UCLA

The current best practice in geotechnical engineering in determining lateral capacity of piles is to replace the soil reaction with a series of independent springs. Basically, the model uses beam theory to represent the pile and uncoupled, non-linear load transfer functions, called p-y curves to represent the soil.

Most of the existing methods for determining p-y curves are highly empirical, based on a limited number of cases of laterally loaded piles, which were instrumented, enabling to measure the pile deflection in discrete depth intervals subject to different lateral load (i.e. Matlock 1970, Reese 1975). In essence, these methods have their own limitations, and are mainly applicable for the conditions similar to the tested conditions.

Although later, more detailed investigations by different people addressed some of the problems, still the basis of the existing design programs such as LPILE, or procedures introduced in applicable codes such as API (American Petroleum Institute), is the same original recommendations made by Matlock and Reese during seventies.

In recent era, demand in employment of in-situ direct-pushed based methods using multi-measurement in-situ devices, such as the seismic cone penetration test with pore water measurement (SCPTu) and Seismic Flat Dilatometer Test (SDMT) is significantly increased.

The main objective of this research is to introduce a unified CPT-based approach for determining p-y curves and pile responses to lateral loads. The suggested approach will provide explicit and defined steps/criteria to develop p-y curves for piles subjected to lateral loads using CPT data. CPT data will be used to determine soil strength parameters. Recent developments in relating CPT data to soil basic parameters using Critical State Soil Mechanics (CSSM) framework will be implemented in the suggested model.

In all current common models, pre-determination of the soil behavior and the model to be used (e.g. Matlock clay, 1970 or Reese sand, 1975), will become warranted even before commencement of the analysis. On the contrary, in the proposed model, the need for the said pre-determination of soil behavior is eliminated. As discussed in Section 2.3.5, soil behavior in the model is being classified into four broad and general groups: drained-dilative, drained- contractive, undrained-dilative and undrained-contractive

The main factor driving the suggested analytical approach is Soil Behavior Type Index, Ic. In the proposed approach, the SBT index, Ic, will be used to determine the in-situ characteristics and behavior of the soil. Based on the value of Ic calculated from CPT data, it could be determined that the soil behaves as a sand-like or a clay-like soil, and during the shearing would behave in undrained or drained condition. The measured shear wave velocity during field test using seismic cone penetration test or other methods such as SASW (Spectral Analysis of Surface Waves) or Cross-Hole logging, may be used to determine the small strain shear modulus, G0, which corresponds to the initial stiffness of the linear part of the p-y curve.

In this research, the proposed model will be verified using collected case histories of laterally loaded piles with available CPT data at the same site. The p-y curves, and pile force-head displacements determined from the model will be compared to the field-resulted p-y curves and pile head displacement measurements available from the case histories.