Internship Report
De-risking structural loading calculations for marine renewable energy devices.
Dr Adrian Macleod in partnership with PML Applications Ltd.
De-risking structural loading calculations for Marine Renewable Energy Devices NERC INTERNSHIP PLACEMENT SCHEME REPORT Adrian K. A. Macleod BSc University of Aberdeen - MSc Heriot-Watt University
MAY, 2013
1. Contents 2.
Introduction ................................................................................................................................................. 3
3.
Hydrodynamic loading and MREDs ......................................................................................................... 4 3.1
Introduction to estimating hydrodynamic loading ................................................................................ 5
4.
Novel method for estimating biofouling roughness, thickness and volume............................................ 6
5.
Collecting data from off-shore navigation buoys ..................................................................................... 9 5.1
Site selection and experimental approach............................................................................................. 9
5.2
Statistical analysis............................................................................................................................... 15
5.3
Result................................................................................................................................................... 16
6.
Discussion................................................................................................................................................... 17
7.
The impact of biofouling characteristics on hydrodynamic loading..................................................... 17 7.1
Results ................................................................................................................................................. 21
7.2
Discussion ........................................................................................................................................... 24
8.
Conclusions ................................................................................................................................................ 25
9.
References .................................................................................................................................................. 29
2. Introduction The large scale deployment of marine renewable energy devices (MREDs) capable of converting natural energy into electrical energy is set to increase significantly in coming years (Gill, 2005). The United Kingdom is well placed to extract large quantities of marine renewable energy offering large economic incentives. To exploit this opportunity engineers and developers must overcome some considerable challenges (Mueller and Wallace, 2008); not the least of which is biofouling. The term biofouling commonly refers to the accumulation of unwanted biological material on manmade structures such as boats, off-shore oil platforms and MREDs (Durr and Thomason, 2010). Biofouling is a major engineering concern, influencing the loading of off-shore structures by: increasing the size of structural elements, increasing the drag and inertia coefficients, as well as increasing the structural weight (Theophanatos and Wolfram, 1989, Jusoh and Wolfram, 1996a, Shi et al., 2012). The consideration of marine growth at the design stage relies on information dealing with biofouling growth characteristics (average growth thickness, average peak to valley height, density and drag and inertia coefficients) which are currently largely dependent on those recorded for the North Sea oil and gas industry.
Renewable energy developments have a whole suite of additional considerations. Firstly, abundant growth on these relatively small structures may have a disproportionately large influence on the loading experienced and mechanical performance compared to the superstructures used in the offshore oil and gas industry (Jusoh and Wolfram, 1996b). Secondly, biofouling type and composition may vary from published data currently used by the industry as many of these structures will be deployed in under-developed areas where biofouling characteristics are poorly understood. For example, high water flow may be the one of the most significant agents shaping biofouling communities found in areas suitable for tidal and wave energy extraction (Nowell and Jumars, 1984, Vogel, 1994, Denny et al., 1998, Leonard et al., 1998, Coutts et al., 2010). However, little is known about how these communities might influence the extreme hydrodynamic stresses imposed on wave and tidal devices. Lastly, anti-fouling technologies and methods for removing biofouling are presently poorly placed to deal with the long term deployments and the logistical considerations of working in extreme marine environments like tidal and wave energy sites.
3
3. Hydrodynamic loading and MREDs Biofouling will interfere with many key processes in the life cycle of deployed MREDs, including, energy conversion, maintenance and device longevity (Table 1). For example, biofouling has been shown to reduce the efficiency of turbine blades decreasing overall power generation (Orme et al., 2001). Similarly, biofouling may decrease the power conversion rate for Wave Energy Devices (WEDs) through added inertia (Eriksson et al., 2005). Furthermore, biofouling may damage protective coating and interfere with sensitive areas necessary for monitoring and maintenance (Durr and Thomason, 2010). Therefore, knowledge of biofouling composition will help inform effective design and maintenance procedures to optimise the performance of devices. This study focuses on a novel method which will be used to estimate hydrodynamic loading on fouled structures paying special attention to the needs of the renewable energy industry. Table 1: Industrial considerations related to biofouling on renewable energy devices.
