Racking Analysis Wang
Racking Analysis Wang
Racking Analysis Wang
Jaw-Nan (Joe) Wang, Ph.D., P.E. Professional Associate Parsons Brinckerhoff Quade & Douglas, Inc. June 1993
First Printing 1993 Copyright Jaw-Nan Wang and Parsons Brinckerhoff Inc.
All rights reserved. No part of this work covered by the copyright thereon may be reproduced or used in any form or by any means graphic, electronic, or mechanical, including photocopying, recording, taping, or information storage or retrieval systems without permission of the publisher.
Published by Parsons Brinckerhoff Inc. One Penn Plaza New York, New York
CONTENTS
Foreword 1.0 Introduction 1.1 Purpose 1.2 Scope of this Study 1.3 Background Importance of Seismic Design Seismic Design before the 90s 1.4 General Effects of Earthquakes Ground Shaking Ground Failure 1.5 Performance Record in Earthquakes 2.0 Seismic Design Philosophy for Tunnel Structures 2.1 Seismic Design vs. Conventional Design 2.2 Surface Structures vs. Underground Structures Surface Structures Underground Structures Design and Analysis Approaches 2.3 Seismic Design Philosophies for Other Facilities Bridges and Buildings Nuclear Power Facilities Port and Harbor Facilities Oil and Gas Pipeline Systems 2.4 Proposed Seismic Design Philosophy for Tunnel Structures Two-Level Design Criteria Loading Criteria 3.0 Running Line Tunnel Design 3.1 Overview 3.2 Types of Deformations Axial and Curvature Deformations Ovaling or Racking Deformations
ix 1 3 4 4 4 5 7 7 8 8 13 15 15 15 16 16 17 17 17 18 18 19 19 20 25 27 27 27 29
i
3.3 Free-Field Axial and Curvature Deformations Background A Practical Approach to Describing Ground Behavior Simplified Equations for Axial Strains and Curvature 3.4 Design Conforming to Free-Field Axial and Curvature Deformations Background and Assumptions Design Example 1: The Los Angeles Metro Applicability of the Free-Field Deformation Approach 3.5 Tunnel-Ground Interaction Simplified Interaction Equations Design Example 2: A Linear Tunnel in Soft Ground 3.6 Special Considerations Unstable Ground Faulting Abrupt Changes in Structural Stiffness or Ground Conditions
31 31 31 33 35 35 35 37 37 38 43 48 48 48 49
4.0
Ovaling Effect on Circular Tunnels 4.1 Ovaling Effect 4.2 Free-Field Shear Deformations Simplified Equation for Shear Deformations 4.3 Lining Conforming to Free-Field Shear Deformations 4.4 Importance of Lining Stiffness Compressibility and Flexibility Ratios Example 1 Example 2 Summary and Conclusions 4.5 Lining-Ground Interaction Closed Form Solutions Numerical Analysis Results and Recommendations
53 55 55 56 58 60 60 61 62 63 64 64 76 76
5.0
Racking Effect on Rectangular Tunnels 5.1 General 5.2 Racking Effect 5.3 Dynamic Earth Pressure Methods
83 85 86 87
ii
5.4
5.5
5.6
5.7
Mononobe-Okabe Method Wood Method Implications for Design Free-Field Racking Deformation Method San Francisco BART Los Angeles Metro Flexibility vs. Stiffness Applicability of the Free-Field Racking Method Examples Tunnel-Ground Interaction Analysis Factors Contributing to the Soil-Structure Interaction Effect Method of Analysis Flexibility Ratio for Rectangular Tunnels Results of Analysis Recommended Procedure: Simplified Frame Analysis Models Step-by-Step Design Procedure Verification of the Simplified Frame Models Summary of Racking Design Approaches
6.0
Summary Vulnerability of Tunnel Structures Seismic Design Philosophy Running Line Tunnel Design Ovaling Effect on Circular Tunnels Racking Effect on Rectangular Tunnels
References
141
iii
LIST OF FIGURES
Figure 1 2 3 4 5 6 7 Title Ground Response to Seismic Waves Damage Statistics Axial and Curvature Deformations Ovaling and Racking Deformations Geometry of a Sinusoidal Shear Wave Oblique to Axis of Tunnel Sectional Forces Due to Curvature and Axial Deformations Page 6 11 28 30 32 39
Free-Field Shear Distortions of Ground Under Vertically Propagating Shear Waves 57 Free-Field Shear Distortion of Ground (Non-Perforated Medium) Shear Distortion of Perforated Ground (Cavity In-Place) Lining Response Coefficient, K1 (Full-Slip Interface) Lining Response Coefficient, K1 (Full-Slip Interface) Lining Response (Thrust) Coefficient, K2 (No-Slip Interface) Lining Response (Thrust) Coefficient, K2 (No-Slip Interface) Lining Response (Thrust) Coefficient, K2 (No-Slip Interface) Normalized Lining Deflection (Full-Slip Interface) Normalized Lining Deflection (Full-Slip Interface) Finite Difference Mesh (Pure Shear Condition) Influence of Interface Condition on Bending Moment Influence of Interface Condition on Lining Deflection 59 59 66 67 69 70 71 73 74 75 78 80
8 9 10 11 12 13 14 15 16 17 18 19
iv
Figure
Title
Page
20
Typical Free-Field Racking Deformation Imposed on a Buried Rectangular Frame Structure Stability for Buried Rectangular Frames Soil-Structure System Analyzed in Example Subsurface Shear Velocity Profiles Free-Field Shear Deformations (from Free-Field Site Response Analysis, SHAKE) Structure Deformations vs. Free-Field Deformations, Case I (from Soil/Structure Interaction Analysis, FLUSH) Structure Deformations vs. Free-Field Deformations, Case ll (from Soil/Structure Interaction Analysis, FLUSH) Typical Finite Element Model (for Structure Type 2) Earthquake Accelerograms on Rock West Coast Northeast Design Response Spectra on Rock (West Coast Earthquake vs. Northeast Earthquake) Types of Structure Geometry Used in the Study Relative Stiffness Between Soil and a Rectangular Frame Determination of Racking Stiffness Normalized Racking Deflections (for Cases 1 through 25) Normalized Structure Deflections Normalized Structure Deflections
89 91 93 95
21 22 23 24
97
25
98
26
99
27
103
28
104 105
29
30 31 32 33
34 35
Figure
Title
Page
36 37 38 39
Effect of Embedment Depth on Racking Response Coefficient, R Normalized Structure Deflections Simplified Frame Analysis Models Moments at Roof-Wall Connections Concentrated Force Model (for Cases 1 through 5) Moments at Invert-Wall Connections Concentrated Force Model (for Cases 1 through 5) Moments at Roof-Wall Connections Triangular Pressure Distribution Model (for Cases 1 through 5) Moments at Invert-Wall Connections Triangular Pressure Distribution Model (for Cases 1 through 5)
129
40
130
41
131
42
132
vi
LIST OF TABLES
Table 1 2 3 4 5 6 7 Title Free-Field Ground Strains Cases Analyzed by Finite Difference Modeling Influence of Interface Conditions on Thrust Cases Analyzed by Dynamic Finite Element Modeling Cases Analyzed to Study the Effect of Burial Depth Cases Analyzed to Study the Effect of Stiff Foundation Seismic Racking Design Approaches Page 34 77 81 113 120 123 134
vii
viii
FOREWORD
For more than a century, Parsons Brinckerhoff (PB) has been instrumental in advancing state-of-the-art design and construction of underground structures, and the fields of seismic design and earthquake engineering are no exceptions. Almost three decades ago PBs engineers pioneered in these fields in the design and construction of the San Francisco BART system, whose toughness during earthquakes, including the recent Loma Prieta event, has been amply tested. Recently, PB developed state-of-theart, two-level seismic design philosophy in its ongoing Los Angeles Metro and Boston Central Artery/Third Harbor Tunnel projects, taking into account both performance-level and life-safety-level earthquakes. This monograph represents PBs continuous attempts in the seismic design and construction of underground structures to: Improve our understanding of seismic response of underground structures Formulate a consistent and rational seismic design procedure Chapter 1 gives general background information including a summary of earthquake performance data for underground structures. Chapter 2 presents the seismic design philosophy for tunnel structures and the rationale behind this philosophy. Differences in seismic considerations between surface structures and underground structures, and those between a seismic design and a static design are also discussed. Chapter 3 focuses on the seismic design considerations in the longitudinal direction of the tunnels. Axial and curvature deformations are the main subjects. The free-field deformation method and the methods accounting for tunnel-ground interaction effects are reviewed for their applicability. Chapter 4 takes a look at the ovaling effect on circular tunnel linings. Closed-form solutions considering soil-lining interaction effects are formulated and presented in the form of design charts to facilitate the design process. Chapter 5 moves to the evaluation of racking effect on cut-and-cover rectangular tunnels. This chapter starts with a review of various methods of analysis that are currently in use, followed by a series of dynamic finite-element analyses to study the various factors influencing the tunnel response. At the end, simplified frame analysis models are proposed for this evaluation. Chapter 6 ends this monograph with a general summary.
ix
Acknowledgments
I wish to express my thanks to the Career Development Committee and Paul H. Gilbert, the original initiator of the William Barclay Parsons Fellowship Program, for selecting my proposal and providing continuous support and guidance throughout this study. Thanks are also due to the Board of Directors of Parsons Brinckerhoff Inc. for making the growth and flowering of an engineers idea possible. The fruitful results of this exciting study would never have been possible without technical guidance from three individuals my fellowship mentors, Dr. George Munfakh and Dr. Birger Schmidt, and the technical director of underground structures, Dr. James E. Monsees. Their constant critiques and advice were sources of inspiration and motivation. Appreciation is due also to Tom Kuesel, who gave constructive technical comments on the content of this study, and to Tim Smirnoff, who provided much of the tunnel structural data of the LA Metro project. Ruchu Hsu and Rick Mayes deserve my thanks for generously giving their time and comments on the draft of this monograph. Gratitude is offered to many other individuals for numerous technical discussions on real world seismic design issues for the ongoing Central Artery/Third Harbor Tunnel project and the Portland Westside LRT project. They include: Louis Silano, Vince Tirolo, Anthony Lancellotti, Dr. Sam Liao, Brian Brenner, Alexander Brudno, Mike Della Posta, Dr. Edward Kavazanjian, Richard Wilson, and many others. Very special thanks to Willa Garnick for her exquisite editing of the manuscript, and to Randi Aronson who carefully proofread the final draft of the monograph. Their wonderful work gave this fellowship study a beautiful finish. I also acknowledge the support and contribution of personnel of the New York office Graphics Department, particularly Pedro Silva who prepared the graphics and tables and laid out the text. I simply could not put a period to this study without expressing thanks to my wife Yvonne Yeh, my son Clinton and my daughter Jolene. Their sacrificing support of my work through many late nights and weekends contributed the greatest part to this monograph.
Jaw-Nan (Joe) Wang, Ph.D., P.E. Professional Associate Parsons Brinckerhoff Quade & Douglas, Inc. June 1993
x
1.0 INTRODUCTION
1.0 INTRODUCTION
1.1 Purpose
The purpose of this research study was to develop a rational and consistent seismic design methodology for lined transportation tunnels that would also be applicable to other underground lined structures with similar characteristics. The results presented in this report provide data for simple and practical application of this methodology. While the general public is often skeptical about the performance of underground structures, tunnel designers know that underground structures are among the safest shelters during earthquakes, based primarily on damage data reported in the past. Yet one certainly would not want to run away from a well designed building into a buried tunnel when seismic events occur if that tunnel had been built with no seismic considerations. Most tunnel structures were designed and built, however, without regard to seismic effects. In the past, seismic design of tunnel structures has received considerably less attention than that of surface structures, perhaps because of the conception about the safety of most underground structures cited above. In fact, a seismic design procedure was incorporated into a tunnel project for the first time in the 1960s by PB engineers. In recent years, however, the enhanced awareness of seismic hazards for underground structures has prompted an increased understanding of factors influencing the seismic behavior of underground structures. Despite this understanding, significant disparity exists among engineers in design philosophy, loading criteria, and methods of analysis. Therefore, this study, geared to advance the state of the art in earthquake engineering of transportation tunnels, has the following goals: To maintain a consistent seismic design philosophy and consistent design criteria both for underground structures and other civil engineering facilities. To develop simple yet rational methods of analysis for evaluating earthquake effects on underground structures. The methodology should be consistent for structures with different section geometries.
1.3 Background
Importance of Seismic Design One of the significant aspects of the 1989 Loma Prieta earthquake in the San Francisco area was its severe impact on the aboveground transportation system:
4
The collapse of the I-880 viaduct claimed more than 40 lives. The direct damage costs to the transportation facilities alone totalled nearly $2 billion (Werner and Taylor, 1990). The indirect losses were several times greater as a result of major disruptions of transportation, particularly on the San Francisco-Oakland Bay Bridge and several major segments of the Bay area highway system. The San Francisco Bay Area Rapid Transit (BART) subway system was found to be one of the safest places during the event, and it became the only direct public transportation link between Oakland and San Francisco after the earthquake. Had BART been damaged and rendered inoperative, the consequences and impact on the Bay area would have been unthinkable. The 60-mile BART system was unscathed by the earthquake because PB engineers had the foresight 30 years ago to incorporate state-of-the-art seismic design criteria in their plans for the subway tunnels (SFBARTD, 1960; Kuesel, 1969; and Douglas and Warshaw, 1971). The Loma Prieta earthquake proved the worth of their pioneering efforts. Seismic Design Before the 90s Based on the performance record, it is undoubtedly fair to say that underground structures are less vulnerable to earthquakes than surface structures (Dowding and Rozen, 1978; Rowe, 1992). Interestingly, some tunnels and shafts built without special earthquake provisions have survived relatively strong earthquakes in the past for example, the Mexico City subway during the 1985 Mexico City earthquake. On the other hand, some underground structures have been damaged severely in other events (see Section 1.5). Limited progress has been made in seismic design methodology for underground tunnels since the work for BART, possibly because of favorable performance data, and limited research work has been done toward a practical solution. The lack of a rational methodology for engineers and the nonexistence of applicable codes has led to widely varied measures taken by different engineers. For example: Some ignore seismic effects and fail to check the resistance of the structures to earthquakes, even in highly seismic areas. Others conduct their seismic design for underground structures using the same methodology developed for aboveground structures, without recognizing that underground structures are constrained by the surrounding medium.
Design based on such inappropriate measures may lead to the construction of unsafe structures or structures that are too conservatively designed. Although the progress of underground seismic design methodology is lagging, the earthquake awareness in the country is not. Recent discoveries in seismology, geology and geotechnical engineering have led to the belief that earthquake hazard is no longer only a California problem. Many regions throughout the United States, Puerto Rico and the Virgin Islands are now known to have the potential for tremors of similar or larger magnitude than that of the Loma Prieta. This situation demands rethinking of the current seismic design practice for our underground transportation systems.
