Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and Arctic Engineering

OMAE2013
June 9-14, 2013, Nantes, France

OMAE2013-10306

DRILLING RISER VIV ANALYSIS USING TIME DOMAIN CODE SIMVIV

Djoni E. Sidarta
Horton Wison Deepwater
Houston, Texas, USA

ABSTRACT introduced in Ref. 3. At the current stage of modeling in-line

VIV in SimVIV, the mean drag loads from currents are not
Drilling risers are often subjected to VIV from ocean included in the code.
currents, which may vary in directions over depth. VIV of
drilling riser has commonly been analyzed using frequency SimVIV computes fluctuating cross-flow and in-line VIV
domain code. This paper presents an alternative tool of forces per element over time. These forces are computed
analyzing VIV of drilling riser using time domain code using VIV coefficients.
SimVIV. With this tool it is possible to apply currents in
varying directions over depth. Cross-flow VIV coefficients are derived based on test data
on single-degree-of-freedom cylinders subjected to steady
Measured currents and VIV responses of a drilling riser currents. In addition, these coefficients for bare pipe have
available in the literature are used in this study. The results of been calibrated to SHEAR7 version 4.2-f [4] for simply
time domain analysis using SimVIV are compared against supported pipe, pinned at both ends and subjected to uniform
measured responses. The effect of current directionality over and sheared flows [3].
depth on drilling riser VIV response is also analyzed.
The in-line VIV forces modeled in SimVIV correspond to
a single excitation region associated with a vortex shedding
INTRODUCTION frequency corresponding to twice the Strouhal number, and
this is assumed to cover the entire range of the reduced
VIV of a drilling riser has commonly been analyzed using velocity of the in-line VIV (from 1 to about 8).
frequency domain code in the industry. Several frequency
domain codes are available such as SHEAR7, VIVA and The in-line VIV coefficients used in the code are functions
VIVANA. The advantage of using frequency-domain code of A/D and reduced velocity. Although there is scope for
over time domain code is speed of computation, allowing a further improvement, these coefficients have been adjusted so
range of sensitivity analyses to be performed quickly. that the predicted response is close to the measured response
However, frequency domain approach is limited by observed in the riser VIV experiments in the Delta Flume at
linearization; and, therefore, cannot accurately model non- Delft Hydraulics [5]. The results of this calibration effort of
linear boundary conditions. In cases where non-linear the in-line VIV coefficients are presented in Ref. 3.
boundary conditions are important, time domain approach has
a clear advantage over frequency domain approach. There is an option in the code to apply or not to apply in-
line VIV forces. This study focuses on the cross-flow VIV
This paper presents the results of VIV analysis of a response of a drilling riser, and in-line VIV forces are not
drilling riser using time domain code SimVIV. Time domain included in the time domain simulations.
VIV code SimVIV is an ABAQUS user subroutine for
computation of cross-flow and in-line VIV forces. The SimVIV assumes lock-in condition when the vortex
algorithm in SimVIV for cross-flow VIV analysis is based on shedding frequency is close to the vibration frequency, and the
the algorithm described in Ref. 1 and Ref. 2, and the fluctuating VIV forces become in-phase with the riser velocity.
implementation of in-line VIV analysis in SimVIV has been Meanwhile, if lock-in condition does not occur, the fluctuating

1 Copyright © 2013 by ASME

VIV forces are set to the vortex shedding frequency based on Table 2 shows the locations of the data loggers from the
Strouhal number with a random phase angle. mean water level, and Figure 4 illustratively shows the
locations of the data loggers on the drilling riser.
The correlation model used in SimVIV is based on
Blevins [6], and this accounts for the variation of the
fluctuating VIV forces with the vibration amplitude and the FE Model of Drilling Riser
closeness of the vibration frequency to the vortex shedding
frequency. The vibration amplitudes and frequencies of VIV A finite element model of this drilling riser was built in
motions are determined based on Prony’s method. ABAQUS v. 6.10-1 [9] using beam elements type B31H.

Figure 5 shows the coordinate system of the FE model.

