Skip to content

Foundations, Deep Excavations and Retaining Systems

Session Summary

Session 4 of GeoVadis (GAIC 2025) encompasses critical advances across foundation systems, deep excavations, and earth retaining structures. Contributions examine piled raft performance under combined vertical-lateral (V-H) loads, soil-cement composite piles, trench stability in diaphragm walls with enlarged panels, bi-directional static load testing (BDSLT), deep soil mixing with steel insertions for 10-meter cut slopes, and thermal behavior of energy micropile arrays.


1. Piled Raft Foundations Under Combined V-H Loading (A. Garg, V.A. Sawant, S. Mehndiratta)

3D Finite Element Interaction Mechanics

Tall structures, bridge towers, and wind turbine jackets impose severe concurrent vertical (\(V\)), horizontal (\(H\)), and moment (\(M\)) loads on piled rafts. Garg et al. conduct coupled 3D elastoplastic FE simulations to analyze load-sharing between raft base and pile group: - Raft-Soil Contact Mobilization: Under pure vertical loads, the raft transmits \(30\% - 45\%\) of total gravity loads directly to surface soil. Under high lateral shear (\(H/V > 0.2\)), raft base shear friction engages rapidly, shedding up to \(40\%\) of horizontal demand before significant lateral pile deflection mobilizes. - Pile Group P-y Softening: Cyclic lateral displacement induces soil plastic yielding in the top \(3D - 5D\) pile depth, diminishing pile group lateral stiffness by \(25\% - 35\%\). - Design Guidelines: Optimizing pile length distribution (longer central settlement-reducing piles, stiffer perimeter shear-resisting piles) minimizes raft differential settlement and perimeter cracking.


2. Soil-Cement Composite Piles (S. Koga, K. Watanabe, T. Naito, N. Tsuchiya)

Concept & Centrifuge Modeling

Soil-cement composite piles comprise a rigid precast concrete core or steel pipe inserted concentric to an outer deep-mixed soil-cement column. Koga and Watanabe utilize geotechnical centrifuge testing (\(50g\)) and full-scale 1g models to elucidate vertical and lateral load-transfer mechanisms:

        Vertical Load P
              │
              ▼
    ┌───────────────────┐
    │ Concrete/Steel    │ ◄─── High Axial Stiffness Core (Ec)
    │ Core Pile         │
    │                   │
    ├─┬───────────────┬─┤
    │ │ Soil-Cement   │ │ ◄─── Enlarged Column Shell (E_sc, D_col)
    │ │ Outer Shell   │ │      Expands Bearing Perimeter & Tip Area
    ├─┼───────────────┼─┤
    │ │               │ │
    │ │ Soil Shear    │ │ ◄─── High Frictional Mobilization across
    │ │ Resistance    │ │      Rough Soil-Cement/Ground Interface
    └─┴───────────────┴─┘

Key Mechanical Insights

  1. Vertical Load Transfer: Axial stress is initially borne by the high-modulus inner core, then transferred progressively into the soil-cement shell via core-grout interface friction. The enlarged base diameter (\(D_{col} = 2 - 3 \times D_{core}\)) expands the end-bearing area by \(400\% - 900\%\).
  2. Lateral Resistance: Under horizontal loads, the outer soil-cement column mobilizes high passive resistance in the upper strata, reducing pile head lateral deflection by over \(50\%\) compared to a bare precast pile.

3. Trench Stability in Diaphragm Walls with Enlarged Panels (A. Iwata, K. Watanabe, T. Watanabe)

Diaphragm Walls with Enlarged T- and Cross-Shaped Panels

To resist massive bending moments in ultra-deep urban excavations, diaphragm walls often incorporate enlarged flange panels (T-panels, cruciform panels). However, the increased trench surface area elevates the risk of slurry trench collapse during excavation: - Slurry Pressure Maintenance: 3D limit equilibrium and finite element analyses indicate that hydrostatic slurry head must exceed the phreatic groundwater table by at least \(1.5\,\text{m}\) to prevent localized inward plastic sloughing at re-entrant corners. - Polymer vs. Bentonite Slurry: Polymer slurries with viscoelastic modifiers maintain superior rheological stability and filter cake continuity across variable sand-gravel lenses, preventing slurry loss and collapse.


4. Deep Soil Mixing with Steel Insertion for 10m Excavation Support (Z.W. He, J. Si, K.W. Leong)

Retaining System for Constrained Urban Sites

He et al. present a successful design and field monitoring case history for a 10-meter deep basement excavation in sensitive urban clay: - System Architecture: Overlapping triple-shaft deep mixed soil-cement columns (\(D = 850\,\text{mm}\)) reinforced with inserted structural steel H-beams (SMW method) installed at alternating column centers. - Tieback Anchoring: Two levels of prestressed ground anchors drilled through the composite wall provide lateral restraint. - Field Performance: Inclinometer measurements recorded a maximum lateral retaining wall deflection of only \(18\,\text{mm}\) (\(< 0.2\% H\)), well within the allowable serviceability limit for adjacent building protection.


5. Bi-Directional Static Load Pile Testing (BDSLT) (A.K. Chakraborti, S.K. Golchha, A. Uppadhyay)

Osterberg Cell (BDSLT) Testing of High-Capacity Bored Piles

Conventional top-down static load tests on deep bored piles (\(Q_{ult} > 25,000\,\text{kN}\)) require colossal reaction kentledge or tension anchor piles, posing severe safety and cost challenges. Chakraborti et al. evaluate deep bored pile tests using hydraulically expandable Osterberg cells (O-cells) embedded at the pile bottom: - Self-Balancing Principle: The O-cell pushes upward against the upper shaft skin friction while simultaneously pushing downward against pile tip end-bearing and lower shaft friction. - Equivalent Top-Load Curve: Using the Schmertmann / O-cell synthesis method, upward and downward load-movement curves are combined into an equivalent conventional top-load movement curve: $\(S_{top}(Q) = S_{up}(Q_{shaft}) + S_{elastic\_shortening}\)$ - Strain Gauge Instrumentation: Multi-level sister bars installed throughout the pile shaft accurately delineate unit shaft friction (\(f_s\)) degradation along stratified clay-shale strata.


6. Excavation and Foundation Instrumentation Summary

Subsurface Structure Primary Diagnostic Sensor Secondary Sensor Key Engineering Parameter
Piled Raft Foundation Earth Pressure Cells (Raft base) Pile Head Vibrating Wire Strain Gauges Raft/pile load sharing ratio, contact pressure
Diaphragm Wall (H-beam / Barrette) Inclinometer Casing (Within wall) Load Cells on Anchor Heads Lateral deflection profile, anchor load loss
Deep Excavation Ground Response Multi-Point Magnetic Extensometers Surface Tiltmeters on Adjacent Structures Subsurface vertical ground settlement profile
BDSLT / O-cell Pile Test Displacement Transducers across Cell Multi-Level Sister Bars in Cage Upward/downward stroke, unit skin friction \(f_s\)
Deep Soil Mixing (SMW Wall) Inclinometers in Steel H-beams Vibrating Wire Piezometers behind Wall Wall deflection, groundwater drawdown