Ground Failure Ground failure broadly includes various types of ground instability such as faulting, landslides, liquefaction, and tectonic uplift and subsidence. Each of these hazards may be potentially catastrophic to tunnel structures, although the damages are usually localized. Design of a tunnel structure against ground instability problems is often possible, although the cost may be high. For example, it may be possible to remedy the ground conditions against liquefaction and landslides with proper ground improvement techniques and appropriate earth retaining measures. It may not be economically or technically feasible, however, to build a tunnel to resist potential faulting displacements. As suggested by Rowe (1992), the best solution to the problem of putting a tunnel through an active fault is - dont. Avoidance of faults may not always be possible, however, because a tunnel system may spread over a large area. In highly seismic areas such as California, tunnels crossing faults may be inevitable in some cases. The design approach to this situation is to accept the displacement, localize the damage, and provide means to facilitate repairs (Kuesel, 1969).
Based on their study, Dowding and Rozen concluded, primarily for rock tunnels, that: Tunnels are much safer than aboveground structures for a given intensity of shaking. Tunnels deep in rock are safer than shallow tunnels. No damage was found in both lined and unlined tunnels at surface accelerations up to 0.19g. Minor damage consisting of cracking of brick or concrete or falling of loose stones was observed in a few cases for surface accelerations above 0.25g and below 0.4g. No collapse was observed due to ground shaking effect alone up to a surface acceleration of 0.5g. Severe but localized damage including total collapse may be expected when a tunnel is subject to an abrupt displacement of an intersecting fault. Owen and Scholl (1981) These authors documented additional case histories to Dowding and Rozens, for a total of 127 case histories. These added case histories, in addition to rock tunnels, included: Damage reports on cut-and-cover tunnels and culverts located in soil Data on underground mines, including shafts The authors discussion of some of the damaged cut-and-cover structures is of particular interest. These structures have the common features of shallow soil covers and loose ground conditions: A cut-and-cover railroad tunnel with brick lining (two barrels, each approximately 20 feet wide) was destroyed by the 1906 San Francisco earthquakes. In this case, where brick lining with no moment resistance was used, the tunnel structure collapsed. Five cases of cut-and-cover conduits and culverts with reinforced concrete linings were damaged during the 1971 San Fernando earthquake. The damages experienced by the linings included: - The failure of longitudinal construction joints - Development of longitudinal cracks and concrete spalling
- Formation of plastic hinges at the top and bottom of walls The conclusions made by Owen and Scholl, based on their study, echoed the findings by Dowding and Rozen discussed above. In addition, they suggested the following: Damage to cut-and-cover structures appeared to be caused mainly by the large increase in the lateral forces from the surrounding soil backfill. Duration of strong seismic motion appeared to be an important factor contributing to the severity of damage to underground structures. Damage initially inflicted by earth movements, such as faulting and landslides, may be greatly increased by continued reversal of stresses on already damaged sections. Wang (1985) In describing the performance of underground facilities during the magnitude 7.8 Tang-Shan earthquake of 1976, the author reported the following: An inclined tunnel passing through 13 feet of soil into limestone was found to have cracks up to 2 cm wide on the side wall. The plain concrete floor heaved up 5 to 30 cm. Damage to underground facilities decreased exponentially with depth to 500 m. Schmidt and Richardson (1989) attributed this phenomenon to two factors: - The increasing competence of the soil/rock with depth - The attenuation of ground shaking intensity with depth Sharma and Judd (1991) The authors extended Owen and Scholls work and collected qualitative data for 192 reported observations from 85 worldwide earthquake events. They correlated the vulnerability of underground facilities with six factors: overburden cover, rock type (including soil), peak ground acceleration, earthquake magnitude, epicentral distance, and type of support. It must be pointed out that most of the data reported are for earthquakes of magnitude equal to 7 or greater. Therefore, the damage percentage of the reported data may appear to be astonishingly higher than one can normally conceive. The results are summarized in the following paragraphs. Readers should be aware that these statistical data are of a very qualitative nature. In many cases, the damage statistics, when correlated with a certain parameter, may show a trend that violates an engineers intuition. This may be attributable to the statistical dependency on other parameters which may be more influential.
10
The effects of overburden depths on damage are shown in Figure 2A for 132 of the 192 cases. Apparently, the reported damage decreases with increasing overburden depth. Figure 2B shows the damage distribution as a function of material type surrounding the underground opening. In this figure, the data labeled Rock (?) were used for all deep mines where details about the surrounding medium were not known. The data indicate more damage for underground facilities constructed in soil than in competent rock. The relationship between peak ground acceleration (PGA) and the number of damaged cases are shown in Figure 2C. - For PGA values less than 0.15g, only 20 out of 80 cases reported damage. - For PGA values greater than 0.15g, there were 65 cases of reported damage out of a total of 94 cases. Figure 2D summarizes the data for damage associated with earthquake magnitude. The figure shows that more than half of the damage reports were for events that exceeded magnitude M=7. The damage distribution according to the epicentral distance is presented in Figure 2E. As indicated, damage increases with decreasing epicentral distance, and tunnels are most vulnerable when they are located within 25 to 50 km from the epicenter. Among the 192 cases, unlined openings account for 106 cases. Figure 2F shows the statistical damage data for each type of support. There were only 33 cases of concrete-lined openings including 24 openings lined with plain concrete and 9 cases with reinforced concrete linings. Of the 33 cases, 7 were undamaged, 12 were slightly damaged, 3 were moderately damaged, and 11 were heavily damaged. It is interesting to note that, according to the statistical data shown in Figure 2F, the proportion of damaged cases for the concrete and reinforced concrete lined tunnels appears to be greater than that for the unlined cases. Sharma and Judd attributed this phenomenon to the poor ground conditions that originally required the openings to be lined. Richardson and Blejwas (1992) offered two other possible explanations: -Damage in the form of cracking or spalling is easier to identify in lined openings than in unlined cases. -Lined openings are more likely to be classified as damaged because of their high cost and importance.
12
13
14
loads arising from other causes. To design a bridge to remain elastic and undamaged for such infrequent loads is uneconomical and sometimes not possible (Buckle, et al, 1987). Therefore, it is clearly not practical to use the same design approach to earthquakes as is used for other types of loads. The seismic design philosophy developed for bridges (AASHTO, 1991) is discussed briefly in Section 2.3. Surface structures are not only directly subjected to the excitations of the ground, but also experience amplification of the shaking motions depending on their own vibratory characteristics. If the predominant vibratory frequency of the structures is similar to the natural frequency of the ground motions, the structures are excited by resonant effects. Underground Structures In contrast, underground structures are constrained by the surrounding medium (soil or rock). It is unlikely that they could move to any significant extent independently of the medium or be subjected to vibration amplification. Compared to surface structures, which are generally unsupported above their foundations, the underground structures can be considered to display significantly greater degrees of redundancy thanks to the support from the ground. These are the main factors contributing to the better earthquake performance data for underground structures than their aboveground counterparts. Design and Analysis Approaches The different response characteristics of aboveground and underground structures suggest different design and analysis approaches: Force Method for Surface Structures. For aboveground structures, the seismic loads are largely expressed in terms of inertial forces. The traditional methods generally involve the application of equivalent or pseudostatic forces in the analysis. Deformation Method for Underground Structures. The design and analysis for underground structures should be based, however, on an approach that focuses on the displacement/deformation aspects of the ground and the structures, because the seismic response of underground structures is more sensitive to such earthquake induced deformations. The deformation method is the focus of this report.
16
Port and Harbor Facilities Neither standard seismic codes nor universally accepted seismic design criteria exist for waterfront facilities such as berthing (wharf) structures, retaining structures, and dikes. Recent advances in seismic design practice for other facilities, however, have prompted the development of several project specific seismic design criteria for waterfront facilities in high seismic areas (POLA, 1991; Wittkop, 1991; Torseth, 1984). The philosophy employed in the design, again, is based on two-level criteria: Under an Operating Level Earthquake (OLE), a smaller earthquake, the structures should experience little to no damage and the deformations of wharf structures should remain within the elastic range. Generally, the OLE is defined to have a probability of exceedance of 50 percent in 50 years. Under a Contingency Level Earthquake (CLE), a larger earthquake, the structures should respond in a manner that prevents collapse and major structural damage, albeit allowing some structural and nonstructural damage. Damage that does occur should be readily detectable and accessible for inspection and repair. Damage to foundation elements below ground level should be prevented (POLA, 1991). Generally, the CLE is to have a probability of exceedance of 10 percent in 50 years. The risk level defined for the CLE is similar to that of the design earthquake adopted in bridge and building design practice. Oil and Gas Pipeline Systems The seismic design guidelines recommended by ASCE (Nyman, et al, 1984) for oil and gas pipeline systems are in many ways similar to the principles used in the design for other important facilities. For important pipeline systems, the design should be based on two-level earthquake hazard: The Probable Design Earthquake (PDE), the lower level, is generally associated with a return period of 50 to 100 years. The Contingency Design Earthquake (CDE), the higher level, is represented by an event with a return period of about 200 to 500 years. The general performance requirements of the pipeline facilities under the two design events are also similar to those for other facilities.
18
19
Loading Criteria Maximum Design Earthquake (MDE). Given the performance goals of the MDE (i.e., public safety), the recommended seismic loading combinations using the load factor design method are as follows: For Cut-and-Cover Tunnel Structures
U = D + L + E1+ E2 + EQ
Where U = required structural strength capacity D = effects due to dead loads of structural components L = effects due to live loads E1 = effects due to vertical loads of earth and water E2 = effects due to horizontal loads of earth and water EQ = effects due to design earthquake (MDE) For Mined (Circular) Tunnel Lining
(Eq. 2-1)
U = D + L + EX + H + EQ
where U, D, L, and EQ are as defined in Equation 2-1
(Eq. 2-2)
EX = effects of static loads due to excavation (e.g., ORourke, 1984) H = effects due to hydrostatic water pressure
Comments on Loading Combinations for MDE The structure should first be designed with adequate strength capacity under static loading conditions. The structure should then be checked in terms of ductility as well as strength when earthquake effects, EQ, are considered. The EQ term for conventional surface structure design reflects primarily the inertial effect on the structures. For tunnel structures, the earthquake effect is governed by the displacements/deformations imposed on the tunnels by the ground. In checking the strength capacity, the effects of earthquake loading should be
20
expressed in terms of internal moments and forces, which can be calculated according to the lining deformations (distortions) imposed by the surrounding ground. If the strength criteria expressed by Equation 2-1 or 2-2 can be satisfied based on elastic structural analysis, no further provisions under the MDE are required. Generally the strength criteria can easily be met when the earthquake loading intensity is low (i.e., in low seismic risk areas) and/or the ground is very stiff. If the flexural strength of the tunnel lining, using elastic analysis and Equation 2-1 or 22, is found to be exceeded (e.g., at certain joints of a cut-and-cover tunnel frame), one of the following two design procedures should be followed: (1) Provide sufficient ductility (using proper detailing procedure) at the critical locations of the lining to accommodate the deformations imposed by the ground in addition to those caused by other loading effects (see Equations 2-1 and 2-2). The intent is to ensure that the structural strength does not degrade as a result of inelastic deformations and the damage can be controlled at an acceptable level. In general the more ductility is provided, the more reduction in earthquake forces (the EQ term) can be made in evaluating the required strength, U. As a rule of thumb, the force reduction factor can be assumed equal to the ductility provided. This reduction factor is similar by definition to the response modification factor used in bridge design code (AASHTO). Note, however, that since an inelastic shear deformation may result in strength degradation, it should always be prevented by providing sufficient shear strengths in structure members, particularly in the cut-and-cover rectangular frame. (2) Re-analyze the structure response by assuming the formation of plastic hinges at the joints that are strained into inelastic action. Based on the plastic-hinge analysis, a redistribution of moments and internal forces will result. If new plastic hinges are developed based on the results, the analysis is re-run by incorporating the new hinges (i.e., an iterative procedure) until all potential plastic hinges are properly accounted for. Proper detailing at the hinges is then carried out to provide adequate ductility. The structural design in terms of required strength (Equations 2-1 and 2-2) can then be based on the results from the plastic-hinge analysis. As discussed earlier, the overall stability of tunnel structures during and after the MDE has to be maintained. Realizing that the structures also must have sufficient capacity (besides the earthquake effect) to carry static loads (e.g., D, L, E1, E2 and H terms), the potential modes of instability due to the development of plastic
21
hinges (or regions of inelastic deformation) should be identified and prevented (Monsees, 1991; see Figure 21 for example). The strength reduction factor, f , used in the conventional design practice may be too conservative, due to the inherently more stable nature of underground structures (compared to surface structures), and the transient nature of the earthquake loading. For cut-and-cover tunnel structures, the evaluation of capacity using Equation 2-1 should consider the uncertainties associated with the loads E1 and E2, and their worst combination. For mined circular tunnels (Equation 2-2), similar consideration should be given to the loads EX and H. In many cases, the absence of live load, L, may present a more critical condition than when a full live load is considered. Therefore, a live load equal to zero should also be used in checking the structural strength capacity using Equations 2-1 and 2-2. Operating Design Earthquake (ODE). For the ODE, the seismic design loading combination depends on the performance requirements of the structural members. Generally speaking, if the members are to experience little to no damage during the lowerlevel event (ODE), the inelastic deformations in the structure members should be kept low. The following loading criteria, based on load factor design, are recommended: For Cut-and-Cover Tunnel Structures
(Eq. 2-3)
b1 = 1.05 if extreme loads are assumed for E1 and E2 with little uncertainty. Otherwise, use b1 = 1.3. For Mined (Circular) Tunnel Lining
(Eq. 2-4)
b2 = 1.05 if extreme loads are assumed for E1 and E2 with little uncertainty. Otherwise, use b2 = 1.3.