VIV MEASURED DATA Current direction is “toward” and counter-clockwise from the
positive X axis. In this case zero degree is toward the positive
BP has donated some measured VIV data of a drilling X axis.
riser in the Gulf of Mexico to “Vortex Induced Vibration Data
Repository” [7]. This data is from field measurements of an Structural damping, when applicable, is modeled in
actual full scale drilling riser [8]. The data is from April 13, ABAQUS using Rayleigh damping constants that are applied
2007 to July 11, 2007. This time frame covers three to the riser’s beam elements.
conditions of the drilling riser: (1) connected to well-1, (2)
hanging condition when moving from well-1 to well-2, and (3)
connected to well-2. This study analyzed VIV of this drilling TIME DOMAIN ANALYSIS USING SIMVIV
riser in hanging position for selected current profiles.
Figure 6 shows the directions of cross-flow VIV motions
Measurements of VIV response on a full-scale drilling at one element of the riser. Vector t denotes the element’s
riser and comparison of the measured response to the predicted tangential vector that connects the first to the second node of
response using available analysis tools have been presented in the element. Vector n1 represents the direction of current.
Ref. 10 and 11. Then the cross product of vector t and vector n1 is vector n2,
which represents the direction of cross-flow VIV motions. In-
line VIV is not included in this analysis.
Properties of Drilling Riser
Each SimVIV analysis of a current profile was run for 250
The properties of the drilling riser are shown in Table 1 in seconds. This takes about 5 hours on a computer with CPU
Imperial and SI units, and its schematic is shown in Figure 5. speed of 2.6 GHz. The results of the last 150 seconds of the
simulation were used to compute standard deviations of
accelerations at the locations of the loggers.
Measured Currents and Riser Accelerations

Current measurements are reported every 10 minutes, and Results of Analysis

riser accelerations are recorded for 15 minutes every 2 hours.
Three current profiles were selected for this study. They are Dynamic VIV response of the drilling riser was analyzed
from: (1) June 12, 2007 at 20:10 pm, (2) June 13, 2007 at using SimVIV for three selected current profiles. In each
02:10 am, and (3) June 13, 2007 at 04:10 am. The current dynamic analysis for each current profile, current directions
profiles and the corresponding current directions at these 3 that vary over depth as shown in Figure 1 to Figure 3 were
time stamps were used in the analysis. applied to the drilling riser.

Figure 1 to Figure 3 show the measured current speeds For each current profile, three dynamic simulations were
and directions for the above 3 time stamps and the speeds and performed with variation in the riser’s structural damping.
directions measured 10 minutes before and after. These Three levels of riser’s structural damping (0%, 1%, and 2%)
figures show consistency in the current speeds and directions were investigated.
for 20 minutes. Therefore, the analysis has been done using
one profile of current speed and one profile of current Full-scale testing that has been done for measuring
direction for each selected time window. structural damping of a drilling riser’s buoyant joint indicates
that the tested buoyant joint has structural damping
coefficients from 2.5% to 3% for A/D greater than 0.02 [12].