Consideration
Description
Reduced running costs
Reduced capital costs
Information concerned with biofouling will allow better predictions of Construction costs
hydrodynamic loading thus allowing for structural components to be adequately specified but not over-engineered. Profits may be increased for longer running projects by ensuring
Deployment time
biofouling has a minor impact on the functioning of the device i.e. the integrity of protective coatings.
Inspection and maintenance of
If occurrences of maintenance (most notable the removal of devices)
sensitive areas i.e. efficient heat
increase the running costs will be greater. Removal of biofouling may be
transfer systems
necessary for inspection, recoating and ensuring device efficiency.
Maintenance of power generation
Biofouling impacts on power generation efficiency reducing power
efficiency
generation capacity over the life-span of the project Microbial biofouling increases corrosion rates by creating anaerobic
Maintenance of protective coatings
conditions. Mechanical damage of protective coating also increase corrosion rates.
4
3.1 Introduction to estimating hydrodynamic loading Estimates of hydrodynamic load on offshore structures are commonly made using the Morrison’s equation to estimate the hydrodynamic force, F (Jusoh and Wolfram, 1996b). The force on a structure or element of a structure is the sum of the drag force and the inertia force (Equation 1). Equation 1:
Equation 2:
̇
| |
Where; ρ is the water density, CD is the drag coefficient, CM is the added mass coefficient, De is the members effective diameter, U if the water velocity in the direction of force and ̇ is the water acceleration in the direction of force. Biofouling has the effect of increasing the effective diameter (De) (Figure 1) (Equation 3 and 4). Equation 3: Equation 4:
A
B
Figure 1: Modification of external structure due to biofouling. A) Biofouling is composed of soft and hard structures. B) The area (diameter) and roughness of structural elements exposed to flow is increased as a result of biofouling. Figure taken and adapter from Shi et al (2012). Both the thickness and roughness characteristics will change as community composition is altered. This means that as the physical and environmental conditions are altered, such as, depth and geographical area, the characteristics important for estimating hydrodynamic loading are also altered. As a general rule, rougher structures tend to produce thicker
5
boundary layers. Therefore, estimates of the drag coefficient and added mass coefficient tend to be dependent on both the roughness height relative to the diameter of the structural element (k/De) and to the Reynolds number (Re) of the system (Theophanatos and Wolfram, 1989). In summary, increases in the drag coefficient, added mass coefficient and effective diameter can result in a larger Force acting on that structure (Gudmestad and Moe, 1996). Due to the variability in biofouling characteristics and environmental conditions, off-shore hydrodynamic design considerations tend to follow common recommendations such as the Stiftelsen Det Norske Veritas (DNV) classification society (Gudmestad and Moe, 1996). Parameters used for hydrodynamic loading estimates tend to be generalisations and have been acquired largely through experience and practice. Information regarding the key characteristics of biofouling on MREDs will assist engineers to make better predictions of loading. The majority of marine renewable energy devices will be deployed in challenging off-shore locations. Therefore, the least problematic and cost-effective method to assess the characteristics of biofouling present will be to develop in situ methods. In this way, costly techniques to manage biofouling, such as the removal and cleaning of the devices by chartering large vessels, may be kept as a final option. Routine inspection of the device may be achieved via visual methods such as the examination of video taken from Remotely Operated Vehicles (ROVs). Video footage collected from ROVs also presents an opportunity to investigate the characteristics of biofouling allowing engineers to better assess the need to manage the impact on the device. Novel visual computing techniques predominantly used in terrestrial environments may provide a method to utilises readily available video footage from routine inspection.
4. Novel method for estimating biofouling roughness, thickness and volume Structure from Motion (SfM) is a recently developed digital technique that allows the interpretation of 3D structure from overlapping photographs (Snavely et al., 2008, Chen et al., 2011) (Figure 2). The SfM technique can be used to obtain a sparse point cloud from a collection of overlapping photographs. This technique relies on matching unique features in images taken from differing viewpoints. In this way a surface can be recreated digitally from a set of related photographs. This technique can be used to study biofouling and has a number of advantages over alternate methods to obtain equivalent data. Firstly, this passive technique
6
uses no complex sampling equipment and utilises tools already in used for visual inspection of underwater structures. Secondly, digital created models of biofouled structures are directly compatible with existing Computer-Aided Design (CAD) drawings of the device. This would allow for a more accurate estimation of the volume of biofouling covering the device. Thirdly, surface characteristics such as mean roughness (k) may be estimated from surface reconstruction allowing for better estimates to be made for drag and added mass coefficients. Lastly, soft biofouling structures which move between images are lost during the recreation of the point cloud and therefore allow for more faithful estimates of the hydrodynamic characteristics of the resident biofouling community.