22
Comments on Loading Combinations for ODE The structure should first be designed with adequate strength capacity under static loading conditions. For cut-and-cover tunnel structures, the evaluation of capacity using Equation 2-3 should consider the uncertainties associated with the loads E1 and E2, and their worst combination. For mined circular tunnels (Equation 2-4), similar consideration should be given to the loads EX and H. When the extreme loads are used for design, a smaller load factor is recommended to avoid unnecessary conservatism. Note that an extreme load may be a maximum load or a minimum load, depending on the most critical case of the loading combinations. Use Equation 2-4 as an example. For a deep circular tunnel lining, it is very likely that the most critical loading condition occurs when the maximum excavation loading, EX, is combined with the minimum hydrostatic water pressure, H. For a cut-and-cover tunnel, the most critical seismic condition may often be found when the maximum lateral earth pressure, E2, is combined with the minimum vertical earth load, E1. If a very conservative lateral earth pressure coefficient is assumed in calculating the E2, the smaller load factor b1 = 1.05 should be used. Redistribution of moments (e.g., ACI 318) for cut-and-cover concrete frames is recommended to achieve a more efficient design. If the strength criteria expressed by Equation 2-3 or 2-4 can be satisfied based on elastic structural analysis, no further provisions under the ODE are required. If the flexural strength of the tunnel lining, using elastic analysis and Equation 2-3 or 24, is found to be exceeded, the structure should be checked for its ductility to ensure that the resulting inelastic deformations, if any, are small. If necessary, the structure should be redesigned to ensure the intended performance goals during the ODE. Zero live load condition (i.e., L = 0) should also be evaluated in Equations 2-3 and 2-4.
23
24
25
26
27
Ovaling or Racking Deformations The ovaling or racking deformations of a tunnel structure may develop when waves propagate in a direction perpendicular or nearly perpendicular to the tunnel axis, resulting in a distortion of the cross-sectional shape of the tunnel lining. Design considerations for this type of deformation are in the transverse direction. Figure 4 shows the ovaling distortion and racking deformation associated with circular tunnels and rectangular tunnels, respectively. The general behavior of the lining may be simulated as a buried structure subject to ground deformations under a two-dimensional, plane-strain condition. Ovaling and racking deformations may be caused by vertically, horizontally or obliquely propagating seismic waves of any type. Many previous studies have suggested, however, that the vertically propagating shear wave is the predominant form of earthquake loading that governs the tunnel lining design against ovaling/racking. The following reasons are given: Ground motion in the vertical direction is generally considered less severe than its horizontal component. Typically, vertical ground motion parameters are assumed to be 1/2 to 2/3 of the horizontal ones. (Note that a vertically propagating shear wave causes the ground to shake in the horizontal direction.) This relation is based on observation of California earthquakes, which are most commonly of the strike-slip variety in which horizontal motion predominates. For thrust faults, in which one rock block overrides another, vertical effects may equal or exceed the horizontal ones. The effects of thrust faulting are usually more localized, however, than those of the strike-slip faulting, and they are attenuated more rapidly with distance from the focus. For tunnels embedded in soils or weak media, the horizontal motion associated with vertically propagating shear waves tends to be amplified. In contrast, the ground strains due to horizontally propagating waves are found to be strongly influenced by the ground strains in the rock beneath. Generally, the resulting strains are smaller than those calculated using the properties of the soils.
29
31
Axis of Tunnel
Simplified Equations for Axial Strains and Curvature Using the simplified approach, the free-field axial strains and curvature due to shear waves and Rayleigh waves (surface waves) can be expressed as a function of angle of incidence, as shown in Table 1. The most critical angle of incidence and the maximum values of the strains are also included in the table. Equations caused by compressional P-waves are also available, but it is generally considered that they would not control the design. It is difficult to determine which type of wave will dominate due to the complex nature of the characteristics associated with different wave types. Generally, strains produced by Rayleigh waves may govern only when the site is at a large distance from the earthquake source and the structure is built at shallow depth. Application of the strain equations presented in Table 1 requires knowledge of: The effective wave propagation velocity The peak ground particle velocity The peak ground particle acceleration The peak velocity and acceleration can be established through empirical methods, field measurements, or site-specific seismic exposure studies. The effective wave propagation velocity in rock can be determined with reasonable confidence from in-situ and laboratory tests. Estimating the effective wave propagation velocity in soil overburden presents the major difficulty. Previous studies have shown that, except possibly for vertically propagating shear waves, the use of soil properties in deriving the wave velocity in soil overburden may be overly conservative. It has been suggested that for horizontally or obliquely propagating waves the propagation velocities in soil overburden are affected significantly by the velocities in the underlying rock. That is to say, the actual velocity values in the soils may be much higher than those calculated based on the soil properties alone (Hadjian and Hadley, 1981). This phenomenon is attributable to the problem of deformation compatibility. The motion of a soil particle due to a horizontally propagating wave above the rock cannot differ greatly from the motion of the rock, unless the soil slides on top of the rock (a very unlikely occurrence) or the soil liquifies. For a very deep (thick) soil stratum, however, the top of the soil stratum is less coupled to the rock and is more free to follow a motion that is determined by its own physical properties.
33
Wave Type
e=
e max =
e=
e max =
VR for q = 0 CR
Radius of Curvature
VS, VR =
Peak Particle Velocity for Shear Wave and Rayleigh Wave, respectively
CS, CR =
Effective Propagation Velocity for Shear Wave and Rayleigh Wave, respectively
AS, AR =
Peak Particle Acceleration for Shear Wave and Rayleigh Wave, respectively
35
Other assumptions and parameters used in this example are: Design Earthquake Parameters: Peak Ground Acceleration, As = 0.6 (Maximum Design Earthquake, MDE) Peak Ground Velocity, Vs = 3.2 ft/sec Soil surrounding Tunnel: Fernando Formation Effective Shear Wave Velocity: Cs = 1360 ft/sec (in Fernando Formation under MDE) Tunnel Structure: Cast-in-place circular segmented reinforced lining, with Radius R =10 feet From Table 1, the combined maximum axial strain and curvature strain would be:
e max =
36
Applicability of the Free-Field Deformation Approach The example presented above demonstrates the simplicity of the free-field deformation approach. Because it is an upper-bound assessment of the tunnel response, it often becomes the first tool an engineer would use to verify the adequacy of his design. This approach offers a method for verification of a design rather than a design itself. Note, however, that this method is: Pertinent to a tunnel structure that is flexible relative to its surrounding medium, such as all tunnels in rock and most tunnels in stiff soils. In this case it is reasonable to assume that the tunnel deforms according to its surrounding medium. Not desirable for situations involving stiff structures buried in soft soil, because under this condition, the calculated ground deformations may be too great (due to the soft nature of the soil) for the stiff structures to realistically accommodate. Once the calculated ground strain exceeds the allowable strain of the lining material, there is very little an engineer can do to improve his design. For instance, if the effective shear wave velocity of the previous example is reduced to 350 ft/sec to reflect a much softer soil deposit, the tunnel lining will then be subjected to a combined maximum axial strain of 0.0052 in compression (see Design Example 2 in the next section). It is essentially unrealistic to provide an adequate concrete lining design resisting an axial strain of this amount. If the free-field deformation approach were used in this case, it appears that the only solution to this problem would be to provide needless flexible joints, forming a chainlink-like tunnel structure to accommodate the ground deformation. In the next section, a design approach considering the tunnel-ground interaction effect is presented. This design approach, based on results from previous studies, may effectively alleviate the design difficulty discussed above.
1987). In general, the tunnel-ground system is simulated as an elastic beam on an elastic foundation, with the theory of wave propagating in an infinite, homogeneous, isotropic medium. When subjected to the axial and curvature deformations caused by the traveling waves in the ground, the tunnel will experience the following sectional forces (see Figure 6): Axial forces, Q, on the cross-section due to the axial deformation Bending moments, M, and shear forces, V, on the cross-section due to the curvature deformation Simplified Interaction Equations Maximum Axial Force: Qmax. Through theoretical derivations, the resulting maximum sectional axial forces caused by a shear wave with 45 degree angle of incidence can be obtained:
(Eq. 3-1)
Ka =longitudinal spring coefficient of medium (in force per unit deformation per unit length of tunnel) D = free-field displacement response amplitude of an ideal sinusoidal shear wave Ec = modulus of elasticity of tunnel lining Ac = cross-section area of tunnel lining The calculated maximum axial force, Qmax, shall not exceed an upper limit defined by the ultimate soil drag resistance in the longitudinal direction. This upper limit is expressed as:
Q limit =
fL 4
(Eq. 3-2)
where f = ultimate friction force (per unit length of tunnel) between the tunnel and the surrounding medium
38
Maximum Bending Moment, Mmax. The bending moment resulting from curvature deformations is maximized when a shear wave is traveling parallel to the tunnel axis (i.e., with an angle of incidence equal to zero). The mathematical expression of the maximum bending moment is:
L 2 2p M max = D Kt L 4 1+ EcIc 2p
Kt
(Eq. 3-3)
where
L, Ec and D are as defined in Equation 3-1 Ic = moment of inertia of the tunnel section Kt = transverse spring coefficient of medium (in force per unit deformation per unit length of tunnel).
Maximum Shear Force, Vmax. The maximum shear force corresponding to the maximum bending moment is derived as:
(Eq. 3-4)
The tunnel-ground interaction effect is explicitly accounted for in these formulations. The ground stiffness and the tunnel stiffness are represented by spring coefficients (Ka or Kt) and sectional modulus (EcAc or EcIc), respectively. The application of these equations is necessary only when tunnel structures are built in soft ground. For structures in rock or stiff soils, the evaluation based on the freefield ground deformation approach presented in Section 3.3 will, in general, be satisfactory. Equations 3-1, 3-3 and 3-4 are general mathematical forms. Other expressions of the maximum sectional forces exist in the literature. The differences are primarily due to the further maximization of the sectional forces with respect to the wavelength, L. For instance:
40
- In the JSCE (Japanese Society of Civil Engineers) Specifications for Earthquake Resistant Design of Submerged Tunnels, the values of wavelength that will maximize Equations 3-1, 3-3 and 3-4 are determined and substituted back into each respective equation to yield the maximum sectional forces. - St. John and Zahran (1987) suggested a maximization scheme that is similar to the Japanese approach except that the spring coefficients (Ka or Kt) are assumed to be functions of wavelength, L, in the maximization process. Both of these approaches assume that the free-field ground displacement response amplitude, D, is independent of the wavelength. This assumption sometimes may lead to very conservative results, as the ground displacement response amplitude generally decreases with the wavelength. It is, therefore, the authors view that Equations 3-1 through 3-4 presented in this section will provide a practical and adequate assessment, provided that the values (or the ranges of the values) of L, D, and Kt (or Ka) can be reasonably estimated. A reasonable estimate of the wavelength can be obtained by
L = T Cs
(Eq. 3-5)
where T is the predominant natural period of the shear wave traveling in the soil deposit in which the tunnel is built, and Cs is the shear wave propagation velocity within the soil deposit. Often, T can also be represented by the natural period of the site. Dobry, Oweis and Urzua (1976) presented some procedures for estimating the natural period of a linear or equivalent linear model of a soil site. The ground displacement response amplitude, D, should be derived based on sitespecific subsurface conditions by earthquake engineers. The displacement amplitude represents the spatial variations of ground motions along a horizontal alignment. Generally, the displacement amplitude increases as the wavelength, L, increases. For example, the displacement spectrum chart prepared by Housner (SFBARTD, 1960) for the SF BART project was expressed by D = 4.9 x 10-6 L1.4, where the units of D and L are in feet. This spectrum is intended for tunnel tubes in soft San Francisco Bay muds and was derived for a magnitude 8.2 earthquake on the San Andreas fault. The equation shows clearly that: - The displacement amplitude increases with the wavelength. - For any reasonably given wavelength, the corresponding ground displacement
41
amplitude is relatively small. Using the given wavelength and the corresponding displacement amplitude, the calculated free-field ground strains would be significantly smaller than those calculated using the simplified equations shown in Table 1. This suggests that it may be overly conservative to use the simplified equations to estimate the axial and curvature strains caused by seismic waves travelling in soils for tunnel design. With regard to the derivations of spring coefficients Ka and Kt, there is no consensus among design engineers. The derivations of these spring coefficients differ from those for the conventional beam on elastic foundation problems in that: -The spring coefficients should be representative of the dynamic modulus of the ground under seismic loads. -The derivations should consider the fact that loading felt by the surrounding soil (medium) is alternately positive and negative due to the assumed sinusoidal seismic wave. Limited information on this problem is available in the literature (SFBARTD 1960, St. John and Zahrah, 1987 and Owen and Scholl, 1981). For preliminary design, it appears that the expressions suggested by St. John and Zahrah (1987) should serve the purpose:
Kt = Ka =
where
16pGm (1 - vm ) d (3 - 4vm ) L
(Eq. 3-6)
Gm = shear modulus of the medium (see Section 4.2 in Chapter 4) nm = Poissons radio of the medium d = diameter (or equivalent diameter) of the tunnel L = wavelength
A review of Equations 3-1, 3-3 and 3-4 reveals that increasing the stiffness of the structure (i.e., EcAc and EcIc), although it may increase the strength capacity of the structure, will not result in reduced forces. In fact, the structure may attract more forces as a result. Therefore, the designer should realize that strengthening of an overstressed section by increasing its sectional dimensions (e.g., lining thickness) may not always provide an efficient solution for seismic design of tunnel structures. Sometimes, a more flexible configuration with adequate reinforcements to provide sufficient ductility is a more desirable measure.
42
Design Example 2: A Linear Tunnel in Soft Ground In this example, a tunnel lined with a cast-in-place circular concrete lining (e.g., a permanent second-pass support) is assumed to be built in a soft soil site. The geotechnical, structural and earthquake parameters are listed as follows: Geotechnical Parameters: - Effective shear wave velocity, CS =350 ft/sec. - Soil unit weight, gt =110 pcf =0.110 kcf. - Soil Poissons ratio, nm =0.5 (saturated soft clay). - Soil deposit thickness over rigid bedrock, H =100 ft. Structural Parameters: - Lining thickness, t =1 ft. - Lining diameter, d =20 ft. - Lining moment of inertia, Ic = 0.5 x 3148 = 1574 ft4 (one half of the full section moment of inertia to account for concrete cracking and nonlinearity during the MDE). - Lining cross section area, Ac =62.8 ft2. - Concrete Youngs Modulus, Ec =3600 ksi =518400 ksf. - Concrete yield strength, fc =4000 psi. - Allowable concrete compression strain under combined axial and bending compression, eallow = 0.003 (during the MDE) Earthquake Parameters (for the MDE): - Peak ground particle acceleration in soil, As =0.6 g. - Peak ground particle velocity in soil, Vs =3.2 ft/sec.
43
First, try the simplified equation as used in Design Example 1. The combined maximum axial strain and curvature strain is calculated as:
. 3.2 0.6 x32.2 x10 cos 3 45 o e max = V s AsR cos 3 q = 2 Cs 2 x 350 ( 350) 2 2Cs = 0.0046 0.0006 = 0.0052
The calculated maximum compression strain exceeds the allowable compression strain of concrete (i.e., emax > eallow = 0.003).
Now use the tunnel-ground interaction procedure. 1. Estimate the predominant natural period of the soil deposit (Dobry, et al, 1976).
