Page 315 - Numerical Analysis and Modelling in Geomechanics
P. 315

296 ANALYSIS AND DESIGN OF PILE GROUPS
            Benchmark  solutions  in  the  linear  and  non-linear  range  are  presented,  and  the
            significant influence of soil non-linearity on load distribution between individual
            piles in a group is highlighted.


                                    Single pile response
            In comparing non-linear solutions for single pile response to axial loading, the
            problem examined is that reported by Poulos (1989) in his Rankine Lecture. This
            example offers the opportunity to assess the validity of the non-linear hyperbolic
            model  adopted  by  PGROUPN  by  comparison  with  well-established  numerical
            solutions. The input parameters are reported in Table 10.3 and, in order to cover
            a  wide  range  of  pile-soil  relative  stiffnesses  (K=E /E ),  two  values  of  pile
                                                       p
                                                          s
            Young’s modulus have been considered, 30GPa and 30,000GPa (the latter would
            be  unrealistically  stiff  in  practice).  Figures  10.2  and  10.3  report  the  pile  head
            load-settlement response obtained from a FEM analysis by Jardine et al. (1986)
            which  can  be  used  as  a  benchmark.  Such  analysis  involves  the  use  of  a  non-
            linear  soil  model  in  which  the  Young’s  modulus  decreases  markedly  from  an
            initial  value  of  1056  MPa  as  the  axial  strain  level  increases.  Figures  10.2  and
            10.3 also show the load-settlement curves obtained from the following two BEM
            analyses  by  Poulos  (1989):  (a)  an  elastic-perfectly  plastic  continuum-based
            interface  model,  using  a  constant  soil  Young’s  modulus  of  1056  MPa;  (b)  a
            hyperbolic non-linear continuum-
            Table 10.3 Parameters for the analyses reported in Figures 10.2–10.4














            based  interface  model  (similar  to  PGROUPN),  using  an  initial  tangent  soil
            Young’s modulus of 1056 MPa and a hyperbolic curve fitting constant (R ) of 0.9
                                                                      f
            for both the shaft and the base. The PGROUPN solutions have been obtained for
            three sets of hyperbolic curve fitting constants: (1) R =0.5 for the shaft and R =0.
                                                      f
                                                                          f
            9 for the base (this set attempts to reproduce the FEM results); (2) R =0 for both
                                                                   f
            the shaft and the base (to be compared with curve (a) by Poulos); (3) R =0.9 for
                                                                      f
            both the shaft and the base (to be compared with curve (b) by Poulos).
              It  is  worth  noting  that,  for  the  more  compressible  and  realistic  pile
            (Figure 10.2), all BEM analyses (perhaps excluding the analyses including R =0.
                                                                          f
            9 for both the shaft and the base) are capable of predicting a very similar load-
            settlement response to that obtained from the FEM solution which utilises a non-
   310   311   312   313   314   315   316   317   318   319   320