stratification （土层）, structure （结 构）and fabric （构造）. boundary and groundwater conditions （边界和地下水的情况）. problem geometry, including 3-dimensionality （几何因素）. construction and loading processes and sequences （施工和加重的 过程）. Successful cases are generally few and far between.
CENTRIFUGE MODELLING B. Site-specific Modelling examples: Bearing capacity of shallow-buried footings (Garnier et al., 1989) 基脚的承载量.
15

CENTRIFUGE MODELLING B. Site-specific Modelling examples: Bearing capacity of shallow-buried footings (Garnier et al., 1989). Simple soil profile, only one type of soil 间单的土层. Simple geometry间单的几何条件. No groundwater considerations, w=5% 没地下水. Remoulded soil used in model and prototype 原状与模型都用 重塑土. Failure-related problem 破坏的问题.
CENTRIFUGE MODELLING B. Site-specific Modelling examples: Cyclic settlement of offshore caissons (Rowe and Craig, 1976) 沉 箱的循环沉降.
OOSTERSCHELDE STORM BARRIER
16

CENTRIFUGE MODELLING B. Site-specific Modelling examples: Cyclic settlement of offshore caissons (Rowe and Craig, 1976). Non-uniform sand bed 不均匀的沙层.(?) Liquefaction and displacement characteristics sensitive to loose zones局部的松散沙对液化和变形有很大的影响.
CENTRIFUGE MODELLING B. Site-specific Modelling examples: Modelling of a reinforced earth embankment (Sekine et al., 1998)
PROTOTYPE 土堤原型
钢筋土堤.
17

MODEL模型
EARTH PRESSURE COEFFICIENTS FOR MODELS AND PROTOTYPES 原型和模型的土压系数
CENTRIFUGE MODELLING B. Site-specific Modelling examples: Modelling of a reinforced earth embankment (Sekine et al., 1998). Model has different soil fraction from prototype原型和模型 的土壤粒级不同. Compaction of model ground not conducted at consistent glevel 模型的压紧过程不在高-g情况下进行. Fixity of model faceplate different from that of prototype during compaction 在压紧过程中，原型和模型面板的 Where 如果
LESSONS LEARNT
parameters to be matched are not dependent on or related to state of ground, such as bearing capacity （参数不受状态影响）, or if topography, ground and loading condition can be adequately represented （土型，土壤和加重条件充分地模拟）, Good site specific modelling may be achieved .
固定性不同.
Main problem appears to be errors from modelling of the construction processes.
Working load parameters such as displacements and working stresses tend to be more dependent upon the initial state of the model ground and thus the model preparation procedures. Thus, more difficult to reproduce. 工作状态的参数，列如位移和应力同常会受到现场状态和施工过程 的影响，所以往往比较难以模拟。
18

PROPOSED STRATEGY FOR CENTRIFUGE MODELLERS
PROPOSED STRATEGY FOR CENTRIFUGE MODELLERS Managing expectation of the engineering clients: Centrifuge modelling should be treated in a similar manner as other forms of modelling such as numerical modelling; it should not be expected to match field measurements exactly. Most important feature must be a conservative, “ballpark” figure. Conservatism may be introduced by use of remoulded soil as well as conservative idealization of the actual problem （用重塑土的效 应：强度，弹模比原状土低=〉保守?）. Main advantages of centrifuge modelling is its ability to deal with real processes and real materials. For certain problems, such as those involve complex impulsive, dynamic or cyclic loadings as well as complicated constructive processes such as compaction, which are often difficult to model using numerical approach, centrifuge can offer very real advantages (hence the need to improve construction process modelling) 数字模型不容以处理的问题？.
离心模型的策略
Development of modelling technology and sophistication 发展新的
模拟技术:
Model preparation for more complex geometries and soil profiles. To avoid boundary effects, larger platform working area advantageous for some problems. We need reasonably small radius centrifuge with relatively large working platform 小直径，大吊篮的 离心机。 Improvement needed on construction processes modelling technology e.g. deep excavation with strutting, sand compaction pile installer and x-y frame 发展模拟施工过程的科技.
EXAMPLES OF PROBLEMS WHICH CANNOT BE READILY MODELLED AT A NUMERICAL LEVEL
processes involving moving boundaries and large displacements e.g. penetration or extrusion problems dynamic or impulsive processes such as heavy tamping, explosions and earthquakes, mixing processes e.g. mixing of soil and cement.
A lot of ground improvement work involves these processes.
19