T=
2.
L = TxCs = 4H = 400 ft
3. Estimate the shear modulus of soil:
Gm = rCs 2 =
4.
Ka = Kt =
44
5.
As discussed before, the ground displacement amplitude is generally a function of the wavelength, L. A reasonable estimate of the displacement amplitude must consider the site-specific subsurface conditions as well as the characteristics of the input ground motion. In this design example, however, the ground displacement amplitudes are calculated in such a manner that the ground strains as a result of these displacement amplitudes are comparable to the ground strains used in the calculations based on the simplified free-field equations. The purpose of this assumption is to allow a direct and clear evaluation of the effect of tunnel-ground interaction. Thus, by assuming a sinusoidal wave with a displacement amplitude D and a wavelength L, we can obtain:
6.
Calculate the maximum axial force (Equation 3-1) and the corresponding axial strain of the tunnel lining:
Q max
eaxial =
45
7.
Calculate the maximum bending moment (Equation 3-3) and the corresponding bending strain of the tunnel lining:
M max
ebending =
= 0.00051
8. Compare the combined axial and bending compression strains to the allowable:
V max = M max x
46
10. Calculate the allowable shear strength of concrete during the MDE:
= shear strength reduction factor (0.85) fc = yield strength of concrete (4000 psi) Ashear = effective shear area = Ac/2
Ashear
11. Compare the induced maximum shear force with the allowable shear resistance:
47
Unstable Ground It is generally not feasible to design a tunnel lining of sufficient strength to resist large permanent ground deformations resulting from an unstable ground. Therefore, the proper design measures in dealing with this problem should consider the following: Ground stabilization (e.g., compaction, draining, reinforcement, grouting, and earth retaining systems) Removal and replacement of problem soils Reroute or deeper burial
Faulting With regard to fault displacements, the best solution is to avoid any potential crossing of active faults. If this is not possible, the general design philosophy is to design a tunnel structure to accept and accommodate these fault displacements. For example, in the North Outfall Replacement Sewer (NORS, City of Los Angeles) project, the amount of fault displacement associated with an M=6.5 design earthquake on the Newport-Inglewood fault was estimated to be about 8 inches at the crossing. To accommodate this displacement, a design scheme using an oversized excavation and a compressible backpacking material was provided. The backpacking material was designed to withstand the static loads, yet be crushable under faulting movements to protect the pipe.
48
It is believed that the only transportation tunnel in the U.S. designed and constructed to take into consideration potential active fault displacements is the Berkeley Hills Tunnel, part of the San Francisco BART system. This horse-shoe-shaped tunnel was driven through the historically active creeping Hayward Fault with a one-foot oversized excavation. The purpose of the over-excavation was to provide adequate clearance for rail passage even when the tunnel was distorted by the creeping displacements. Thus rails in this section could be realigned and train services could be resumed quickly afterward. The tunnel was lined with concrete encased ductile steel ribs on two-foot centers. The concrete encased steel rib lining is particularly suitable for this design because it provides sufficient ductility to accommodate the lining distortions with little strength degradation. The two projects described above have several common design assumptions that allowed the special design to be feasible both technically and economically: The locations of the faults at crossings can be identified with acceptable uncertainty, limiting the lengths of the structures that require such special design. The design fault displacements are limited to be within one foot. The cost associated with special design may become excessively high when significant uncertainty exists in defining the activities and locations of the fault crossings, or when the design fault displacements become large (e.g., five feet). Faced with these situations, designers as well as owners should re-evaluate and determine the performance goals of the structures based on a risk-cost balanced consideration, and design should be carried out accordingly.
Abrupt Changes in Structural Stiffness or Ground Conditions These conditions include, but are not limited to, the following: When a regular tunnel section is connected to a station end wall or a rigid, massive structure such as a ventilation building At the junctions of tunnels When a tunnel traverses two distinct geological media with sharp contrast in stiffness When tunnels are locally restrained from movements by any means (i.e., hard spots)
49
Generally, the solutions to these interface problems are to provide either of the following: A movable joint, such as the one used at the connection between the Trans-Bay Tube and the ventilation building (Warshaw, 1968) A rigid connection with adequate strength and ductility At these critical interfaces, structures are subjected to potential differential movements due to the difference in stiffness of two adjoining structures or geological media. Estimates of these differential movements generally require a dynamic analysis taking into account the soil-structure interaction effect (e.g., SFBARTD, 1991). There are cases where, with some assumptions, a simple free-field site response analysis will suffice. The calculated differential movements provide necessary data for further evaluations to determine whether special seismic joints are needed. Once the differential movements are given, there are some simple procedures that may provide approximate solutions to this problem. For example, a linear tunnel entering a large station may experience a transverse differential deflection between the junction and the far field due to the large shear rigidity provided by the end wall of the station structure. If a conventional design using a rigid connection at the interface is proposed, additional bending and shearing stresses will develop near the interface. These stress concentrations can be evaluated by assuming a semi-infinite beam supported on an elastic foundation, with a fixed end at the connection. According to Yeh (1974) and Hetenyi (1976), the induced moment, M(x), and shear, V(x), due to the differential transverse deflection, d, can be estimated as:
M(x) =
where x = distance from the connection Ic = moment of inertia of the tunnel cross section Ec = Youngs modulus of the tunnel lining Kt = transverse spring coefficient of ground (in force per unit deformation per unit length of tunnel)
50
Based on Equations 3-7 and 3-8, the maximum bending moment and shear force occur at x=0 (i.e., at the connection). If it is concluded that an adequate design cannot be achieved by using the rigid connection scheme, then special seismic (movable) joints should be considered.
51
52
53
54
4.1
Ovaling Effect
As defined in Chapter 3, ovaling of a circular tunnel lining is primarily caused by seismic waves propagating in planes perpendicular to the tunnel axis (see Figure 2). Usually, it is the vertically propagating shear waves that produce the most critical ovaling distortion of the lining. The results are cycles of additional stress concentrations with alternating compressive and tensile stresses in the tunnel lining. These dynamic stresses are superimposed on the existing static state of stress in the lining. Several critical modes may result (Owen and Scholl, 1981): Compressive dynamic stresses added to the compressive static stresses may exceed the compressive capacity of the lining locally. Tensile dynamic stresses subtracted from the compressive static stresses reduce the linings moment capacity, and sometimes the resulting stresses may be tensile.
4.2
As discussed in Chapter 3, the shear distortion of ground caused by vertically propagating shear waves is probably the most critical and predominant mode of seismic motions in many cases. It causes a circular tunnel to oval and a rectangular underground structure to rack (sideways motion), as shown in Figure 3. Analytical procedures by numerical methods are often required to arrive at a reasonable estimate of the free-field shear distortion, particularly for a soil site with variable stratigraphy. Many computer codes with variable degree of sophistication are available (e.g., SHAKE, 1972; FLUSH, 1975; and LINOS, 1991).
55
The most widely used approach is to simplify the site geology into a horizontally layered system and to derive a solution using one-dimensional wave propagation theory (Schnabel, Lysmer, and Seed, 1972). The resulting free-field shear distortion of the ground from this type of analysis can be expressed as a shear strain distribution or shear deformation profile versus depth. An example of the resulting free-field shear distortion for a soil site using the computer code SHAKE is presented in Figure 7. Simplified Equation for Shear Deformations For a deep tunnel located in relatively homogeneous soil or rock, the simplified procedure by Newmark (presented in Table 1) may also provide a reasonable estimate. Here, the maximum free-field shear strain, gmax, can be expressed as:
g max =
where
Vs Cs
(Eq. 4-1)
The values of Cs can be estimated from in-situ and laboratory tests. An equation relating the effective propagation velocity of shear waves to effective shear modulus, Gm , is expressed as:
Cs =
where
Gm r
(Eq. 4-2)
It is worth noting that both the simplified procedure and the more refined SHAKE analysis require the parameters Cs or Gm as input. The propagation velocity and the shear modulus to be used should be compatible with the level of shear strains that may develop in the ground under design earthquake loading. This is particularly critical for soil sites due to the highly non-linear behavior of soils. The following data are available: Seed and Idriss (1970) provide an often used set of laboratory data for soils giving the effective shear wave velocity and effective shear modulus as a function of shear strain. Grant and Brown (1981) further supplemented the data sets with results from a series of field geophysical measurements and laboratory testing conducted for six soil sites.
56
Figure 7. Free-Field Shear Distortions of Ground Under Vertically Propagating Shear Waves
57
g DD = max 2 D
where D = the diameter of the tunnel gmax = the maximum free-field shear strain
(Eq. 4-3)
On the other hand, if the ground deformation is derived by assuming the presence of a cavity due to tunnel excavation (Figure 9, for perforated ground), then the lining is to be designed according to the diametric strain expressed as:
DD = 2g max (1 - vm ) D
where
(Eq. 4-4)
Equations 4-3 and 4-4 both assume the absence of the lining. In other words, tunnelground interaction is ignored. Comparison between Equations 4-3 and 4-4 shows that the perforated ground deformation would yield a much greater distortion than the non-perforated, free-field ground deformation. For a typical ground medium, an engineer may encounter solutions provided by Equations 4-3 and 4-4 that differ by a ratio ranging from 2 to about 3. By intuition:
58
59
Equation 4-4, the perforated ground deformation, should serve well for a lining that has little stiffness (against distortion) in comparison to that of the medium. Equation 4-3, on the other hand, should provide a reasonable distortion criterion for a lining with a distortion stiffness equal to the surrounding medium. It is logical to speculate further that a lining with a greater distortion stiffness than the surrounding medium should experience a lining distortion even less than that calculated by Equation 4-3. This latest case may occur when a tunnel is built in soft to very soft soils. The questions that may be raised are: How important is the lining stiffness as it pertains to the engineering design? How should the lining stiffness be quantified relative to the ground? What solutions should an engineer use when the lining and ground conditions differ from those where Equations 4-3 and 4-4 are applicable? In the following sections (4.4 and 4.5), answers to these questions are presented.
4.4
Compressibility and Flexibility Ratios To quantify the relative stiffness between a circular lining and the medium, two ratios designated as the compressibility ratio, C, and the flexibility ratio, F (Hoeg, 1968, and Peck et al., 1972) are defined by the following equations:
Compressibility Ratio, C =
(Eq. 4-5)
Flexibility Ratio,
where
F=
(Eq. 4-6)
Em = modulus of elasticity of the medium nm = Poissons Ratio of the medium El = the modulus of elasticity of the tunnel lining nl = Poissons Ratio of the tunnel lining
60
R = radius of the tunnel lining t = thickness of the tunnel lining I = moment of inertia of the tunnel lining (per unit width) Of these two ratios, it is often suggested that the flexibility ratio is the more important because it is related to the ability of the lining to resist distortion imposed by the ground. As will be discussed later in this chapter, the compressibility ratio also has an effect on the lining thrust response. The following examples on the seismic design for several tunnel-ground configurations are presented to investigate the adequacy of the simplified design approach presented in the previous section.
Example 1 The first illustrative example is a tunnel cross-section from the LA Metro project. The ground involved is an old alluvium deposit with an effective shear wave propagation velocity, Cs, equal to 1000 ft/sec. The peak shear wave particle velocity, Vs, according to the design criteria, is 3.4 ft/sec. Using Equation 4-1, the maximum free-field shear strain, gmax , is calculated to be 0.0034. The reinforced cast-in-place concrete lining properties and the soil properties are assumed and listed in the following table. Lining Properties Soil Properties
R = 9.5 feet Em = 7200 ksf t = 8.0 inches nm = 0.333 El/(1- nl 2) = 662400 ksf I = 0.0247 ft4/ft Flexibility Ratio, F = 47 Compressibility Ratio, C = 0.35 Note that uncertainties exist in the estimates of many of the geological and structural parameters. For instance: The effective shear wave propagation velocity in the old alluvium may have an uncertainty of at least 20 percent. Uncertainty up to 40 percent may also be applied to the estimates of Em.
61
The moment of inertia, I, for a cracked lining section, or for a segmental lining with staggered joints in successive rings, may be considerably less than that for the typical cross section of a segment as used in this calculation example. (See Section 4.5 for a means of estimating the effective moment of inertia, Ie.) It would be desirable, therefore, to define the ranges of the values considering these uncertainties in the actual design cases. The LA Metro project has adopted Equation 4-4 as the criterion for ovaling of the running lines (SCRTD, 1984). Therefore, a maximum diametric strain, DD/D, of 0.00453 is obtained. The maximum combined bending strain and thrust compression strain as a result of this diametric strain is calculated, with some simple assumptions based on ring theory, by using the following formulation:
etotal =
1 R Em(1- v 12 ) Vs t 1- 3( v m) + 2 t E1(1+ v m) Cs R
(Eq. 4-7)
= 0.00061
To verify the accuracy of the results, a numerical analysis using finite difference code is performed. No-slip interface between the lining and the surrounding ground is assumed in the analysis. A more detailed description of this modeling is presented in Section 4.5. Results from the finite difference analysis yield: A maximum diametric strain of 0.0038 A combined maximum total compression strain in the lining of about 0.0006 The excellent agreement between the simplified approach using Equation 4-4 and the refined numerical analysis is explained by the flexibility ratio (F=47) of the ground-lining system. A flexibility ratio of this magnitude suggests that the lining should be flexible even when compared to ground with a cavity in it, and therefore should conform to the perforated ground deformation. Example 2 In this example, the tunnel is assumed to be built in a very soft soil deposit. The cross-sectional properties of the lining and the surrounding ground are shown in the following table. Note that these properties are made in order to result in a flexibility ratio equal to 1.0.
62
Lining Properties
Soil Properties
R = 10 feet Em = 325 ksf t = 12 inches nm = 0.25 El/(1-nl2) = 518400 ksf I = 0.0833 ft4/ft Flexibility Ratio, F = 1.0 Compressibility Ratio, C = 0.01 It is further assumed that the free-field maximum shear strain, gmax = 0.008, is obtained from one-dimensional site response analysis using SHAKE program. If Equation 4-4 is used, the maximum diametric strain, DD/D , of the lining is calculated to be 0.012. With this diameter change, the lining will be subject to a maximum bending strain of approximately 0.0018 together with an almost negligible amount of thrust compression strain. This additional strain, when superimposed on the existing strain caused by the static load, may exceed the compression capacity of the concrete. It is questionable, however, that designing the lining to conform to the perforated ground deformation (Equation 4-4) is adequate in this case. Flexibility ratio equal to 1.0 implies that the lining may just have enough stiffness to replace that of the soil being excavated. Ideally, the lining should distort in accordance with the free-field, nonperforated ground deformation (Equation 4-3). With this assumption, the maximum diametric strain according to Equation 4-3 is 0.004, a value only one-third of that calculated by Equation 4-4. A computation by finite difference code is performed for comparison. The resulting maximum diametric strain is about 0.0037, which supports the suggestions made immediately above. Summary and Conclusions In conclusion, the simplified seismic design approach can serve its purpose, provided that good judgment is used during the design process. The ovaling effects on the lining, however, may in some cases be overestimated or underestimated, depending on the relative stiffness between the ground and the lining. The main reason for this drawback is the uncertainty of the tunnel-ground interaction. This drawback, however, may be immaterial for most applications in the real world. For most circular tunnels encountered in practice, the flexibility ratio, F, is likely to be large enough (F>20) so that the tunnel-ground interaction effect can be ignored (Peck, 1972). In these cases, the distortions to be experienced by the lining can be reasonably assumed to be equal to those of the perforated ground.
