3D Finite Element Analysis of a Deep Excavation & Ground Response Evaluation

Excavation support along with Sydney Trains (west) boundary
Figure 4: Excavation support along with Sydney Trains (west) boundary

Significant ground movements due to deep surface excavations can seriously impact neighbouring infrastructure and utilities. When it comes to assessing the after-effects of excavations on adjacent infrastructure and planning fundamental mitigation measures, it’s of utmost importance to consider displacements due to deep excavations.

This case study demonstrates the capabilities of RS3’s 3D Finite Element Analysis as a primary design tool in the Observational Method (OM), during an excavation project for property development on Sydney’s North Shore.

Overview of the project

The project development is currently located at 88 Christie St, St Leonards, NSW. Images of the proposed development and aerial perspective of the site are shown in Figure 1.

Aerial view perspective of the excavation
Figure 1: Aerial view perspective of the excavation

This development consists of two residential towers (maximum 47 storeys high) and a commercial tower (15 storeys) over a large retail area with 10 levels of the below-ground basement up to 43 m deep. The 8,000 m2 basement excavation extends to the Sydney Trains boundary and is situated along the major rail and road infrastructure shown in Figure 2.

Location plan showing adjacent transport infrastructure & an aerial view of excavation progress
Figure 2: Location plan showing adjacent transport infrastructure & an aerial view of excavation progress

Adjacent infrastructure details

The site was surrounded by buried utilities, road and rail infrastructure, and buildings in the vicinity. The initial consideration for this development was about 14 m from CBD Rail Link which was planned earlier. This rail transport project consisted of two tunnels connecting south of the city, that added two more tracks on each side of the existing protection corridor, as shown in Figure 3.

Schematic diagram with initial development scheme and earlier-planned CBD rail link
Figure 3: Schematic diagram with initial development scheme and earlier-planned CBD rail link

Excavation Design & Subsurface Conditions

The ground conditions were assessed using 52 boreholes, rock core samples, rock index strengths, borehole characteristics, and rock mass defects including joints and bedding partings.

There were specific restrictions for the development which included:

  • A restriction of using anchor systems in the rail easement
  • The construction and operations couldn’t affect the stability of railway infrastructure
  • A maximum displacement of 30mm on the Pacific Highway
  • Monitoring of ground movements and track structures due to excavation was required

The following supports were installed:

  • A contiguous concrete pile wall along with an anchored shotcrete wall
  • Ground anchors within a 20m wide square buttress of rock on the south corner
  • Ground anchors were placed across 20m in the north corner near the Pacific Highway railway bridge.

Excavation supports can be seen in Figure 4.

Excavation support along with Sydney Trains (west) boundary
Figure 4: Excavation support along with Sydney Trains (west) boundary

Impact Assessment

The impact assessment analyzed the effects of the development including basement excavation and building loads along with the correction of the natural stress field characterized by rock mass quality.

The initial numerical assessment of the rock mass responses and installed support was performed using RS2’s 2D Finite Element analysis and finite difference analysis using FLAC 3D. However, for a more detailed and realistic estimate of how existing infrastructure was affected due to construction, 3D Finite Element numerical modelling and analysis was carried out using RS3.

A shoring system with soldier piles and anchors was designed to control the ground surface deformation due to lateral soil pressure in the upper parts of the proposed excavation site and was calibrated using monitoring results from various deep excavations around Sydney.

The predicted ground movements from the 2D assessment using RS2 can be seen in Figure 5.

Predicted total ground movements of the east-west section from 2D assessment
Figure 5: Predicted total ground movements of the east-west section from 2D assessment

Basement Excavation using RS3

The investigation of the west railway wall suggested there could be a presence of weaker shale/laminate bands at about 40m depth. A temporary high stiffness anchor was installed to control ground movements due to wall deflection. While the bulk excavation was in progress, the automated inclinometer recorded horizontal sliding movement on the two shale bands which can be seen in Figure 6. The sliding movement was due to the release of in-situ stress within the sandstone causing the block of sandstone above the shale band to move more than predicted.

Initially, RS3 was used to calibrate a slice model to match the movement measured by the inclinometer. Further, the shale band was modeled as a ubiquitous joint model with Mohr-Coulomb parameters. After the model was calibrated, several options were tested to reduce the movement and provide additional support to the wall and ensure that the railway track wasn’t impacted by the movements.

Bulk excavation progress from west to east and RS3 slice model including shale bands calibrated with inclinometer measurements – 14 November 2019 to 19 March 2020
Figure 6: Bulk excavation progress from west to east and RS3 slice model including shale bands calibrated with inclinometer measurements – 14 November 2019 to 19 March 2020
RS3 slice model with shale bands including building floor slabs and walls and showing the predicted movement of the railway
Figure 7: RS3 slice model with shale bands including building floor slabs and walls and showing the predicted movement of the railway
RS3 slice model with shale bands calibrated showing the effect on the movement of building floor slabs and walls
Figure 8: RS3 slice model with shale bands calibrated showing the effect on the movement of building floor slabs and walls
Case Study - 3D Finite Element Analysis of a Deep Excavation & Ground Response Evaluation
Figure 9: Case Study – 3D Finite Element Analysis of a Deep Excavation & Ground Response Evaluation
RS3 full staged excavation model leaving a rock buttress below the railway wall to be excavated last.
Figure 10: RS3 full staged excavation model leaving a rock buttress below the railway wall to be excavated last.
RS3 full staged excavation model predicted effects on the railway line leaving a rock buttress below the railway wall to be excavated last and introduction of corner propping and additional anchoring.
Figure 11: RS3 full staged excavation model predicted effects on the railway line leaving a rock buttress below the railway wall to be excavated last and introduction of corner propping and additional anchoring.

Monitoring & Risk Management

During construction, the excavation walls were monitored through instrumentation and regular visual observation. The maximum measured horizontal wall movement was 28 mm at the mid-point of the west wall, as seen in Figure 12, which was representative of the results generated in RS3.

Laser wall scanning on the west wall with a mean displacement of 10mm and maximum of 28mm
Figure 12: Laser wall scanning on the west wall with a mean displacement of 10mm and maximum of 28mm

Conclusion: Successful assessment of basement excavation in RS3

This project highlights the significance of RS3’s finite element analysis in assessing the impact between underground infrastructure and high-rise building foundations/deep basement excavations. This study also concludes the advantage of having FEM as a primary tool in the observation-based approach to make accurate predictions and back-analysis to overcome geotechnical challenges.

We recommend reading the full paper for more details. Discover RS3 and how it can help with your excavation projects.

Be the first to comment

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.