Figure 2: A screenshot of VisualSfM dense reconstruction of a tree
This method was investigated using several available open source programs. However, particular attention was given to the open source software, VisualSfM, developed by Changchang Wu (downloadable at: http://homes.cs.washington.edu/~ccwu/vsfm/ date: 08/03/2013). ROV footage of biofouled man-made structures was then loaded into QuickTime 7.7.3 (Apple Inc.) and relevant images were extracted at a rate of 10 frames per second. These images were then loaded in to VisualSfM and feature detection with full pairwise image matching was accomplished. A sparse reconstruction was then carried out before running a dense reconstruction using the Clustering views for Multi-View Stereo (CMVS)
package
developed
by
Yasutaka
Furukawa
(downloadable
at:
http://www.di.ens.fr/cmvs/ date: 08/03/2013). The final model output was viewed in the open source CAD software MeshLab V1.3.2 (downloadable at: http://meshlab.sourceforge.net/ date: 08/03/2013) (Figure 3).
7
A
B
Figure 3: An example of the output using the Structure from Motion (SfM) technique. A) One of 600 stills extracted from a moving ROV video footage (downloaded at: https://www.youtube.com/watch?v=LdmldiiIamw date: 09/04/2013). B) Two different renders of the digitaly reconstructed surface of a biofouled off-shore structure produced in MeshLab V1.3.2 note the vertical ridges which are due to the underlying structure and are clearly seen in the video stills.
8
Other methods exist to create 3D point cloud of structures underwater for inspection. Perhaps the most advance of these technologies has been developed by Teledyne BlueView. This company has developed 3D acoustic multibeam scanners capable of digitally reconstructing underwater structures. The key advantage of this method is its reduced sensitivity to water quality and light attenuation. However, this method requires costly equipment, expertise and is tailored for larger scale reconstructions. Therefore, this method is not suitable for use with the off-shore renewable energy industry for this application.
5. Collecting data from off-shore navigation buoys Across the ‘spectra’ of renewable energy technologies, deployment will take place within a range of environments from relatively low hydrodynamic stresses, more suitable for off-shore wind, to the extreme hydrodynamic stresses, characteristic of wave and tidal devices. To evaluate the extent of biofouling on off-shore structures we need to first examine what types of communities are common on off-shore artificial structures. To address this question, requires a large number of similar artificial structures in both ‘high’ and ‘low’ flow environments. By utilising a large-scale network of navigation buoys in Scotland and the Isle of Man, Macleod (2013), investigate the community composition of off-shore artificial substrates in a range of environments and geographical locations. Information of community composition along with the reprocessing samples allowed for biological information (present coverage of taxa) be used to predict the density of biofouling on structures. 5.1 Site selection and experimental approach Using the Geographical Information Systems (GIS) software (ArchGIS Esri), a model was constructed using five data layers, including tidal current regimes (mean peak spring tidal speed ms-1) taken from the UK Atlas of Marine Renewable Energy Resources (BERR, 2008). It should be noted that the tidal data layer model produced a relatively large grid size (≤1.8 km). This layer was based on the best available knowledge gathered and functioned primarily to estimate relative tidal conditions to which navigation buoys were exposed. Admiralty data of tidal streams were also used to verify current strength. Bathymetry and navigational buoy data, in addition to other geographical features, were provided by EDINA (Digimap Ordnance Survey Service, 01/01/2011). The use of a GIS model allowed for the querying of data to select floating navigation buoys subject to varying water flow rates in different geographical locations (Figure 4).
9
Figure 4: The location of navigation buoys sampled in different geographical areas.
A total of five trips were made on the NLV Polestar to four NLB areas, sampling 30 navigation buoys, approximately one third of the navigation buoys that the NLB maintain (Table 2). The first trip to the Sound of Harris took place during May 2010, the remaining trips to the Skye area, Orkney Isles, Clyde area and the Isle of Man took place during the spring and summer of 2011. Although not always adhered to, the annual schedule for navigation buoys maintenance and replacement generally followed a similar pattern from spring to summer. The Sound of Harris and Skye buoys were typically visited in spring, whilst the Orkney, Clyde and Isle of Man buoys were typically visited in the summer. All four geographic areas had navigation buoys occupying a range of tidal exposures and the majority were adjacent to proposed future tidal developments (Harrald et al., 2010).
