Pile Foundations in Liquefied and Laterally Spreading Ground
Project funded by:
(2005)
| PI: |
Kohji Tokimatsu, Tokyo Institute of Technology |
| Co-PIs: |
Hiroko Suzuki(Tokyo Institute of Technology) , Ross Boulanger(UC Davis) |
| Description: |
This research project addressed key needs for advancing the design of pile foundations in soil profiles that are
susceptible to liquefaction and lateral spreading during earthquakes.
A series of large-scale dynamic centrifuge model tests were performed to study the behavior of single piles and pile
groups in a soil profile comprised of a nonliquefied crust spreading laterally over a loose saturated sand layer. Detailed
instrumentation and new interpretation and data processing procedures enabled fundamental measurements of soil-pile
interaction behavior in the centrifuge tests. The measurements include the first available time histories of loads from
nonliquefied surface soils and underlying liquefied soils during lateral spreading under realistic earthquake shaking motions.
The identified behaviors that are important to design practice include: (1) peak lateral down-slope loads from the surface
crust may be in-phase or out-of-phase with lateral loads from the deeper liquefied layers; (2) the peak lateral loads imposed
on the pile caps by the laterally spreading ground included significant interface friction loads from along the sides and base of
the pile cap; and (3) the lateral load versus relative displacement response of pile caps embedded in laterally spreading
ground was much softer than predicted by relations derived for static loading conditions.
Monotonic pushover analyses based on limit pressures or on nonlinear p-y analyses with monotonic kinematic loading
were evaluated against the centrifuge data. Guidelines for estimating the lateral spreading loads from both nonliquefied and
liquefied layers were subsequently provided. Pushover design analyses using these guidelines and common relations for
other input parameters produced predictions of peak pile bending moments and pile cap displacements that ranged from
reasonable to conservative.
Nonlinear dynamic time-history analyses were also performed using dynamic p-y, t-z, and q-z materials that were
developed and implemented in connection with this project. Example problems and initial comparisons to centrifuge test data
of pile-supported structures in liquefying sand profiles demonstrate that these modeling methods can reasonably
approximate the essential features of soil and structural response.
Downdrag loads on vertical piles in liquefied soil profiles were evaluated using an adapted version of the neutral plane
concept, and a simple design guideline was subsequently recommended.
|
|
|