Chapter 1

Introduction

1.1 Preface

With fast growing urban populations the need for sustainable and effective infrastructure is increasing. This includes an efficient transportation system, which, in congested cities, is only feasible under ground, as well as water and wastewater facilities, which have to cope with ever increasing quantities. London is currently undergoing several major construction projects, namely the Crossrail project and the London Tideway Improvements, to enhance their transportation and sewage systems, for which tunnels and deep excavations are being constructed in congested and sensitive areas.

The London Tideway Improvements includes the construction of a 25 km long tunnel underneath the river Thames, which will capture sewage overflow. Another 7 km long tunnel is built to conduct the wastewater to the treatment works in East London. Crossrail is a 118 km long high-speed rail, including 21 km twin-bore tunnel underneath central London, connecting Heathrow and Maidenhead in the West to Shenfield and Abbey Wood in the East. Eight central London stations allow rapid transportation to and from the city centre.

Tunnelling projects often require the construction of several deep excavations to lower the tunnel boring machine (TBM) to the tunnel drive or to retrieve it again, as for example shown in Figure 1.1. After tunnel construction, shafts can serve to access the tunnel, e.g. stations, or to conduit sewage. Smaller diameter shafts are usually used for ventilation purposes.

These deep excavations can be of rectangular or circular shape. Circular excavations are often the preferred construction method due to larger structural stiffness and the lack of propping inside the excavation. Nevertheless, in com-

Figure 1.1: Lowering the cutter head portion of a 120 m long TBM called ‘Busy-Lizzie’ into a 80 m deep shaft at Beckton (Courtesy of Thames Water)

parison to rectangular shapes, little attention has been paid to the performance of circular shafts. Well-documented case histories are limited while the database for rectangular excavations is much more comprehensive. Therefore designers often use conservative approaches which result in larger settlements than probably occur in reality. Hence the costs of the construction and the protective measures increase. The study by New and Bowers (1994) is one of the only well-documented case histories in the UK and is hence frequently referred to. The formula to estimate surface settlement around shafts is based on a specific geometry, ground conditions and construction method, and may not easily be transferred to different excavation conditions.

Another uncertainty associated with diaphragm walls is the discontinuous nature of the wall which is built from individual panels. In circular excavations the discontinuity is in circumferential direction, potentially reducing the shafts hoop stiffness, which has a major influence on structural performance and hence ground movements. Designers usually assume fully isotropic walls (i.e. full stiffness in circumferential direction), which may not be adequate for diaphragm wall shafts and may result in unsafe design.

The construction of the Lee Tunnel, an integral part of the £1.7 billion London Tideway Improvements, provided the opportunity to carry out a field study on a deep circular shaft to address these uncertainties (costs as estimated in 2004, Thames Water Utilities Limited, 2013).

1.2 London Tideway Improvements

London’s 150 year old Victorian sewage system was initially designed to deal with foul sewage assuming a population of about four million as well as surface run-off. Following heavy rainfall, the system was designed to overflow into the River Thames when necessary to prevent flooding. The 2011 census revealed that London has a population in excess of eight million; in addition, less frequent but heavier rainfall is now common and many green spaces are paved preventing natural drainage. As a result, discharges into the River Thames are now happening much more frequently – around once a week on average. Every year a total of approximately 39 million tonnes of sewage are discharged into the Thames (Thames Water Utilities Limited, 2014).

In 2000, Thames Water, together with the Environment Agency, the Department of Environment, Food and Rural Affairs and the Greater London Authority, conducted the Thames Tideway Strategic Study (Thames Water Utilities Limited, 2013) in order to assess the environmental impact of intermittent discharges of storm sewage on the River Thames. The comprehensive study concluded that not only discharges from combined sewer overflows have to be prevented, but also that London’s major sewage treatment works need to be expanded to treat considerably more wastewater.

