Critical State Soil Mechanics Model Of Soil Consolidation Stresses Around a Driven Pile

Abstract

ABSTRACT This paper discusses results of a study investigating the radial, plane-strain consolidation of soil around a pile using a Critical State soi1 Mechanics [1, 2] model of the soil The study was conducted to assess the assumption that the stress path of the soil during radial consolidation would occur along the critical state line as a basis for developing an effective stress pile capacity formulation [3, 4]. The study showed that predicted stress paths deviated significantly from the critical state line. Furthermore, the results indicated that if initial pore pressures are high enough, consolidation of the soil at the pile wall occurs along a stress path characterized by a radial-Ko condition (Available in full paper). The implications of these results are that consolidalion pressures are model dependent and that excess pore-pressures are likely to be generated during pile loading. It therefore appears appropriate to modify effective stress pile capacity formulations [3] that are based on the soil being at the critical state at initial pile loading. INTRODUCTION Developing improved methods of predicting axial pile response has been an area of active interest by the offshore industry and the geotechnical community for some time. Until recently, the methods developed have been based on a total stress approach to soil behavior although it is widely appreciated that an effective stress approach is more fundamentally sound. The preference for total stress methods exists because effective stress analysis introduces major experimental and analytical difficulties. As a result, use of effective stress methods require many simplifying assumptions and it has not been clear they provide a real advantage. Furthermore, the total stress methods have proven to be sound, workable engineering tools, however many engineers feel that they have been extended to their limits and that further improvements can only be achieved through the more fundamental approach. Recently several efforts have been made to develop axial pile capacity prediction techniques that are based on an effective stress representation of soil strength (e.g. Burland [5], Parry and Swain [6], Esrig, et al. [3]). Burland assumed that pile installation and loading did not change the soil's stress state. Parry attempted to remove an important assumption by considering changes in stress state due to pile loading. Although these methods do constitute important background work and have indicated some promise they do not demonstrate a clear advantage over total stress approaches when compared with available pile test data. In order to improve on these methods, the in-situ stress state and stress history of the soil around the in-place pile must be determined. This exercise requires several important tools; chief among these are accurate constitutive models of soil behavior, models of pile insertion and loading, and a means of solving difficult nonlinear equations.