Summary
Soil-pipe interaction is an important factor in the design of steel catenary
risers. Pertinent aspects of this issue include the backbone curve defining
soil resistance under conditions of virgin penetration, equivalent soil spring
stiffness during unload-reload cycles, and permanent displacements under cyclic
loading. This paper presents the findings of single-gravity model tests in soft
kaolin that were conducted to investigate these issues. Cyclic tests were
conducted for small amplitude loading involving no reversal in the direction of
the soil-resisting force, and large amplitude loading involving such reversals.
Presented in this paper are measured relationships of soil secant stiffness as
a function of displacement and load cycle for conditions of unloading and
reloading. The rate of accumulation of permanent displacements under cyclic
loading is strongly affected by the magnitude of loading. Backbone curve
measurements during virgin penetration are in general agreement with analytical
estimates. The laboratory-model test data were used to derive seabed stiffness
parameters for a seabed-riser interaction model. An example analysis is
presented for a typical steel catenary riser touching down on a soft
seabed.
Introduction
Interaction effects between seafloor soils and shallowly embedded pipes are
relevant to a variety of pipeline problems. A topic of considerable current
interest is soil-pipe interaction within the zone where a riser pipe touches
down on the seabed, which proves to be a region where changes in bending
stresses are largest and therefore a critical location for fatigue (Bridge et
al. 2003, 2004). A P-y approach similar to that applied to laterally
loaded piles is often adopted for this problem, where P is soil force
per unit length of pipe (F/L) and y is deflection normal to the
pipe axis. In the study presented herein, the focus is on vertical motions, and
y denotes vertical deflection.
Development of a P-y model requires characterization of the following
aspects of soil-pipe interaction:
- The "backbone" curve describing force-penetration (P-y)
behavior under conditions of virgin penetration.
- Equivalent soil spring stiffness under conditions of unloading and
reloading, and the change in stiffness during cyclic loading.
- Permanent deformations under cyclic loading.
This paper presents the results of a testing program developed for the
purpose of elucidating basic aspects of soil-pipe interaction behavior with
regard to the points listed previously.
Test Program
The test program presented herein comprises three tests conducted under
single gravity conditions. The first is a reference test that establishes the
backbone curve and stiffness relationships to be used as a basis for evaluating
cyclic test data. The second test involves "small-amplitude" cyclic
loading conditions, which in this paper will refer to cyclic loads of
sufficiently small magnitude such that no reversal of direction occurs in the
net soil force acting on the pipe. The third test is a
"large-amplitude" cyclic load test in which the pipe undergoes a large
(1D) upward displacement during each load cycle and a reversal of the
net soil force on the pipe occurs.
Test Apparatus. The tests were conducted in a kaolin test bed
contained in strong box with internal dimensions, 650 mm by 390 mm by 325 mm
deep. The consolidated thickness of the kaolin was 220 mm. The model pipe
dimensions were 25 mm diameter by 125 mm long. An electronically driven
actuator allows both T-bar and pipe penetration tests to be undertaken. The
data acquisition and control software permits loading with either displacement
or force control. A displacement control mode with load reversal at a
prescribed force level is also possible, a feature which was utilized in the
large-amplitude cyclic load test.
Soil Properties. Shear strength of the kaolin test bed was
measured with a 5-mm diameter by 20-mm long T-bar (Stewart and Randolph 1994)
penetrated at a rate of 1 mm/sec. Fig. 1 shows the inferred strength assuming a
T-bar factor of 10.5. A linear regression analysis of the strength profile,
excluding the top 20 mm owing to free surface effects, yields a strength
intercept of 3.7 kPa with a slight negative strength gradient of -0.0029
kPa/mm. The negative strength gradient is a likely consequence of incomplete
consolidation in the test bed. Because the intent of this testing program is to
test the pipe under conditions of known soil strength, but not necessarily
under strength gradient conditions representative of field conditions, the
resultant strength profile is considered satisfactory.
© 2008. Society of Petroleum Engineers
View full textPDF
(
4,892 KB
)
History
- Original manuscript received:
1 February 2008
- Meeting paper published:
5 May 2008
- Revised manuscript received:
2 July 2008
- Manuscript approved:
8 July 2008
- Version of record:
15 December 2008
View Cart