In line with these recommendations, Thames Water has developed a number of major engineering schemes to help stop sewers overflow and improve water quality in the River Thames:

• The Lee Tunnel project (Figure 1.2)
• The Thames Tideway Tunnel project
• Improving London’s five principal sewage works

The Lee Tunnel has an internal diameter of 7.2 m. With an average depth of 75 m below the ground it is London’s deepest tunnel. It runs 7 km from the Abbey Mills Pumping Station complex to the Beckton Sewage Treatment Works in East London. Figure 1.2 illustrates the key components of the Lee Tunnel project, which requires the formation of five permanent large diameter shafts, three sited at Beckton and two at Abbey Mills. These shafts are some of the largest ever constructed in the UK. Figure 1.1 shows the Beckton Overflow shaft which served as a TBM launch shaft.

This thesis presents the details of a case study conducted at Abbey Mills Shaft F which has an external diameter of 30 m (Figure 1.2). Due to its depth of

Figure 1.2: Overview Lee Tunnel project (Courtesy of Thames Water)

71 m the shaft penetrates mixed and challenging ground conditions. Its 1.2 m thick primary lining was constructed from twenty 84 m deep diaphragm wall panels. A comprehensive monitoring scheme, including fibre optic (FO) sensors, inclinometers, extensometers and optical surveys, has been implemented and is detailed in this thesis. The results of this study will allow designers to make more confident predictions of structural behaviour and ground movements for the deep excavations of the forthcoming Thames Tunnel project.

1.3 Aims and objectives

The aim of this thesis is to investigate the performance of circular excavations with respect to structural behaviour and ground movements. A detailed case study and careful numerical back-analysis were conducted to improve design assumptions for future shaft construction projects. In detail, this thesis pursues the following objectives:

Enhance database: The database of circular excavations contains a limited number of studies, which makes it difficult for designers to base estimates on previous experience, as advised by codes and standards. This study contributes to the database with a well-documented case history.

Ground movement: As a result of the limited knowledge concerning ground movements around shafts, conservative approaches such as New and Bowers (1994) are used which probably lead to larger movements being predicted than occur in reality. Therefore surface and subsurface ground movements are measured and compared to existing empirical approaches to determine their adequacy or inadequacy.

Structural performance: Curvatures, bending and hoop strains are measured to gain a better understanding of the shaft wall’s structural performance with a special focus on a potential anisotropic wall stiffness. Anisotropy could be caused by the discontinuous nature of the diaphragm wall.

Validate/ improve design assumptions: The measurements are compared to the initial PLAXIS design. In addition, a further numerical model is built in FLAC to carry out a more flexible parametric study. This helps to understand whether the design assumptions were appropriate and how they can be improved.

Overall, this work aims to contribute a well-documented case study to the database of circular excavations and to improve the design assumptions for shafts under similar conditions. It is envisaged that the results will be directly applicable for the design of the forthcoming Thames Tideway Tunnel project, which includes several deep circular excavations.

1.4 Organisation of thesis

• Chapter 2 provides an overview of common shaft construction methods and current design practice. Due to the very few field and model studies, which are examined in detail in this Chapter, design assumptions are conservative and highly uncertain.
• Chapter 3 introduces construction method of Abbey Mills Shaft F and the implemented monitoring scheme. It focuses on the FO instrumentation and details the background principle, the installation procedure, the data acquisition and analysis.
• Chapter 4 deals with the conventional instrumentation. Extensometers, inclinometers and optical surface settlement and diaphragm wall surveys are introduced. A focus is placed on the error correction methods for inclinometers.
• Chapter 5 presents the monitoring results during shaft excavation in comparison to the initial design (carried out in PLAXIS). Ground and wall movements as well as structural bending and hoop strains are presented. Further data from the diaphragm wall construction, a dewatering trial prior to excavation and the annulus infill are also discussed.
• Chapter 6 introduces the numerical model carried out in FLAC, which is developed to back-analyse the design and conduct a parametric study.
• Chapter 7 presents the result of the numerical simulations from FLAC in comparison to the field data. The influence of constitutive soil models, wall installation, wall stiffness/ anisotropy, wall thickness, Chalk stiffness and lateral earth pressure coefficient is examined.
• Chapter 8 summarises this thesis and its findings. It also gives suggestions for further work in this area.