2 Copyright © 2013 by ASME

 All of SimVIV simulations were based on Strouhal The Effect of Current Directionality
number of 0.20. Table 3 shows the dominant frequencies of
the cross-flow VIV motions of the drilling riser for the three Numerical simulations using time domain VIV code
selected current profiles. This table shows the dominant SimVIV were also performed on these three selected current
frequencies from the measured data and also those computed profiles to investigate the effect of directionality of currents
from SimVIV simulations. SimVIV estimates on the dominant through depths. The profiles of current speeds remain the
frequencies are close to those of the measured data for current same, but the direction of current was made uniform through
profiles 1 and 2, but it is quite different for current profile 3. depths. Figure 1 to Figure 3 show the directionality of the
current profiles through depth.
The maximum current speeds are about 1.20, 1.25, and
1.03 m/sec for the three current profiles. The estimated Figure 12 to Figure 14 show the comparison of standard
shedding frequencies for these maximum current speeds based deviation of the computed lateral accelerations from two
on Strouhal number of 0.20 are 0.187 Hz, 0.195 Hz, and 0.161 SimVIV simulations, with and without directionality, for the
Hz, respectively. These estimates are close to the dominant three current profiles. The results with riser’s structural
frequencies obtained from SimVIV simulations, and they are damping of 2% were used in this comparison.
close to the dominant frequencies of the measured data for
current profiles 1 and 2. It is not clear that the estimated There are two plots in each figure. The computed
dominant frequency for current profile 3 is quite different from accelerations (referred to X-Acc and Y-Acc in the plot) from
that of the measured data. Further study, perhaps with also the simulation with current directionality were not rotated for
including other current data with similar profile and direction, the plot on the left hand side. Meanwhile, they are rotated for
is required to find the explanation. the plot on the right hand side. The rotation of the coordinate
system was based on the direction of current near surface
Figure 7, Figure 8, and Figure 9 show the comparison of which has the maximum speed.
standard deviation of accelerations on the riser at the locations
of the loggers for the X-, Y-, and Z-accelerations, respectively, After the computed accelerations from the simulation with
for current profile 1. The X, Y, and Z refer to the FE global current directionality were rotated, the standard deviation of
coordinate system. Each figure shows comparison of standard the rotated Y-accelerations (minor components) becomes fairly
deviation of accelerations from the three runs with different small. This shows that the computed VIV response of the riser
levels of riser’s structural damping to the standard deviation of using SimVIV is nearly in one plane for the given
the measured accelerations. There was no adjustment on the directionality of the currents. The standard deviation of the
orientation of the measured accelerations, in case the locations rotated X-accelerations (major components) is less than that
of the loggers along the riser are twisted. computed using SimVIV without current directionality for
current profiles 1 and 2. Meanwhile, they are quite close for
The standard deviation of the computed lateral current profile 3. These results of numerical simulation using
accelerations without riser’s structural damping is about three SimVIV, in general, show the effect of current directionality.
times higher than that of the measured accelerations. The results for the first two current profiles show that the
Meanwhile, the standard deviation of the computed vertical riser’s VIV response with the given current directionality is
acceleration is much higher compared to that of the measured less than that without current directionality.
accelerations.

Improvement on the standard deviation of the computed SUMMARY AND CONCLUSIONS

accelerations can be seen by introducing structural damping
(1% and 2%), and significant improvement was seen for the This paper demonstrates SimVIV application for time-
vertical accelerations. domain VIV analysis of a drilling riser. Two main objectives
of this study are to benchmark SimVIV’s results against
Figure 10 and Figure 11 show similar comparison for the measured data for an actual drilling riser, and to investigate the
lateral accelerations for current profiles 1 and 2, respectively. effect of current directions on the VIV response of a drilling
riser based on numerical simulations using SimVIV.
For all three current profiles, the standard deviation of the
computed accelerations even with riser’s structural damping of For three selected current profiles, which have maximum
2% is still higher than that of the measured one. current speed in the range of 1.0 to 1.25 m/sec, SimVIV’s
estimates of standard deviations of riser accelerations at
several locations on the riser are higher than the measured
data. This observation is applicable to SimVIV’s results
without riser’s structural damping and with riser’s structural