63
This rule of thumb procedure may present some design problems in the real world too. These problems arise when a very stiff structure is surrounded by a very soft soil. A typical example would be to construct a very stiff immersed tube in a soft lake or river bed. In this case the flexibility ratio is very low, and the tunnel-ground interaction must be considered to achieve a more efficient design. In the following section a refined procedure, equally simple, if not simpler, will be presented. This refined procedure considers the tunnel-ground interaction effect and provides a more accurate assessment of the seismic effect upon a circular lining.
4.5
Lining-Ground Interaction
Closed Form Solutions Closed form solutions for estimating ground-structure interaction for circular tunnels have been proposed by many investigators. These solutions are commonly used for static design of tunnel lining. They are generally based on the assumptions that: The ground is an infinite, elastic, homogeneous, isotropic medium. The circular lining is generally an elastic, thin walled tube under plane strain conditions. The models used in these previous studies vary in the following two major assumptions, the effects of which have been addressed by Mohraz et al. (1975) and Einstein et al. (1979): Full-slip or no-slip conditions exist along the interface between the ground and the lining. Loading conditions are to be simulated as external loading (overpressure loading) or excavation loading. Most of the recent developments in these models fall into the category of excavation loading conditions, as they represent a more realistic simulation of actual tunnel excavation (Duddeck and Erdmann, 1982). To evaluate the effect of seismic loading, however, the solutions for external loading should be used. Peck, Hendron, and Mohraz (1972), based on the work by Burns and Richard (1964) and Hoeg (1968), proposed closed form solutions in terms of thrusts, bending moments and displacements under external loading conditions.
64
The expressions of these lining responses are functions of flexibility ratio and compressibility ratio as presented previously in Equations 4-5 and 4-6. The solutions also depend on the in-situ overburden pressure, gtH, and the at rest coefficient of earth pressure, Ko. To be adapted to the loading caused by seismic shear waves, it is necessary to replace the in-situ overburden pressure with free-field shear stress, t and , assign Ko=1, to simulate the simple shear condition in the field. The shear stress, t can , be expressed as a function of shear strain, g. With some mathematical manipulations, the resulting expressions for maximum thrust, Tmax, bending moment, Mmax, and diametric strain, DD/D, can be presented in the following forms:
T max =
1 Em K1 Rg max 6 (1 + vm )
(Eq. 4-8)
(Eq. 4-9)
(Eq. 4-10)
where
K1 =
(Eq. 4-11)
where
Em, nm = modulus of elasticity and Poissons Ratio of medium R = radius of the tunnel lining gmax = maximum free-field shear strain F = flexibility ratio
K1 is defined herein as lining response coefficient. The earthquake loading parameter is represented by the maximum shear strain, gmax, which may be obtained through a simplified approach (such as Equation 4-1), or by performing a site-response analysis. To ease the design process, Figures 10 and 11 show the lining response coefficient, K1, as a function of flexibility ratio and Poissons Ratio of the ground. It should be noted that the solutions provided here are based on the full-slip interface assumption.
65
Response Coefficient, K1
Response Coefficient, K1
Comments on Closed Form Solutions According to previous investigations, during an earthquake slip at interface is a possibility only for tunnels in soft soils, or when seismic loading intensity is severe. For most tunnels, the condition at the interface is between full-slip and no-slip. In computing the forces and deformations in the lining, it is prudent to investigate both cases and the more critical one should be used in design. The full-slip condition gives more conservative results in terms of maximum bending moment, Mmax , and lining deflections DD. This conservatism is desirable to offset the potential underestimation (10 to 15 percent) of lining forces resulting from the use of equivalent static model in lieu of the dynamic loading condition (Mow and Pao, 1971). Therefore, the full-slip model is adopted for the present study in evaluating the moment and deflection response of a circular tunnel lining. The maximum thrust, Tmax, calculated by Equation 4-8, however, may be significantly underestimated under the seismic simple shear condition. The full-slip assumption along the interface is the cause. Therefore, it is recommended that the no-slip interface assumption be used in assessing the lining thrust response. The resulting expressions, after modifications based on Hoegs work (Schwartz and Einstein, 1980), are:
(Eq. 4-12)
68
A review of Equation 4-12 and the expression of K2 suggests that lining thrust response is a function of compressibility ratio, flexibility ratio and Poissons Ratio. Figures 12 through 14 graphically describe their interrelationships. As the plots show: The seismically induced thrusts increase with decreasing compressibility ratio and decreasing flexibility ratio when the Poissons Ratio of the surrounding ground is less than 0.5. When the Poissons Ratio approaches 0.5 (e.g., for saturated undrained clay), the linings thrust response is essentially independent of the compressibility ratio. Figures 12 through 14, along with data contained in Figures 10 and 11 provide a quick aid for designers. The theoretical solutions and the influence of interface assumptions will be further verified for their reasonableness by numerical analysis presented in the next section. Another useful and important information, for illustration purpose, is to express the deformation ratio between the lining and the free-field as a function of flexibility ratio, F. This relationship can be obtained by dividing Equation 4-10 with Equation 4-3. The resulting expression is:
(Eq. 4-13)
The normalized lining deflection is plotted and presented in Figures 15 and 16. The results indicate that the lining tends to resist and therefore deforms less than the free-field when the flexibility ratio, F, is less than approximately 1. This situation may occur only when a stiff lining is built in soft to very soft soils. As the flexibility ratio increases, the lining deflects more than the free-field and may reach an upper limit as the flexibility ratio becomes infinitely large. This upper limit deflection is equal to the perforated ground deformations calculated by Equation 4-4, signaling a perfectly flexible lining situation. The relationship shown in Figures 15 and 16 supports and supplements the discussions presented in Examples 1 and 2 of Section 4.3.
72
(DDlining)/(DDfree-field)
Numerical Analysis A series of computer analyses using finite difference code (FLAC, 1989) is performed to verify the proposed procedure in the previous section. The mesh and the lining-ground system used in these analyses are shown in Figure 17. The assumptions made for these analyses include the following: Plane strain conditions are assumed. Seismic shear wave loading is simulated by pure shear conditions with shear stresses applied at far external boundaries. Taking advantage of the anti-symmetric loading conditions, only one quarter of the entire lining/ground system is analyzed. Rollers are provided at planes of antisymmetry. Lining is modeled by a series of continuous flexural beam elements of linear elasticity. Ground (medium) is modeled as linear elastic material. No-slip condition along the lining-ground interface is assumed. A total of 13 analyses are performed. In order to cover a wide range of possible effects of lining-ground interaction, the parameters for lining and ground are varied. Following is a list of the range of the variations: Range of Em = from 325 ksf to 72000 ksf
maximum bending moment than the no-slip interface condition. The differences are within approximately 20 percent under seismic shear loading condition. It should be realized, however, that these results are based on pseudo-static solutions that do not consider the potential dynamic amplification and stress concentrations at the tunnel excavation boundary (Mow and Pao, 1971). Previous studies suggest that a true dynamic solution would yield results that are 10 to 15 percent greater than an equivalent static solution, provided that the seismic wavelength is at least about 8 times greater than the width of the excavation (cavity). Therefore, it is prudent to adopt the more conservative full-slip assumption for the calculation of bending moments. With this more conservative assumption, the effects of stress amplification need not be considered. Maximum Lining Deflection, D Dlining. Figure 19 presents a plot of the maximum lining deflections from full-slip closed form solution versus those from no-slip finite difference analysis (noting that these lining deflections are normalized with respect to the free-field ground deflections). Similar to the discussion presented above, lining tends to oval (distort) more under the full-slip interface assumption. The differences, however, are very small. The full-slip assumption (Equation 4-10 or Equation 4-13) is recommended for calculating the lining distortion. The effects of stress amplification need not be considered when the conservative full-slip assumption is adopted. It is interesting to note from the plot that almost no difference exists between the two assumptions for Case No. 12. This can be explained by the fact that a nearly perfectly flexible lining is used and little lining-ground interaction is involved in the Case No.12 analysis. Maximum Lining Thrust, Tmax. For comparison, the maximum lining thrusts are calculated using closed form solutions for both assumptions (Equations 4-8 and 4-12). The results, along with those from the finite difference analysis, are tabulated in Table 3. The table shows excellent agreement on the thrust response between the numerical finite difference analysis and the closed form solution for the no-slip condition. It also verified that the full-slip assumption will lead to significant underestimation of the lining thrust under seismic shear condition. Therefore, it is recommended that Equation 4-12 be used for thrust calculation. To account for the dynamic stress amplification due to the opening, it is further recommended that thrusts calculated from Equation 4-12 be multiplied by a factor of 1.15 for design purpose.
79
Lining Stiffness, I. The results presented above are based on the assumption that the lining is a monolithic and continuous circular ring with intact, elastic properties. Many circular tunnels are constructed with bolted or unbolted segmental lining. Besides, a concrete lining subjected to bending and thrust often cracks and behaves in a nonlinear fashion. Therefore, in applying the results presented herewith, the effective (or, equivalent) stiffness of the lining will have to be estimated first. Some simple and approximate methods accounting for the effect of joints on lining stiffness can be found in the literature: Monsees and Hansmire (1992) suggested the use of an effective lining stiffness that is one-half of the stiffness for the full lining section. Analytical studies by Paul, et al., (1983) suggested that the effective stiffness be from 30 to 95 percent of the intact, full-section lining. Muir Wood (1975) and Lyons (1978) examined the effects of joints and showed that for a lining with n segments, the effective stiffness of the ring was:
4 2 Ie = Ij + I n
where Ie < I and n > 4 I =lining stiffness of the intact, full-section Ij = effective stiffness of lining at joint Ie = effective stiffness of lining
(Eq. 4-14)
82
83
84
5.1 General
Shallow depth transportation tunnels are often of rectangular shape and are often built using the cut-and-cover method. Usually the tunnel is designed as a rigid frame box structure. From the seismic design standpoint, these box structures have some characteristics that are different from those of the mined circular tunnels, besides the geometrical aspects. The implications of three of these characteristics for seismic design are discussed below. First, cut-and-cover tunnels are generally built at shallow depths in soils where seismic ground deformations and the shaking intensity tend to be greater than at deeper locations, due to the lower stiffness of the soils and the site amplification effect. As discussed in Chapter 2, past tunnel performance data suggest that tunnels built with shallow soil overburden cover tend to be more vulnerable to earthquakes than deep ones. Second, the dimensions of box type tunnels are in general greater than those of circular tunnels. The box frame does not transmit the static loads as efficiently as the circular lining, resulting in much thicker walls and slabs for the box frame. As a result, a rectangular tunnel structure is usually stiffer than a circular tunnel lining in the transverse direction and less tolerant to distortion. This characteristic, along with the potential large seismic ground deformations that are typical for shallow soil deposits, makes the soilstructure interaction effect particularly important for the seismic design of cut-and-cover rectangular tunnels, including those built with sunken tube method.
85
Third, typically soil is backfilled above the structure and possibly between the in-situ medium and the structure. Often, the backfill soil may consist of compacted material having different properties than the in-situ soil. The properties of the backfill soil as well as the in-situ medium should be properly accounted for in the design and analysis.
86
Wood Method Another theoretical form of dynamic earth pressure was derived by Wood (1973). By assuming infinite rigidity of the wall and the foundation, Wood derived a total dynamic
87
thrust that is approximately 1.5 to 2.0 times the thrust calculated by the Mononobe-Okabe method. Model experiments by Yong (1985) confirmed these theoretical results. This method is possibly adequate for a volume structure (e.g., a basement) resting on a very stiff/hard medium (such as rock) and rigidly braced across (e.g., by transverse shear wall diaphragms). A possible application of this method in a cut-and-cover tunnel construction is at the end walls of a subway station, where the end walls act as rigid shear wall diaphragms and prevent the structure from making sideways movements during earthquakes. For regular rectangular cross-sections under plane strain condition, the Wood theory, like the Mononobe-Okabe method, would lead to unrealistic results and is not recommended for use in typical tunnel sections with significant soil cover thickness.
Implications for Design It is logical to postulate that the presence of a rectangular frame structure in the ground will induce dynamic earth pressures acting upon the structure. This earth pressure loading, however, is in a form of complex distributions of shear stresses as well as normal pressures along the exterior surfaces of the roof, the walls and the invert. To quantify these external earth loads accurately requires a rigorous dynamic soil-structure analysis. Realizing that the overall effect of this complex external earth loading is to cause the structure to rack, engineers find it more realistic to approach the problem by specifying the loading in terms of deformations. The structure design goal, therefore, is to ensure that the structure can adequately absorb the imposed racking deformation (i.e., the deformation method), rather than using a criterion of resisting a specified dynamic earth pressure (i.e., the force method). The focus of the remaining sections of this chapter, therefore, is on the method based on seismic racking deformations.
88
Figure 20. Typical Free-Field Racking Deformation Imposed on a Buried Rectangular Frame
(Source: St. John and Zahrah, 1987) 89
San Francisco BART In his pioneering development of the seismic design criteria for the San Francisco BART subway stations, Kuesel (1969) presented this approach and developed projectspecific soil distortion profiles for design purpose. The elastic and plastic distortion limits of the reinforced concrete box structure were studied and compared to the design freefield soil distortions. For the BART project, Kuesel concluded that: The structure would have sufficient capacity to absorb the imposed free-field soil distortions elastically in most cases, and that no special provisions need be made for seismic effects. When the imposed shear distortions caused plastic rotation of joints, such joints should be designed with special structural details. The soil deformation profiles and some of the assumptions used by Kuesel at that time are applicable only for the SFBART project. The design philosophy and the general approach proposed are still valid, however, even when viewed more than two decades later.
Los Angeles Metro In setting forth the seismic design criteria for the LA Metro project, Monsees and Merritt (1991), also adopted the free-field deformation method for the racking evaluation of rectangular frame structures. They specified that joints being strained into plastic hinges should be allowed under the Maximum Design Earthquake (MDE) provided that no plastic hinge combinations were formed that could lead to a potential collapse mechanism. The acceptable and unacceptable hinging conditions specified in the LA Metro project are described in Figure 21.
