Measurements of global and local effects of wave impact on a fixed platform deck
Summary (4 min read)
Introduction
- Current regulations used in the design of an offshore platform for a specific site 1, 2 require a minimum air gap of 1.5 m between the expected magnitude of a 100-year wave crest (including tide and storm surge) and the underside of the lowest deck of the platform.
- Extreme wave heights were recorded at North Rankin platform off the Western Australia (WA) coast in 1989 during tropical cyclone Orson 11; examination of the damage sustained by the base of the platform indicated that the platform had experienced impacts from waves with a height in excess of 20 m.
- The air gap for an offshore fixed or floating structure may be smaller in reality than it has been designed to be.
- In order to mitigate the potential effects of wave-in-deck impacts, a thorough understanding of wave induced loads is required.
- Besides, measurement repeatability were analysed and the observed variations in forces and pressures were discussed.
Experimental investigation
- A series of model tests was conducted at the Australian Maritime College towing tank, which is 100 m long, 3.55 m wide and 1.5 m deep.
- To minimise this undesired effect the stiffness and rigidity of the system was improved by attaching the model to a 4500 mm long steel Hbeam (334 mm x 170 mm x 6/11 mm) mounted on the tank rails and placed 15 m away from the wavemaker as shown in Figure 2.
- The remaining 85 m of towing tank allowed for sufficiently long run times without interference from reflected waves travelling back up the tank 26.
- Three deck clearances were nominated based on the platform’s loading conditions, as shown in Figure 3 at model scale with the z-coordinate vertical and positive upward.
- The wave height of incoming/incident waves, travelling in positive x-direction along the tank, was measured by WP1 and WP2.
Measurements of wave-in-deck loads
- Both global and local effects of wave-in-deck impact loading were investigated.
- Two load cells were used to measure the global forces generated due to the impact of the wave crest against the deck structure.
- The layout of the two AMTI MC3A-100 load cells, denoted by LC1 and LC2, is illustrated in Figure 6.
- As can be seen in Figure 6 and Table 2, the pressure transducers (denoted as PT) were placed along the diagonal of the bottom plate.
- A sampling frequency of 20 kHz was chosen for all channels (including wave probes) in order to capture the short-duration slamming pressures 29.
Experimental procedure
- The experiments were conducted using a combination of the following procedures:.
- Wave calibration tests – carried out to identify the extreme waves within long-crested irregular wave trains without the model in-place.
- Free oscillation tests – to find the natural frequencies of the complete test system when subjected to free oscillation tests in air and in water.
- Wave impact tests – to measure the impact wave forces and localised slamming pressures.
- In addition the deck accelerations were monitored to identify the structural dynamic response and its effect on the force magnitudes by estimating the inertial force contribution in the load cell responses.
Wave calibration tests
- The wave calibration tests were conducted by measuring the wave elevation profile, using five wave probes spread longitudinally down the tank, whilst running long-crested irregular waves without the deck model in place.
- Three long-crested irregular wave trains, with a duration of 120 s each, were generated.
- The wave crests of each wave event were identified from the measured wave elevation time histories with and without the deck structure in-place (at different deck clearances, a0).
- The maximum wave steepness of the identified wave events was found to be approximately 0.10 (WE5) i.e. nonbreaking wave conditions.
- In addition, since the magnitude of the peak horizontal force depends on the associated wave velocity, u, in x-direction at the wave crest, the later was estimated using the Stokes second order wave kinematics 29, as given in Table 4.
Free oscillation tests
- Once the instrumented model was attached to the heavy beam mounted on the tank rails, the full testing assembly (deck model, instruments and force supports) was subjected to a series of oscillatory decay tests.
- By obtaining the lowest natural frequency of the system in x- and z-direction, the force contribution into the load cell signals due to the system’s dynamics may be identified.
- The lowest natural frequencies (denoted as fn) obtained in the dry and wet free oscillation tests are summarised in Table 5, based on time traces measured by both load cells and accelerometer.
- The second and the third modal frequencies obtained from the dry test (16.0 Hz and 21.8 Hz) are not present when the bottom plate of the deck is aligned on the water surface, due to the contribution of the relatively small added mass and viscous damping to the system in the longitudinal direction (x-direction).