3 Copyright © 2013 by ASME

damping of 1% and 2%. This result suggests that there is still 6. Blevins, R.D. (2001). “Flow-Induced Vibration,” Second
conservatism in SimVIV’s prediction on the drilling riser’s Edition, Krieger Publishing Company, Malabar, Florida,
VIV response based on these measured data, and this may U.S.A.
imply two possibilities: (1) VIV forces are smaller than
currently estimated in SimVIV, and / or (2) drilling riser’s 7. Vortex Induced Vibration Data Repository,
structural damping is higher than 2% for this level of http://oe.mit.edu/VIV/index.html, web site hosted by
vibrations. Center for Ocean Engineering, MIT, Cambridge, MA,
U.S.A.
Further study that includes a range of current velocities
that also covers smaller levels of vibrations will be beneficial. 8. BP Report (2008): VIV Benchmarking Repository,
Experiments on a full-scale drilling riser joint suggest that Drilling Riser without VIV Suppression, Data Package
drilling riser’s structural damping varies with vibration Technical Note.
amplitudes [12].
9. ABAQUS Version 6.10-1, SIMULIA.
Numerical analysis investigation using time domain code
SimVIV on the effect of current directions on the VIV 10. Tognarelli, M.A., Taggart, S., Campbell, M. (2008).
response of a drilling riser shows that riser’s VIV response “Actual VIV Fatigue Response of Full Scale Drilling
with current directionality is less than that without current Risers: With and Without Suppression Devices,”
directionality. Proceedings of the 27th International Conference on
Offshore Mechanics and Arctic Engineering, OMAE
2008, No. 57046, Estoril, Portugal.
ACKNOWLEDGMENTS
11. Beynet, P., Shilling, R., Campbell, M., Tellier, E.,
The author wishes to thank BP for donating the VIV data Howells, H. (2008). “Full Scale VIV Response
of drilling riser to the VIV data repository [7] and to thank the Measurements of A Drill Pipe in Gulf of Mexico Loop
organizer of the VIV data repository to make the data available Currents,” Proceedings of the 27th International
to public. Conference on Offshore Mechanics and Arctic
Engineering, OMAE 2008, No. 57610, Estoril, Portugal.

REFERENCES 12. Padelopoulos, P., Ritchie, M., Tognarelli, M., Chitwood,

J. (2012). “Drilling Riser Structural Damping Test,” The
1. Finn, L., Lambrakos, K., Maher, J. (1999). “Time Domain International Society of Offshore and Polar Engineers.
Prediction of Riser VIV,” 4th International Conference on
Advances in Riser Technologies (Aberdeen), May 1999.

2. Highly Compliant Rigid Riser Large Scale Model Test

and Analysis JIP - Phase 3, Time Domain Algorithm
Development, Moffatt & Nichol International, 29 March
2001.

3. Sidarta, D.E., Finn, L.D., Maher, J. (2010). “Time

Domain FEA for Riser VIV Analysis,” Proceedings of the
29th International Conference on Offshore Mechanics and
Arctic Engineering, OMAE 2010, No. 20688, Shanghai,
China.

4. Vandiver, J.K., Lee, L. (2002), “User Guide for SHEAR7

Version 4.2”, Massachusetts Institute of Technology,
Cambridge, Massachusetts, U.S.A.

5. Chaplin, J.R., Bearman, P.W., Huera Huarte, F.J.,

Pattenden, R.J. (2005). “Laboratory measurements of
vortex-induced vibrations of a vertical tension riser in a
stepped current,” Journal of Fluids and Structures, 21, 3-
24.

4 Copyright © 2013 by ASME

 Table 1 Riser Properties
3000-ft 4000-ft 5000-ft
Drilling Riser
Buoyant Buoyant Buoyant Slick Joint Lower FJ LMRP BOP
Components
Joint Joint Joint
Pipe OD (in) 21.000 21.000 21.000 21.000 21.000 26.580 33.550
Pipe WT (in) 0.875 0.875 0.875 0.875 0.875 3.915 7.400
Pipe ID (in) 19.250 19.250 19.250 19.250 19.250 18.750 18.750
Hydro OD (in) 50.500 51.500 52.500 34.000 58.000 222.500 222.500
Number of Joints 23 10 12 26 1 1 1
Length (ft) 1725.000 750.000 900.000 1950.000 8.900 18.900 23.500
Joint Length (ft) 75.000 75.000 75.000 75.000 8.900 18.900 23.500
Joint Dry Weight (kip) 43.428 45.608 48.271 29.570 6.418 220.000 420.000
Joint Wet Weight (kip) 0.888 1.388 0.758 25.726 5.580 191.400 365.400