Flexibility vs. Stiffness In contrast to the static design, where the loads are well defined and the analysis is based on a force method, the seismic effect based on the deformation method is highly dependent on the structural details. The seismic forces induced in structural members decrease as the structures flexibility increases. Therefore, from the seismic design standpoint it is desirable to make the structure flexible rather than to stiffen it. In
90
general, flexibility can be achieved by using ductile reinforcement at critical joints. In contrast, increasing the thickness of the members makes the structure less flexible. The special structural details suggested by Kuesel and the plastic-hinge design specified by Monsees and Merritt are in fact based on this philosophy. Another design concept that can increase the flexibility of the cut-and-cover box structure is to specify pinned connections at walls/slabs joints. This design detail becomes attractive when cofferdam retaining structures are used as permanent walls because pinned connections are less difficult to build than fixed connections in this case.
Applicability of the Free-Field Racking Method The free-field deformation method serves as a simple and effective design tool when the seismically induced ground distortion is small, for example when the shaking intensity is low or the ground is very stiff. Given these conditions, most practical structural configurations can easily absorb the ground distortion without being distressed. The method is also a realistic one when the structure, compared to its surrounding medium, is flexible. Cases arise, however, when this simple procedure leads to overly conservative design for box structures. These situations generally occur in soft soils. Seismically induced free-field ground distortions are generally large in soft soils, particularly when they are subjected to amplification effects. Ironically, rectangular box structures in soft soils are generally designed with stiff configurations to resist the static loads, making them less tolerant to racking distortions. Imposing free-field deformations on a structure in this situation is likely to result in unnecessary conservatism, as the stiff structure may actually deform less than the soft ground. An example to demonstrate the effect of structure stiffness on racking deformation is given below.
Examples Soil Parameters. In this example a simplified subsurface profile is used in the free-field deformation analysis and the soil-structure interaction analysis. Figure 22 shows the soil stratigraphy of this profile. Shear wave velocities are used to represent the stiffness of the soil layers overlying the bedrock. For parametric study purposes, the analysis is performed for two cases with the silty clay layer being represented by a shear wave velocity of:
92
254 ft/sec for case I 415 ft/sec for case II These shear wave velocities are assumed to be compatible with the shear strains the soil experiences during the design earthquake. Assuming a unit weight of 115 pcf for the silty clay, the corresponding shear moduli are: G = 230 ksf for case I G = 615 ksf for case II Figure 23 shows the shear wave velocity profiles used in the analysis. Structure Properties. A reinforced one-barrel concrete box structure with the following properties is assumed: Structure Member Side Wall Base Slab Roof Slab Elastic* Modulus(ksi) 3640 3640 3640 Moment of Inertia(ft4/ft) 42.7 51.2 51.2 Thickness (ft) 8.0 8.5 8.5 Length (ft) 26 90 90
* Plane Strain Elastic Modulus The structure members are modeled as rigid continuous beam elements under a twodimensional plane strain condition. Analytical Model. Earthquake excitation is represented by a vertically propagating shear wave accelerogram originated from the rigid bedrock. The relative geometric relationship between the soil and the tunnel structure is described in Figure 22. To assess the effect of soil-structure interaction the analysis is conducted using the dynamic finite element program FLUSH (1975). Under horizontal earthquake excitation the seismic loading condition is anti-symmetrical. Therefore, only one half of the soilstructure system need be analyzed, by imposing horizontal rollers along the vertical axis of anti-symmetry (see Figure 21). A more detailed description of the time-history finite element analysis including the input ground motions and the structural modeling will be given in Section 5.5. Results. Figure 24 shows results based on free-field analysis, ignoring the presence of structure and the opening. The free-field differential deformations between the projected
94
locations of roof and invert are approximately 0.26 inch and 0.17 inch for case I and case II respectively. When both soil and structure are included in the analysis, the calculated racking distortions (between the roof and the invert) were only about 13 percent and 32 percent of the free-field deformations for case I and case II, respectively (see Figures 25 and 26). Conclusions. The results of the analysis lead to the following conclusions: It may be very conservative to design a rectangular tunnel structure to accommodate all the shear deformations in the free-field, particularly when the structure is stiff and the surrounding ground is soft. This finding coincides with results from several previous studies (Hwang and Lysmer, 1981; and TARTS, 1989). As the relative stiffness between the soil and the structure decreases (e.g., from case II to case I), the actual structure racking deformation would also decrease, when expressed as a percentage of the free-field deformation. This suggests that the soilstructure interaction effect on the racking of a rectangular tunnel should be: -Similar to that on the ovaling of a circular tunnel (Chapter 4) -A function of the relative stiffness between the ground and the structure A series of analyses performed to define this relationship and their results are presented and discussed next.
96
Figure 24. Free-Field Shear Deformation (from Free-Field Site Response Analysis, SHAKE)
97
Figure 25. Structure Deformations vs. Free-Field Deformations, Case I (from Soil/Structure Interaction Analysis, FLUSH)
98
Figure 26. Structure Deformations vs. Free-Field Deformations, Case II (from Soil/Structure Interaction Analysis, FLUSH)
99
Factors Contributing to the Soil-Structure Interaction Effect Many factors contribute to the soil-structure interaction effect. In this study, the main factors that may potentially affect the dynamic racking response of rectangular tunnel structures are investigated. These factors are: Relative Stiffness between Soil and Structure. Based on results derived for circular tunnels (see Chapter 4), it is anticipated that the relative stiffness between soil and structure is the dominating factor governing the soil/structure interaction. Therefore, a series of analyses using ground profiles with varying properties and structures with varying racking stiffness was conducted for parametric study purpose. A special case where a tunnel structure is resting directly on stiff foundation materials (e.g., rock) was also investigated. Structure Geometry. Five different types of rectangular structure geometry were studied, including one-barrel, one-over-one two-barrel, and one-by-one twin-barrel tunnel structures. Input Earthquake Motions. Two distinctly different time-history accelerograms were used as input earthquake excitations. Tunnel Embedment Depth. Most cut-and-cover tunnels are built at shallow depths. To study the effect of the depth factor, analyses were performed with varying soil cover thickness. A total number of 36 dynamic finite element analyses were carried out to account for the variables discussed above.
Method of Analysis Computer Program. The dynamic finite element analyses were performed using the computer code FLUSH (1975), a two-dimensional, plane strain, finite element program in frequency domain. Besides calculating the internal forces in the structure members, FLUSH analysis: Produces data in the form of maximum relative movements between any two locations within the soil/structure system being analyzed Allows a simultaneous free-field response analysis and compares the relative movement between any two locations in the soil/structure system and in the free field
100
These features are ideal for this study because design of the tunnel structures is based on the deformation method. A detailed description of this program can be found in Lysmer, et al. (1975). Soil-Structure Model. Figure 27 shows the typical soil-structure finite element model used. The assumptions related to the model were as follows: The structure members are modeled by continuous flexural beam elements of linear elasticity. Structural frames with rigid connections are considered. A rigid base underlies the soil (medium) deposit. The soil overburden generally consists of a soft layer overlying a stiffer layer. Except for 7 cases where the top of the stiffer layer is raised to the invert elevation (to study the effect of stiff foundation), all cases assume the stiffer layer is below the base of the structure by a vertical distance of at least one time the full height of the structure. Materials of both layers are linearly elastic. No-slip condition along the soil/structure interface is assumed. Taking advantage of the anti-symmetric loading condition, only one half the entire soil/structure system is analyzed. Horizontal rollers are provided at planes of antisymmetry. To minimize the boundary effect on the geometric dissipation of seismic energy, an energy absorbing boundary is placed at the far side of the mesh (i.e., transmitting boundary). Earthquake Accelerograms. The two digitized ground motion accelerograms employed in the analyses (see Figures 28A and 28B) were generated synthetically from the two sets of design response spectra presented in Figure 29. The following should be noted: The W. EQ spectra and the corresponding accelerogram represent the rock outcrop ground motions that are typical in the western states of the United States. They were obtained from the San Francisco BART extension project. The N.E. EQ spectra and the corresponding accelerogram represent rock outcrop earthquake motions in the northeastern part of the country. They are taken from the Seismic Design Criteria of Underground Structures for the Boston Central Artery and Third Harbor Tunnel project (1990). Horizontal earthquake accelerograms are input at the rigid base to simulate the vertically propagating shear waves.
101
As Figures 28A and 28B show, earthquake motions of these two types have very different frequency characteristics, with the N.E. EQ motions displaying significantly increased high frequency components. The purpose of using two sets of design response spectra instead of one was to evaluate the effect of ground motion characteristics on soil/structure interaction. Note that these design spectra were developed for motions expected at rock outcrop (ground surface). For motions to be used as rigid base input in the FLUSH analysis, a suitable modification of ground motion characteristics should be made. This was achieved in this study by using the one-dimensional site response analysis program SHAKE based on wave propagation theory. Details of this de-convolution process can be found in Schnabel, et al.(1972).
Flexibility Ratio for Rectangular Tunnels Figure 30 shows the five different types of structure configurations that were analyzed. Note that although the configurations were limited to five types, the racking stiffness of each structure type was varied further (for parametric studies) by varying the properties of the structure members (e.g., EI and EA values). Similarly, the stiffness of the surrounding soil, as represented by shear modulus, was also varied in such a manner that the resulting relative stiffness between the soil medium and the structure covered a range that was of interest. This relative stiffness, as represented by the Flexibility Ratio, F, will be defined in detail in the following paragraphs. The flexibility ratio for a rectangular tunnel, just as for a circular tunnel, is a measure of the flexural stiffness of the medium relative to that of the tunnel structure. Under a seismic simple shear condition, this relative stiffness may be translated into the shear stiffness of the medium relative to the lateral racking stiffness of the rectangular frame structure. General Cases. Consider a rectangular soil element in a soil column under simple shear condition (see Figure 31). Assume the soil element has a width, L, and a height, H, that are equal to the corresponding dimensions of the rectangular tunnel. When subjected to the simple shear stress, t, the shear strain (or angular distortion, g) of the soil element is given by:
g=
D t = H G
(Eq. 5-1)
Acceleration (g)
Acceleration (g)
Figure 29. Design Response Spectra (West Coast Earthquake vs. Northeast Earthquake)
106
Figure 31. Relative Stiffness Between Soil and a Rectangular Frame (from Soil/Structure Interaction Analysis, FLUSH)
108
The shear (or flexural) stiffness of the soil element is taken as the ratio of the shear stress to the corresponding angular distortion as expressed by:
t t = =G g D/H
(Eq. 5-2)
When the rectangular frame structure is subjected to the same shear stress, t, the stress can be converted into a concentrated force, P, by multiplying the shear stress by the width of the structure (P= tL). The resulting expression for the angular distortion of the structure becomes:
g=
D P tL = = H HS1 HS1
(Eq. 5-3)
where
The flexural (or, racking) stiffness of the structure is, therefore, given by:
t t SH = = 1 g D/H L
(Eq. 5-4)
The flexibility ratio, F, is obtained by dividing Equation 5-2 by Equation 5-4. The resulting expression is:
F=
GL S1 H
(Eq. 5-5)
In the expression above, the unit racking stiffness, S1, is simply the reciprocal of lateral racking deflection, S1=1/D 1 caused by a unit concentrated force (i.e., p=1 in Figure 32A). For a rectangular frame with arbitrary configuration, the flexibility ratio can be determined by performing a simple frame analysis using conventional frame analysis programs such as STAAD-III (see Figure 32A). Additional effort required to perform this type of analysis should be minimal as most of the computer input is readily established for static design. Special Case 1. For some of the simple one-barrel frames (Figure 32B), it is possible to derive the flexibility ratio without resorting to computer analysis. The expression of F
109
developed for a one-barrel frame with equal moment of inertia, IL, for roof and invert slabs and equal moment of inertia, IH, for side walls is given by:
(Eq. 5-6)
E = plane strain elastic modulus of frame G = shear modulus of soil IL, IH = moments of inertia per unit width for slabs and walls, respectively
Note that the expressions by Equation 5-6 and Equation 5-7 that follow are valid only for homogeneous, continuous frames with rigid connections. Reinforced framed concrete structures are examples of this type of construction. Special Case 2. The flexibility ratio derived for a one-barrel frame with roof slab moment of inertia, IR, invert slab moment of inertia, II, and side wall moment of inertia, IW , is expressed as:
F=
G HL2 EIR Y 12
(Eq. 5-7)
where
Y =
I a1 = R and II
I H a2 = R IW L
E = plane strain elastic modulus of frame G = shear modulus of soil IR, II, IW = moments of inertia per unit width Implications of Flexibility Ratios. The derivation of the flexibility ratio presented in this section is consistent with that for the circular tunnels. The theoretical implications are: A flexibility ratio of 1.0 implies equal stiffness between the structure and the ground. Thus, the structure should theoretically distort the same magnitude as estimated for the ground in the free-field.
110
Figure 32. Determination of Racking Stiffness (from Soil/Structure Interaction Analysis, FLUSH)
111
For flexibility ratios less than 1.0, the structure is considered stiff relative to the freefield and should distort less. An infinitely large flexibility ratio represents a perfectly flexible structure. At this state, the deformed shape of the structure should be identical to that of a perforated ground. The size and shape of the perforation, of course, should match the structure.