- The lowest natural frequency observed while the deck was tested in air (16.0 Hz) was found to reduce to 14.5 Hz.
Wave impact tests
- It is important to note that, in order to ensure the repeatability of the results each test condition was repeated 3 – 5 times resulting in a total of 138 runs.
- Each run covered the measurements of force components (Horizontal and vertical) and localised pressures (impact and downward) such that approximately 2620 peaks were analysed and averaged to obtain reliable experimental data.
- As can be seen in Table 6, the dynamic air gap (denoted by a) was obtained for each condition using (a0 - ηc).
- Nevertheless, as discussed earlier the associated wave events (WE7 and WE9) caused wave impact nearby the TE for these conditions.
Results and discussion
- Wave-in-deck forces in the x-direction (Fx) and z-direction (Fz) as well as localised pressures associated with the test conditions shown in Table 6 are presented and discussed in this section.
- WE1 (H = 253 mm) and WE8 (H = 249 mm), which are considered the most extreme amongst the investigated wave events, have been selected for detailed discussion in terms of repeatability analyses.
Wave-in-deck forces
- To assess the uncertainty in the experimental data each test condition was repeated between 3 and 5 times.
- The force peaks in the z-direction were found for both the upward direction, Fz(↑), and the downward direction, Fz(↓).
- E.g. in condition 1 a relative difference of approximately 12% was obtained between Run 1 and Run 2.
- Hence, there is a sufficient confidence in the mean values reported in this paper.
- Nevertheless, a close view was done in regards to the dynamic response of the impacted deck structure and its effect on the wave-in-deck forces.
Force time history
- In order to investigate the force variation, time histories of the force measurements are presented for multiple runs.
- Both time histories show an impulse-like impact i.e. force magnitude sharply increased and then followed by a rapid decrease.
- The vertical force signal, illustrated in Figure 13, has a large downwards component - approximately double the upwards component.
- The figure clearly shows a substantial amount of downwards flowing water from the LE of the deck after the wave impact.
- Causes the vertical wave-in-deck force Fz) is controlled by the magnitude of the horizontal force.
Deck clearance effect
- The effect of deck clearance on the measured forces is discussed in this section.
- Peak forces from different runs were averaged and listed against each wave event.
- Likewise, Table 11 shows the effect of deck clearance on the peak values of the vertical upward force component, Fz(↑), due to the most severe wave events observed in this investigation.
- As can be noted the time history in both runs has a similar trend in the uplift direction i.e. positive cycle.
- The ratio between the maximum positive pressures (impact pressure, Pi) in both runs was found to be 0.92 (almost an 8% difference).
Pressure repeatability
- Repeatability of pressure measurements is introduced in this section by demonstrating the impact pressures associated with WE1 measured in multiple runs.
- Impact pressures measured by sixteen PTs in multiple runs for condition 1 [WE1 at a0 = 120 mm]. ), the maximum peak pressure was also captured by PT#1 with a mean value of approximately 2.48 kPa, whilst for condition 18 ( Figure 20.
- The averaged impact pressure measured by each transducer is normalised by the dynamic pressure (0.5𝜌C2) associated with the wave event.
- The markers represent the number of pressure transducers along the deck (PT#1 – PT#16).
- A consistent finding can be reached based upon these graphs such that the results confirm that the location of the maximum impact pressure(s) moves towards the trailing edge (TE) as the deck clearance is reduced.
Conclusions
- This paper described a series of model tests conducted to examine extreme wave events associated with tropical cyclonic conditions and their impacts on an offshore deck structure.
- Extreme waves of a representative cyclonic sea state were observed in a towing tank within long-crested irregular wave trains.
- This finding should be considered conclusive provided that there are not successive large waves impacting the structure; in which case, the structural and viscous damping appear to strongly influence the force signals, particularly in the direction of wave propagation.
- Most of the test conditions demonstrated that the magnitude of the impact pressure varies considerably among repeated runs, even if identical wave condition was used.
- By investigating the effect of deck clearance on the localised impact pressures, it was found that the reduction in the original deck clearance (10 mm or 20 mm) can increase the impact pressure magnitude at many locations along and across the deck underside.
