Re-Examination Of P-Y Curve Formulations

Abstract

In the analysis of piles supporting offshore structures, the key element in predicting the response to lateral loads is the determination of the appropriate lateral load - deformation relationships (p-y curves) for the soil. The present practice of constructing p-y curves is based on the results of lateral load tests on instumented piles and strength-deformation characteristics of the soil. (Matlock,1970; Reese et al, 1975; and API RP 2A, 1978) A study was undertaken to compare the results from lateral load tests on piles up to 59 inches in diameter driven in soft to medium clays to that predicted using the above ways of constructing p-y curves. In general, it was found that the predicted pile deflections were significantly greater than the observed deflections. Also the maximum bending moments were underestimated. Two factors were identified as sources of the differences; The first was the assumed linear dependence of the soil-deformation characteristics on pile diameter. The second was the lateral soil resistance at shallow depths was greater than those given by the empirical methods. Based on the study results, modifications are suggested to the present methods of constructing p-y curves. INTRODUCTION P-y curves are generally used to characterize the lateral soil resistance-deformation relationship for piles supporting offshore structures. These curves are related to the strength-deformation characteristics of the soil as obtained from conventional soil tests. The methods of constructing p-y curves for clays are empirically derived from the results of lateral load tests on instrumented piles. The use of empirical methods for design purposes is justifiable when the design conditions approximate those upon which the method is based. One of the important considerations in using empirical methods in geotechnical engineering is the scale effect. Ideally the size of the test element should be the same as the actual element. However, in most cases, it is impractical to perform a full scale test. The methods presently used for constructing p-y curves for clays were based on piles having diameters ranging from 6 to 24 inches. With the trend of installing platforms in deeper waters and more hostile enviroments, piles with diameters ranging from 5 to 8 feet are common. An extreme example is a number of 15-ft diameter piles driven in the Scheldt River delta to support a cable way. (Lubking, 1977) Piles of these diameters are much larger than those upon which the empirical methods are based. Lateral load tests on instrumented piles comparable to those presently used offshore have been published. In addition, other researchers have presented results which suggest ways to account for scale effects. This paper compares the test results to those predicted using the present methods of constructing p-y curves and suggests some modifications.

Numerical analysis of pile behaviour under lateral loads in layered elastic–plastic soils

“…In general, there is a good agreement between the results from FEM and LPILE in uniform sand profile. In uniform soft clay profile, it is noted that the pressures at shallow depth from LPILE are smaller than those computed by FEM, which agrees with one of the findings by the work of Steven and Audibert [15]. For example, the pressures at lateral load of 120 and 200 kN from LPILE are only about half of those from FEM.…”

Re-assessment of P-Y curves for soft clays from centrifuge testing and finite element modeling

“…Reese and Cox, 1971, Stevens and Audibert, 1979, Murff and Hamilton, 1993, Randolph and Houlsby, 1984 to suggest that the API (2000) method for soft normally consolidated clays, will produce springs that underestimate the ultimate unit pressure that can be generated at the face of the pile. Reese and Cox, 1971, Stevens and Audibert, 1979, Murff and Hamilton, 1993, Randolph and Houlsby, 1984 to suggest that the API (2000) method for soft normally consolidated clays, will produce springs that underestimate the ultimate unit pressure that can be generated at the face of the pile.…”

Characterisation of model uncertainties for laterally loaded rigid drilled shafts

“…The ultimate lateral soil stress typically increases with depth to a limiting value that is produced by a deep failure mechanism in which soil flows around the shaft. Common lateral soil stress distributions include: (a) the undrained models proposed by Reese (1958), Hansen (1961), Broms (1964a), Stevens & Audibert (1979), and Randolph & Houlsby (1984); and (b) the drained models proposed by Hansen (1961), Broms (1964b), and Reese et al (1974). Broms (1964b) further simplified his lateral soil stress distribution by replacing the passive stresses developed below the depth of rotation with a concentrated load at the shaft tip; this is called the 'simplified Broms' method herein.…”