Results of Analysis Analyses were first performed for 25 cases of soil/structure systems with varying combinations of soil profile, structure configuration, input ground motion type and flexibility ratio. Table 4 lists the details of the combinations for all 25 cases. Note that: The backfilled overburden thickness (soil cover) used in these analyses was limited to a range between 15 and 22.5 feet. The soil medium surrounding the embedded structure was assumed to be homogeneous, except for Cases 10, 14 and 15 where a soil profile with linearly increasing shear modulus with depth was assumed. An average soil shear modulus taken at the mid-height of the structure was used to represent the soil stiffness and to calculate the flexibility ratio for these three cases. For each of the 25 cases, a free-field site response analysis (i.e., with no structure and no opening in ground) was first performed, followed subsequently by a corresponding soil/structure interaction analysis. The free-field site response analysis calculated the freefield shear deformation of the ground, g free-field, at the depth where the structure was to be placed, specifically, the differential shear distortion between the projected locations of the roof and the invert. The corresponding soil/structure interaction analysis then calculates the actual racking distortion, g s, of the structure. Racking Coefficient. A racking coefficient, R, defined as the normalized structure racking distortion with respect to the free-field ground distortion is given as:
R=
gs gfree- field
(Eq. 5-8)
112
where
gs = angular distortion of the structure D s = lateral racking deformation of the structure gfree-field = shear distortion/strain of the free-field D free-field = lateral shear deformation of the free-field
The racking coefficients, R, obtained from the analyses are presented in the last column of Table 4 for all 25 cases. Note that the total structural deformation obtained from the finite element analyses contains a rigid body rotational movement, which causes no distortion to the cross-section of the structure. Therefore, this portion of the movement is excluded in the calculation of the structure racking deformation. Effect of Relative Stiffness. As expected, results of the analyses indicate that the relative stiffness between the soil medium and the structure has the most significant influence on the structure response. This is demonstrated in Figure 33, where the structure racking coefficients, R, are plotted against the flexibility ratios, F. When the flexibility ratio approaches zero, representing a perfectly rigid structure, the structure does not rack regardless of the distortion of the ground in the free-field. The normalized structure distortion (i.e., R) increases with the increasing flexibility ratio. At F=1, the structure is considered to have the same stiffness as the ground and therefore is subjected to a racking distortion that is comparable in magnitude to the ground distortion in the free field (i.e., R1). With a flexibility ratio greater than 1.0, the structure becomes flexible relative to the ground and the racking distortion will be magnified in comparison to the shear distortion experienced by the ground in the free field. This latter phenomenon is not caused by the effect of dynamic amplification. Rather, it is primarily attributable to the fact that the ground surrounding the structure has a cavity in it (i.e., a perforated ground). A perforated ground, compared to the non-perforated ground in the free field, has a lower stiffness in resisting shear distortion and thus will distort more than will the non-perforated ground. An interesting presentation of these data for rectangular structures is shown in Figures 34 and 35, where the closed-form solutions obtained for the normalized circular lining deflections (Figure 15 in Chapter 4) are superimposed. Note that the definitions of flexibility ratio, F, are different. For circular tunnels, Equation 4-6 is used. For rectangular tunnels, Equation 5-5, 5-6 or 5-7, as appropriate, is used.
114
Because the Poissons Ratios of the soil used in all the rectangular cases are between 0.4 and 0.48, for comparison, the data for circular tunnels are shown only for Poissons Ratios of 0.4 and 0.5. The figures show excellent consistency in distortion response between the two distinctly different types of tunnel configurations. Generally speaking, for a given flexibility ratio the normalized distortion of a rectangular tunnel tends to be less than that of a circular tunnel by approximately 10 percent. The results presented above lead to the following conclusions: The conventional seismic design practice for rectangular tunnels (see Section 5.4) is too conservative for cases involving stiff structures in soft soils (specifically, when F<1.0). Designing a rectangular tunnel according to the free-field deformation method leads to an underestimation of the tunnel response when the flexibility ratio, F, becomes greater than 1.0. From a structural standpoint, fortunately, this may not be of major concern in most cases because F>1.0 may imply the medium (soil/rock) is very stiff, and therefore the free-field deformation can be expected to be small. F>1.0 may also imply the structure is very flexible so that the structure can, in general, absorb greater distortions without being distressed. From a practical standpoint, the data presented in Figures 34 and 35 can be used for design purposes. The normalized deflection curves derived for circular tunnels (Figures 15 and 16) may serve as upper-bound estimates for tunnels with rectangular shapes. Note that Figures 15 and 16 are based on Equation 4-13 in Chapter 4. Effect of Structure Geometry. The effect of structure geometry was studied by using five different types of box structure configurations (Figure 30) in the 25 cases of analyses listed in Table 4. The results presented in Figure 33, however, clearly demonstrate that: The normalized racking deformations are relatively insensitive to the structure geometry. The soil/structure interaction is mainly a function of the relative stiffness between the soil and the structure, regardless of the variations of structure types. Effect of Ground Motion Characteristics. The effect of ground motion characteristics on the normalized racking deformations is negligible. Consider the comparisons of the following pairs of analyses listed in Table 4: Cases 7 and 9 for structure type 2 Cases 20 and 21 for structure type 3 Cases 22 and 23 for structure type 4
118
In each pair of analyses, the parameters characterizing the soil/structure system are identical except for the input ground motions (i.e., the northeastern versus the western earthquakes). The seismically induced racking distortions of the structures are much greater under the assumed western design earthquake than the northeastern design earthquake. However, for the three comparisons made in this study, the normalized racking response with respect to the free-field, R, is very little affected by the type of ground motions used in the analysis. For instance, the calculated racking response coefficients show negligible difference (R=0.445 vs.R=0.448) between cases 22 and 23. Effect of Embedment Depth. To determine the effect of shallow embedment depth on the normalized racking response, finite-element analyses were performed using Type 2 structure as an example. Here, the burial depths of the structure were varied. Table 5 presents the cases that were analyzed for this purpose. Note that flexibility ratio, F, remained the same for all cases. The normalized racking distortions from these analyses versus the dimensionless depth of burial, h/H, are presented in Figure 36. Based on the results, it appears that: The normalized racking distortion, R, is relatively independent of the depth of burial for h/H>1.5 (i.e., soil cover thickness equal to structure height). At this burial depth the structure can be considered to respond as a deeply buried structure. For cases where the depth of embedment is less than 1.5, the normalized racking distortion decreases as the depth of burial decreases, implying that design based on data presented in Figures 34 and 35 is on the safe side for tunnels with little to no soil cover. Effect of Stiffer Foundation. The results of analyses discussed thus far are primarily for cases involving structures entirely surrounded by relatively homogeneous soil medium, including soil profiles with linearly increasing stiffness. A frequently encountered situation for cut-and-cover tunnels is when structures are built directly on the top of geological strata (e.g., rock) that are much stiffer than the overlying soft soils. To investigate the effect of stiffer foundation, seven analyses were performed with varying foundation material properties as well as varying overlying soil properties. Table 6 lists the various parameters used in each of these analyses. The flexibility ratios shown in Table 6 are based on the overlying soil modulus only. The stiffness of the more competent foundation material is not taken into account. The calculated racking distortions, as normalized by the free-field shear deformations, are presented as a function of the flexibility ratio in Figure 37.
119
Racking Coefficient, R
In comparison with the results shown in Figure 35 it may be concluded that in general, the presence of a stiffer foundation would result in some, but not significant, increase in the normalized racking distortion of the structure. It should be noted, however, that: Although the magnitude of this increase is not significant when expressed in a normalized form, the actual impact to the structure may be significantly greater due to the increased free-field deformations. Normally, amplification of shear strains is expected near the zone of interface between two geological media with sharp contrast in stiffness. Care should be taken, therefore, in estimating the free-field shear deformations in a soft soil layer immediately overlying a stiff foundation (e.g., rock).
Step-by-Step Design Procedure The simplified frame analysis models shown in Figure 38 are proposed. A step-bystep description of this procedure is given below: (a) Characterize the subsurface conditions at the site and determine the soil/rock properties based on results from field and laboratory investigations.
122
(b) Derive earthquake design parameters. As a minimum, these parameters should include peak ground accelerations, velocities, displacements, design response spectra, and possibly the time-history accelerograms for both Maximum Design Earthquake (MDE) and Operating Design Earthquake (ODE). This work should be carried out by earthquake engineers with assistance from geotechnical engineers and seismologists. (c) Conduct a preliminary design of the structure. Size and proportion members of the structure based on the loading criteria under static loading conditions. Normally, applicable design codes for buildings and bridges should be used, recognizing that the structure is surrounded by geological materials rather than a freestanding configuration. (d) Based on the soil/rock properties from step (a) and the design earthquake parameters from step (b), estimate the free-field shear strains/deformations of the ground at the depth that is of interest. Generally: For a deep tunnel in a relatively homogeneous medium the simplified Newmark method, as presented by Equations 4-1 and 4-2, may be used. For shallow tunnels, for tunnels in stratified soil sites, or for tunnels sitting on stiff foundation medium, a simple one-dimensional site response analysis (e.g., SHAKE) is desirable. The end results of this step provide the free-field deformation data, D free-field, as depicted in Figure 38. (e) Determine the relative stiffness (i.e., the flexibility ratio, F ) between the free-field medium and the structure using the properties established for the structure and the medium in steps (a) and (c) respectively. Equation 5-5, 5-6 or 5-7, as appropriate, may be used to calculate the flexibility ratio for a rectangular structure. (f) Determine the racking coefficient, R, based on the flexibility ratio obtained from step (e), using the data presented in Figures 34 and 35, or Figure 37 as applicable. (g) Calculate the actual racking deformation of the structure, D s, using the values of D free-field and R from steps (d) and (f) as follows:
D s = R D free- field
(Eq. 5-9)
(h) Impose the seismically induced racking deformation, D s, upon the structure in simple frame analyses as depicted in Figures 38A and 38B.
125
Pseudo-Concentrated Force Model for Deep Tunnels (Figure 38A). For deeply buried rectangular structures, the primary cause of the racking of the structure generally is attributable to the shear force developed at the exterior surface of the roof. Thus, a simplified pseudo-concentrated force model provides a reasonable means to simulate the racking effects on a deep rectangular tunnel. Using a conventional frame analysis program, this may be achieved by applying a horizontal support movement or an equivalent concentrated force at the roof level. Pseudo-Triangular Pressure Distribution Model for Shallow Tunnels (Figure 38B). For shallow rectangular tunnels, the shear force developed at the soil/roof interface will decrease as the soil cover (i.e., soil overburden) decreases. The predominant external force that causes the structure to rack may gradually shift from the shear force at the soil/roof interface to the normal earth pressures developed along the side walls. Therefore, for shallow tunnels, the racking deformation, D s, should be imposed by applying some form of pressure distribution along the walls instead of a concentrated force. The triangular pressure distribution is recommended for this purpose. Generally, for a given racking deformation, D s, the triangular pressure distribution model (Figure 38B) provides a more critical evaluation of the moment capacity of rectangular structure at its bottom joints (e.g., at the invert-wall connections) than the concentrated force model (Figure 38A). On the other hand, the concentrated force model gives a more critical moment response at the roof-wall joints than the triangular pressure distribution model. For design, it is prudent to employ both models in the frame analyses. The more critical results should govern to account for the complex distributions of shear stresses as well as normal earth pressures along the exterior surfaces of the structures. (i) Add the racking-induced internal member forces, obtained from step (h), to the forces due to other loading components by using the loading combination criteria specified for the project. The loading criteria presented in Chapter 2 (Equations 2-1 through 2-4) are recommended for this purpose. (j) If the results from step (i) show that the structure has adequate strength capacity according to the loading combination criteria (for both MDE and ODE), the design is considered satisfactory and no further provisions under the seismic conditions are required. Otherwise, proceed to step (k) below. (k) If the flexural strength of the structure is found to be exceeded from the step (i) analysis, the structural members rotational ductility should be checked. Special design provisions using practical detailing procedures should be implemented if inelastic
126
deformations result. Section 2.4 in Chapter 2 includes a detailed discussion on the strength and ductility requirements for both MDE and ODE loading combinations. (l) The structure, including its members and the overall configurations, should be redesigned if: The strength and ductility requirements based on step (k) evaluation could not be met, and/or The resulting inelastic deformations from step (k) evaluation exceed the allowables (which depend on the performance goals of the structure) In this case, repeat the procedure from step (e) to step (l), using the properties of the redesigned structure section until all criteria are met.
Verification of the Simplified Frame Model The simplified frame models according to Equation 5-8 and Figures 38A and 38B were performed for Cases 1 through 5 (see Table 4) to verify the models validity. The bending moments induced at the exterior joints of the one-barrel rectangular framed structure (simplified analyses) were compared to those calculated by the dynamic finiteelement soil/structure interaction analyses (rigorous analyses). The comparisons are presented, using the concentrated force model, in Figures 39 and 40 for bending moments at the roof-wall connections and the invert-wall connections, respectively. Similar comparisons made for the triangular pressure distribution model are shown in Figures 41 and 42. As Figures 39 and 40 show, the simplified frame analyses using the concentrated force model provide a reasonable approximation of the structure response under the complex effect of the soil/structure interaction. One of the cases, however, indicates an underestimation of the moment response at the bottom joints (i.e., invert-wall connections) by about fifteen percent (Figure 40). When the triangular-pressure distribution model is used, the simplified frame analyses yield satisfactory results in terms of bending moments at the bottom joints (Figure 42). The triangular-pressure distribution model, however, is not recommended for evaluation at the roof-wall connections, as it tends to underestimate the bending moment response at these upper joints (Figure 41). Through the comparisons made above, and considering the uncertainty and the many variables involved in the seismological and geological aspects, the proposed simplified
128
Figure 39. Moments at Roof-Wall Connections Concentrated Force Model (for Cases 1 through 5)
129
Figure 40. Moments at Invert-Wall Connections Concentrated Force Model (for Cases 1 through 5)
130
Figure 41. Moments at Roof-Wall Connections Triangular Pressure Distribution Model (for Cases 1 through 5)
131
Figure 42. Moments at Invert-Wall Connections Triangular Pressure Distribution Model (for Cases 1 through 5)
132
frame analysis models shown in Figures 38A and 38B are considered to comprise an adequate and reasonable design approach to the complex problem.
133
6.0 SUMMARY
135
136
6.0 SUMMARY
A rational and consistent methodology for seismic design of lined transportation tunnels was developed in this study which was mainly focused on the interaction between the ground and the buried structures during earthquakes. Although transportation tunnels were emphasized, the methods and results presented here would also be largely applicable to other underground facilities with similar characteristics, such as water tunnels, large diameter pipelines, culverts, and tunnels and shafts for nuclear waste repositories (Richardson, St. John and Schmidt, 1989).
Vulnerability of Tunnel Structures Tunnel structures have fared more favorably than surface structures in past earthquakes. Some severe damages including collapse have been reported for tunnel structures, however, during earthquakes. Most of the heavier damages occurred when: The peak ground acceleration was greater than 0.5 g The earthquake magnitude was greater than 7.0 The epicentral distance was within 25 km. The tunnel was embedded in weak soil The tunnel lining was lacking in moment resisting capacity The tunnel was embedded in or across an unstable ground including a ruptured fault plane
Seismic Design Philosophy State-of-the-art design criteria are recommended for transportation tunnel design for the following two levels of seismic events: The small probability event, Maximum Design Earthquake (MDE), is aimed at public life safety.
137
The more frequently occurring event, Operating Design Earthquake (ODE), is intended for continued operation of the facility, and thus economy. Loading combination criteria consistent with current seismic design practice were established in this study for both the MDE and the ODE. The proper seismic design of a tunnel structure should consider the structural requirements in terms of ductility, strength, and flexibility.