Did you find this useful? Give us your feedback
Citations
35 citations
32 citations
17 citations
8 citations
6 citations
References
475 citations
226 citations
209 citations
"Measurements of global and local ef..." refers background in this paper
...Although there is currently a large amount of research effort towards the computation of wave-induced loads on ships and offshore structures, only few studies concentrate on wave loads from abnormal waves.(18) Several authors have conducted experimental and/or numerical investigations to estimate the wave-in-deck loading due to regular waves,(19,20) irregular waves(21,22) and a combination of regular and irregular waves....
[...]
99 citations
"Measurements of global and local ef..." refers result in this paper
...This finding is in contrast with the theoretical models which assume that the deck structure is transparent to the impacting waves.(33)...
[...]
36 citations
"Measurements of global and local ef..." refers background in this paper
...In many cases, insufficient air gap has been reported to be one of the major reasons for damage sustained by offshore structures, for example, in the Gulf of Mexico.(3) Damage to the structure or equipment can have costly economic and safety implications, as highlighted recently by McBride....
[...]
...For instance, similar steep waves were measured by three wave radars on the Marco Polo TLP during hurricane Rita in the Gulf of Mexico.32 The generated waves had good repeatability using multiple runs at different values of a0, with only a very minimal disturbance on crest height of WE#1 seen in Figure 7(a) due to the presence of the deck structure....
[...]
...In many cases, insufficient air gap has been reported to be one of the major reasons for damage sustained by offshore structures, for example, in the Gulf of Mexico.3 Damage to the structure or equipment can have costly economic and safety implications, as highlighted recently by McBride.4 Consequently, there is a requirement by classification societies to ensure that an offshore facility can survive in extreme wave conditions caused by tropical cyclones or hurricanes with long return periods....
[...]
Related Papers (5)
Frequently Asked Questions (11)
Q2. How many runs are required to overcome the impact pressure issue?
To overcome this issue, a sufficient number of repeated runs (seems to be more than five per condition) are required during tanks experiments.
Q3. How many air gaps are required in the design of an offshore platform?
Current regulations used in the design of an offshore platform for a specific site 1, 2 require a minimum air gap of 1.5 m between the expected magnitude of a 100-year wave crest (including tide and storm surge) and the underside of the lowest deck of the platform.
Q4. What is the effect of deck clearance on the localised impact pressures?
By investigating the effect of deck clearance on the localised impact pressures, it was found that thereduction in the original deck clearance (10 mm or 20 mm) can increase the impact pressure magnitudeat many locations along and across the deck underside.
Q5. Why are the second and third modal frequencies not present when the bottom plate of the deck is?
The second and the third modal frequencies obtained from the dry test (16.0 Hz and 21.8 Hz) are not present when the bottom plate of the deck is aligned on the water surface, due to the contribution of the relatively small added mass and viscous damping to the system in the longitudinal direction (x-direction).
Q6. Why do deck impacts occur more frequently than predicted?
This may be predominantly due to wave magnitudes being larger than expected, the actual air gap being smaller than anticipated and the resulting wave-in-deck impact forces being greater than predicted.
Q7. How many times did the test condition be repeated?
It is important to note that, in order to ensure the repeatability of the results each test condition was repeated 3 – 5 times resulting in a total of 138 runs.
Q8. What was used to synthesise short-time wave trains using the towing tank wavemaker?
The JONSWAP spectrum with a peak shape parameter γ = 1.0, which in this case is identical to the Pierson-Moskowitz (PM) spectrum 29, was used to synthesise short-time wave trains using the towing tank wavemaker.
Q9. How many pressure transducers were used to represent the length of the deck structure?
The x-position of the sixteen pressure transducers was usedas x-axis to represent the length of the deck structure (Table 2) where the LE and TE are denoted by vertical dashed lines at x = -304 mm and x = 304 mm, respectively.
Q10. How was the phase celerity of the wave crest estimated?
The phase celerity, C, was estimated by λ/Tz for each WE so that the resulting impact pressure could be related to the associated dynamic pressure (0.5𝜌C2).
Q11. How was the role of the dynamic response of the deck structure identified?
The role of the dynamic response of the deck structure was identified by monitoring the acceleration components during wave impact tests.