914-m 1219-m 1524-m

Drilling Riser
Buoyant Buoyant Buoyant Slick Joint Lower FJ LMRP BOP
Components
Joint Joint Joint
Pipe OD (mm) 533.4 533.4 533.4 533.4 533.4 675.1 852.2
Pipe WT (mm) 22.2 22.2 22.2 22.2 22.2 99.4 188.0
Pipe ID (mm) 489.0 489.0 489.0 489.0 489.0 476.3 476.3
Hydro OD (mm) 1282.7 1308.1 1333.5 863.6 1473.2 5651.5 5651.5
Number of Joints 23 10 12 26 1 1 1
Length (m) 525.78 228.60 274.32 594.36 2.71 5.76 7.16
Joint Length (m) 22.86 22.86 22.86 22.86 2.71 5.76 7.16
Joint Dry Weight (m ton) 19.70 20.69 21.90 13.41 2.91 99.79 190.51
Joint Wet Weight (m ton) 0.40 0.63 0.34 11.67 2.53 86.82 165.74

Table 2 Locations of Data Loggers

Elevation Elevation
Location Joint Logger ID Pod
(ft) (m)
Vessel PH-B012 13 80.6 24.6
4000-ft Buoyant Joint 42 PH-B009 90 -2101.2 -640.4
4000-ft Buoyant Joint 41 PH-B007 207 -2176.2 -663.3
Slick Joint 14 PH-E011 12 -4269.3 -1301.3
Slick Joint 13 PH-B004 40 -4344.3 -1324.1
Slick Joint 12 PH-C007 307 -4419.3 -1347.0
Slick Joint 11 PH-B008 80 -4494.3 -1369.9
Slick Joint 10 PH-B010 missing
Slick Joint 9 PH-F001 10 -4644.3 -1415.6
Slick Joint 7 PH-A003 103 -4794.3 -1461.3
Slick Joint 5 PH-B018 18 -4944.3 -1507.0
Slick Joint 3 PH-B003 203 -5094.3 -1552.7
Slick Joint 1 PH-B014 14 -5244.3 -1598.5
LMRP PH-D006 60 -5263.3 -1604.3

Table 3 Dominant Frequency of VIV Response

SimVIV Simulation
Current Measured
for Different Structural Damping
Profile Data
0% 1% 2%
1 0.193 Hz 0.195 Hz 0.171 Hz 0.171 Hz
2 0.192 Hz 0.195 Hz 0.207 Hz 0.183 Hz
3 0.098 Hz vary from 0.146 Hz to 0.171 Hz

5 Copyright © 2013 by ASME

 6/12/2007 20:00 6/12/2007 20:10 6/12/2007 20:20 6/12/2007 20:00 6/12/2007 20:10 6/12/2007 20:20

0 0

-200 -200

-400 -400

Elevation (m)
Elevation (m)

-600 -600

-800 -800

-1,000 -1,000
0 30 60 90 120 150 180 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Current Direction (deg) Current Speed (m/sec)

Figure 1 Current at June 12, 2007 at 20:10 pm

6/13/2007 2:00 6/13/2007 2:10 6/13/2007 2:20 6/13/2007 2:00 6/13/2007 2:10 6/13/2007 2:20

0 0

-200 -200

-400 -400
Elevation (m)

Elevation (m)

-600 -600

-800 -800

-1,000 -1,000
0 30 60 90 120 150 180 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Current Direction (deg) Current Speed (m/sec)

Figure 2 Current at June 13, 2007 at 02:10 am

6 Copyright © 2013 by ASME

 6/13/2007 4:00 6/13/2007 4:10 6/13/2007 4:20 6/13/2007 4:00 6/13/2007 4:10 6/13/2007 4:20

0 0

-200 -200

-400 -400
Elevation (m)

Elevation (m)
-600 -600

-800 -800

-1,000 -1,000
0 30 60 90 120 150 180 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Current Direction (deg) Current Speed (m/sec)

Figure 3 Current at June 13, 2007 at 04:10 am

Riser Logger
100
0
-100
-200
-300
-400
-500
-600
-700
Z (m)

-800
-900
-1,000
-1,100
-1,200
-1,300
-1,400
-1,500
-1,600
-1,700
-10 0 10
X (m)