Running Line Tunnel Design Seismic effects of ground shaking on a linear running tunnel can be represented by two types of deformations/strains: axial and curvature. The following procedures currently used in quantifying the axial and curvature deformations/strains were reviewed: The simplified free-field method (Table 1 equations), which allows simple and quick evaluations of structure response but suffers the following drawbacks: - By ignoring the stiffness of the structure, this method is not suitable for cases involving stiff structures embedded in soft soils. - The ground strains calculated by simplified free-field equations (see Table 1) are generally conservative and may be overly so for horizontally propagating waves travelling in soft soils. The tunnel-ground interaction procedure (beam on elastic foundation), which provides a more realistic evaluation of the tunnel response when used in conjunction with a properly developed ground displacement spectrum. Through several design examples presented in Chapter 3, it was demonstrated that under normal conditions the axial and curvature strains of the ground were not critical to the design of horizontally or nearly horizontally aligned linear tunnels. Special attention should be given, however, to cases where high stress concentrations may develop as follows (Section 3.6): When tunnels traverse two distinctly divided geological media with sharp contrast in stiffness When abrupt changes in tunnel cross sectional stiffness are present, such as at the connections to other structures or at the junctions with other tunnels When the ground ruptures across the tunnel alignments (e.g., fault displacements)
138
When tunnels are embedded in unstable ground (e.g., landslides and liquefiable sites) When tunnels are locally restrained from movements by any means (i.e., hard spots)
Ovaling Effect on Circular Tunnels Ovaling of a circular tunnel lining is caused primarily by seismic waves propagating in planes perpendicular to the tunnel axis. Usually, the vertically propagating shear waves produce the most critical ovaling distortion of the lining. The conventional simplified free-field shear deformation method was first reviewed, through the use of several design examples in this study, for its applicability and limitations. Then a more precise, equally simple method of analysis was developed to assist the design. This method takes into account the soil-lining interaction effects and provides closed form solutions (Equations 4-9 through 4-13) to the problems. Numerical finite difference analyses using the computer program FLAC were performed to validate the proposed method of analysis. A series of design charts (Figures 10 through 16) was developed to facilitate the engineering design work.
Racking Effect on Rectangular Tunnels The racking effect on a cut-and-cover rectangular tunnel is similar to the ovaling effect on a mined circular tunnel. The rectangular box structure will experience transverse sideways deformations when subjected to an incoming shear wave travelling perpendicularly to the tunnel axis. The most vulnerable part of the rectangular frame structure, therefore, is at its joints. Conventional approaches to seismic design of cut-and-cover boxes consist of: The dynamic earth pressure method (Section 5.3), originally developed for aboveground retaining structures. Its applications in the seismic design of underground structures are limited only to those built with very small backfill cover, and those with structural characteristics that resemble the characteristics of aboveground retaining structures (e.g., a depressed U-section). The free-field shear deformation method (Section 5.4), which assumes that the racking deformation of a tunnel conforms to the shear deformation of the soil in the free-field.
139
Use of this method will lead to a conservative design when a stiff structure is embedded in a soft soil deposit. On the other hand, when the tunnel structure is flexible relative to the surrounding ground, this method may also underestimate the seismic racking response of the structure. A proper design procedure that can avoid the drawbacks discussed above must consider the soil-structure interaction effect. For this purpose, an in-depth study using dynamic finite element soil-structure interaction analysis was conducted (Section 5.5). In this study, many factors that might potentially affect the tunnel response to seismic effects were examined. The results, however, indicate that the relative stiffness between the soil and the structure is the sole dominating factor that governs the soil-structure interaction effect. Flexibility ratios, F , were defined to represent the relative stiffness between soils and rectangular structures. Using these flexibility ratios, a well defined relationship was established between the actual tunnel racking response and the free-field shear deformation of the ground (Figures 34 and 35). This relationship allows engineers to perform their design work by using conventional and simple frame analysis programs without resorting to complex and time consuming finite element soil-structure interaction analyses. A detailed step-by-step design procedure using these simplified frame analysis models was given in Section 5.6 of Chapter 5.
140
REFERENCES
141
142
REFERENCES
Agrawal, P. K., et al, Seismic Response of Buried Pipes and Structural Components, ASCE Committee on Seismic Analysis on Nuclear Structures and Materials, 1983. American Concrete Institute (ACI), Building Code Requirements for Reinforced Concrete, ACI 318-89. American Association of State Highway and Transportation Officials (AASHTO), Guide Specifications for Seismic Design of Highway Bridges, 1983 and 1991 Interim Report. Applied Technology Council (ATC), Tentative Provisions for the Development of Seismic Regulations for Buildings, ATC 3-06, 1978. Bolt, B. A., Earthquakes: A Primer, W. H. Freeman and Company, 1978. Buckle, I. G., Mayes, R. L., and Button, M. R., Seismic Design and Retrofit Manual for Highway Bridges, Prepared for Federal Highway Administration, FHWA-IP-87-6, 1987. Burns, J. Q., and Richard, R. M., Attenuation of Stresses for Buried Cylinders, Proceedings of the Symposium on Soil-Structure Interaction, Tempe, Univ. of Arizona, 1964. Central Artery/ Third Harbor Tunnel (CA/THT), Seismic Design Criteria for Underground Structures, 1990. Converse Consultants, Seismological Investigation and Design Criteria, prepared for Southern California Rapid Transit District, 1983. Dobry, R., Oweis, I., and Urzua, A., Simplified Procedures for Estimating the Fundamental Period of a Soil Profile, Bulletin of the Seismological Society of America, Vol. 66, No. 4, 1976. Douglas, W. S., and Warshaw, R., Design of Seismic Joint for San Francisco Bay Tunnel, Journal of the Structural Division, ASCE, Vol. 97, No. ST4, April 1971. Dowding, C. H., and Rozen, A., Damage to Rock Tunnels from Earthquake Shaking, Journal of the Geotechnical Engineering Division, ASCE, Vol. 104, No. GT2, February 1978.
143
Duddeck, H. and Erdman, J., Structural Design Models for Tunnels, Tunneling 82, Institute of Mining and Metallurgy, London, 1982. East Bay Municipal Utility District, Seismic Design Requirements, Oakland, California, 1973. Einstein, H. H., and Schwartz, C. W., Simplified Analysis for Tunnel Supports, Journal of the Geotechnical Division, ASCE, Vol. 105, No. GT4, April 1979. Fast Lagrangian Analysis of Continua (FLAC), Itasca Consulting Group, Inc., 1989. Grant, W. P., and Brown Jr., F. R., Dynamic Behavior of Soils from Field and Laboratory Tests, Proceedings, International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics, Rolla, Univ. of Missouri, 1981. Hadjian, A. H., and Hadley, D. M., Studies of Apprarent Seismic Wave Velocity, Proceedings of International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis, 1981. Hetenyi, M., Beams on Elastic Foundation, The University of Michigan Press, Ann Arbor, Michigan, 1976. Hoeg, K., Stresses Against Underground Structural Cylinders, Journal of the Soil Mechanics and Foundation Division, ASCE, Vol. 94, SM4, April 1968. Hwang, R. N., and Lysmer, J., Response of Buried Structures to Traveling Waves, Journal of the Geotechnical Engineering Division, ASCE, Vol. 107, No. GT2, February, 1981. Japanese Society of Civil Engineers (JSCE), Specifications for Earthquake Resistant Design of Submerged Tunnels, 1975. Kuribayshi, E., Iwasaki, T., and Kawashima, K., Dynamic Behaviour of A Subsurface Tubular Structure, Proceedings of 5th Symposium on Earthquake Engineering, India, 1974. Kuesel, T. R., Earthquake Design Criteria for Subways, Journal of the Structural Divisions, ASCE, Vol. 95, No. ST6, June 1969. LINOS A Non-Linear Finite Element Program for Geomechanics and Geotechnical Engineering, Users Manual, by J. P. Bardet, Univ. of Southern California, Los Angeles, California, 1991.
144
Lyons, A. C., The Design and Development of Segmental Tunnel Linings in the U. K., Proceedings of the International Tunnel Symposium on Tunneling Under Difficult Conditions, Tokyo, 1978. Lysmer, J. et al, FLUSH A Computer Program for Approximate 3-D Analysis of SoilStructure Interaction Problems, EERC Report No. 75-30, Berkeley, Univ. of California, 1975. Mohraz, B., Hendron, A. J., Ranken, R. E., and Salem, M. H., Liner-Medium Interaction in Tunnels, Journal of the Construction Division, ASCE, Vol. 101, No. CO1, January 1975. Monsees, J. E., and Hansmire, W. H., Civil Works Tunnels for Vehicles, Water, and Wastewater, Chapter 24.1, SME Mining Engineering Handbook, 2nd Edition, 1992. Monsees, J. E., and Merritt, J. L., Earthquake Considerations in Design of the Los Angeles Metro, ASCE Conference on Lifeline Earthquake Engineering, 1991. Monsees, J. E., Underground Seismic Design, ASCE Structural Group Lecture Series, Boston Society of Civil Engineers Section, 1991. Mow, C. C., and Pao, Y. H., The Diffraction of Elastic Waves and Dynamic Stress Concentrations, R-482-PR, Report for the U.S. Air Force Project, The Rand Corp., 1971 Muir Wood, A. M., The Circular Tunnel in Elastic Ground, Geotechnique 25, No. 1, 1975. Nakamura, M., Katayama, T., and Kubo, K., Quantitative Analysis of Observed Seismic Strains in Underground Structures, Proceedings of Review Meeting of U.S.-Japan Cooperative Research on Seismic Risk Analysis and Its Application to Reliability-Based Design of Lifeline System, 1981. Newark, N. M., Problems in Wave Propagation in Soil and Rock, International Symposium on Wave Propagation and Dynamic Properties of Earth Materials, 1968. Nyman, D. J., et al, Guidelines for the Seismic Design of Oil and Gas Pipeline Systems, ASCE Technical Council on Lifeline Earthquake Engineering, 1984. ORourke, T. D., Guidelines for Tunnel Lining Design, ASCE Technical Committee on Tunnel Lining Design of the Underground Technology Research Council, 1984. Owen, G. N., and Scholl, R. E., Earthquake Engineering of Large Underground Structures, prepared for the Federal Highway Administration, FHWA/RD-80/195, 1981. Paul, S.L., et al, Design Recommendations for Concrete Linings for Transportation
145
Tunnels, Report No. UMTA-MA-06-0100-83-1 and 2, prepared for U.S. DOT, Urban Mass Transportation Administration, Vols. 1 and 2, 1983. Peck, R. B., Hendron, A. J., and Mohraz, B., State of the Art of Soft Ground Tunnelling, Proceedings of the Rapid Excavation and Tunnelling Conference, Chicago, IL., Vol. 1, 1972. Preliminary Seismic Design Guidelines for Pier 300 Container Wharf and Retaining Structures, Port of Los Angeles, 1991. Richardson, A., and Blejwas T. E., Seismic Design of Circular-Section Concrete-Lined Underground Openings Preclosure Performance Considerations for the Yucca Mountain Site, ASCE Symposium on Dynamic Analysis and Design Considerations for High-Level Nuclear Waste Repositories, 1992. Richardson, A., St. John, C.M., and Schmidt, B., A Proposed Concrete Shaft Liner Design Method for an Underground Nuclear Waste Repository, Proceedings of the Conference on Structural Mechanics in Reactor Technology, Pasadena, 1989. Rowe, R., Tunnelling in Seismic Zones, Tunnels & Tunnelling, December 1992. Sakurai, A., and Takahashi, T., Dynamic Stresses of Underground Pipeline during Earthquakes, Proceedings of the 4th World Conference on Earthquake Engineering, 1969. Schnabel, P. B., Lysmer J., and Seed, B. H., SHAKE A Computer Program for Earthquake Response Analysis of Horizontally Layered Sites, EERC Report No. 72-12, Berkeley, Univ. of California, 1972. Schmidt, B., and Richardson, A. M., Seismic Design of Shaft Linings, ASME Symposium on Recent Development on Lifeline Earthquake Engineering, Honolulu, July 1989. Schwartz, C. W., and Einstein H. H., Improved Design of Tunnel Supports: Volume 1 Simplified Analysis for Ground-Structure Interaction in Tunneling, Report No. UMTA-MA06-0100-80-4, Prepared for U.S. DOT, Urban Mass Transportation Administration, 1980. Seed, B. H., and Idriss, I. M., Soil Moduli and Damping Factors for Dynamic Response Analysis, EERC Report No. 70-10, 1970. Seed, B. H., and Whitman, R. V., Design of Earth Retaining Structures for Dynamic Loads, ASCE Specialty Conference on Lateral Stresses in the Ground and Design of Earth Retaining Structures, 1970.
146
SFBARTD, Technical Supplement to the Engineering Report for Trans-Bay Tube, July 1960. SFBARTD, Trans-Bay Tube Seismic Joints Post-Earthquake Evaluation, November 1991. Sharma, S., and Judd, W. R., Underground Opening Damage from Earthquakes, Engineering Geology, 30, 1991. Southern California Rapid Transit District (SCRTD), Supplemental Criteria for Seismic Design of Underground Structures, Metro Rail Transit Consultants, 1984. St. John, C. M., and Zahrah, T. F., Aseismic Design of Underground Structures, Tunnelling and Underground Space Technology, Vol. 2, No. 2, 1987. Taipei Metropolitan Area Rapid Transit Systems (TARTS), Railway Tunnel Investigation Sungshan Extension Project, ATC, Inc., 1989. Torseth, D. O., Port of Seattle Seismic Waterfront Design, ASCE Symposium on Lifeline Earthquake Engineering: Performance, Design and Construction, 1984. Uniform Building Code (UBC), International Conference of Building Officials, Whittier, California, 1992. Wang, J. M., The Distribution of Earthquake Damage to Underground Facilities during the 1976 Tangshan Earthquake, Earthquake Spectra, Vol. 1, No. 4, 1985. Warshaw, R., BART Tube Ventilation Building, Civil Engineering, ASCE, December 1968. Werner, S. D., and Taylor, C. E., Seismic Risk Considerations for Transportation Systems, ASCE Technical Council on Lifeline Earthquake Engineering, Monograph No. 1, November 1990. Wittkop, R. C., Seismic Design for Port of Los Angeles, 2020 Program, ASCE Conference on Lifeline Earthquake Engineering, 1991. Wood, J. H., Earthquake-Induced Soil Pressures on Structures, Report No. EERL 73-05, 1973, California Institute of Technology. Yeh, G. C. K., Seismic Analysis of Slender Buried Beams, Bulletin of the Seismological Society of America, Vol.64, No. 5, 1974. Yong, P. M. F., Dynamic Earth Pressures Against A Rigid Earth Retaining Wall, Central Laboratories Report 5-85/5, Ministry of Works and Development, New Zealand, 1985.
147