Figure 4 Locations of Data Loggers on Riser

7 Copyright © 2013 by ASME

 Y

Direction of Current

X
Angle (deg)

Y
X

Figure 5 Coordinate System of the FE Model

n2
~
n1 t xn =n
~ ~ ~1 ~2
t
~
Z

Y
X

Figure 6 Directions of Cross-Flow VIV Motions

8 Copyright © 2013 by ASME

 0% 1% 2% Measured
200

-200

-400

Elevation (m) -600

-800

-1,000

-1,200

-1,400

-1,600

-1,800
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Standard Deviation of X-Acceleration (m/s2)

Figure 7 Standard Deviation of X-Acceleration for Current Profile 1

0% 1% 2% Measured
200

-200

-400

-600
Elevation (m)

-800

-1,000

-1,200

-1,400

-1,600

-1,800
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Standard Deviation of Y-Acceleration (m/s2)

Figure 8 Standard Deviation of Y-Acceleration for Current Profile 1

9 Copyright © 2013 by ASME

 0% 1% 2% Measured
200

-200

-400

Elevation (m) -600

-800

-1,000

-1,200

-1,400

-1,600

-1,800
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Standard Deviation of Z-Acceleration (m/s2)

Figure 9 Standard Deviation of Z-Acceleration for Current Profile 1

0% 1% 2% Measured 0% 1% 2% Measured
200 200

0 0

-200 -200

-400 -400

-600 -600
Elevation (m)

Elevation (m)

-800 -800

-1,000 -1,000

-1,200 -1,200

-1,400 -1,400

-1,600 -1,600

-1,800 -1,800
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Standard Deviation of X-Acceleration (m/s2) Standard Deviation of Y-Acceleration (m/s2)

Figure 10 Standard Deviation of X and Y-Accelerations for Current Profile 2

10 Copyright © 2013 by ASME

 0% 1% 2% Measured 0% 1% 2% Measured
200 200

0 0

-200 -200

-400 -400

-600 -600
Elevation (m)

Elevation (m)
-800 -800

-1,000 -1,000

-1,200 -1,200

-1,400 -1,400

-1,600 -1,600

-1,800 -1,800
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Standard Deviation of X-Acceleration (m/s2) Standard Deviation of Y-Acceleration (m/s2)

Figure 11 Standard Deviation of X and Y-Accelerations for Current Profile 3

X-Acc Y-Acc without directionality X-Acc (rotated) Y-Acc (rotated) without directionality
200 200

0 0

-200 -200

-400 -400

-600 -600
Elevation (m)
Elevation (m)

-800 -800

-1,000 -1,000

-1,200 -1,200

-1,400 -1,400

-1,600 -1,600

-1,800 -1,800
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Standard Deviation of Lateral-Acceleration Standard Deviation of Lateral-Acceleration
(m/s2) (m/s2)

Figure 12 Effect of Directionality on Current Profile 1

11 Copyright © 2013 by ASME

 X-Acc Y-Acc without directionality X-Acc (rotated) Y-Acc (rotated) without directionality
200 200

0 0

-200 -200

-400 -400

-600 -600

Elevation (m)
Elevation (m)

-800 -800

-1,000 -1,000

-1,200 -1,200

-1,400 -1,400

-1,600 -1,600

-1,800 -1,800
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Standard Deviation of Lateral-Acceleration Standard Deviation of Lateral-Acceleration
(m/s2) (m/s2)

Figure 13 Effect of Directionality on Current Profile 2

X-Acc Y-Acc without directionality X-Acc (rotated) Y-Acc (rotated) without directionality
200 200

0 0

-200 -200

-400 -400

-600 -600
Elevation (m)
Elevation (m)

-800 -800

-1,000 -1,000

-1,200 -1,200

-1,400 -1,400

-1,600 -1,600

-1,800 -1,800
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Standard Deviation of Lateral-Acceleration Standard Deviation of Lateral-Acceleration
(m/s2) (m/s2)

Figure 14 Effect of Directionality on Current Profile 3

12 Copyright © 2013 by ASME