10
Table 2: Navigation buoys, duration of deployment, flow conditions and geographical area. Buoy Name
Black Rocks Macosh Rock Whitestone Bank Bahama Bank Barr Rock Egilsay Graand Peter Skerry Roddock shoal Sand Eel Billia crew East Billia crew North Billia crew South Billia crew West Boray Skerries Eday Gruna Flotta Grinds Galt Skerry Linga Skerry Royal Oak Skertours String Rock Black Eye Rock Bow Rock Fork Rock Gulnar Rock Jackal Rock Na Gamhachain Racoon Rock
Duration of deployment (years)
Flow Category
Geographical Area (NLB)
Latitude
Longitude
3 4 4 1 2 1 4 6 1 7 7 7 7 5 3 3 3 2 3 3 4 3 4 2 3 2 5 1
High Flow High Flow High Flow Low Flow High Flow High Flow High Flow High Flow High Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow High Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow Low Flow
Clyde Clyde Isle of Man Isle of Man Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Orkney Skye Skye Skye Skye Skye Skye Skye Skye
55 47.503 N 55 17.950 N 54 24.599 N 54 20.022 N 58 56.605 N 59 06.870 N 58 56.259 N 58 55.890 N 58 56.417 N 58 58.386 N 58 59.522 N 58 57.430 N 58 58.530 N 59 03.659 N 59 08.386 N 58 50.970 N 59 05.225 N 59 02.395 N 58 55.746 N 59 04.118 N 57 16.490 N 57 16.706 N 57 16.762 N 57 16.836 N 57 19.148 N 57 20.340 N 57 35.890 N 57 16.152 N
006 04.082 W 005 36.970 W 004 20.375 W 004 08.549 W 003 17.000 W 002 54.400 W 003 13.515 W 003 15.000 W 003 15.342 W 003 22.388 W 003 23.750 W 003 23.047 W 003 24.634 W 002 57.643 W 002 43.846 W 003 00.773 W 002 54.182 W 002 57.557 W 002 59.186 W 002 56.704 W 005 42.890 W 005 45.305 W 005 45.916 W 005 44.935 W 005 55.856 W 006 04.758 W 005 57.714 W 005 35.198 W
The Isle of Skye area forms part of the inner-Hebridean islands, where a complex tidal system exists between the islands and the mainland. Most notable is the Kyle Rhea and Kyle of Lochalsh, which have strong tidal flow (over 3 ms-1) and tidal renewable energy schemes are currently under development (BERR, 2008). Eight buoys were sampled in this region; 7 designated as ‘low’ flow and 1 designated as ‘high’ flow.
The Orkney Isles in the north-east of Scotland have complex tidal streams, reaching speeds of over 4 ms-1 (BERR, 2008). The Orkney Isles are rich in tidal and wave energy resources and as a result this area is a leading location in the development of wave and current MREDs (Moore et al., 2007). Navigation buoys directly adjacent to the wave test site (Billia crew) and the tidal test site (Falls of Warness) were sampled. Buoys in other locations, such as the Sound of Hoy, where tidal currents regularly exceed 3 ms-1 were also sampled. Sixteen buoys were sampled in this region; 11 designated as ‘low’ flow and 5 designated as ‘high’ flow.
11
The Clyde area included one navigation buoy in the Sound of Islay and one navigation buoy at the southern tip of the Mull of Kintyre. Scottish Power Renewables is currently developing the tidal resources (speeds over 4 ms-1) in the Sound of Islay with the installation of 10 large turbines. Whilst two companies (Nautricity Ltd and Argyll Tidal Ltd) are currently developing the tidal resources (speeds over 4 ms-1) around the southern end of the Mull of Kintyre. In total, two buoys were sampled in this region; both designated as ‘high’ flow.
The Isle of Man was the southernmost site visited. Two buoys were sampled in this area, one of which was exposed to ‘high’ flow over 1 ms-1, where strong tides exist off the northern tip of the island. The other buoy was located in a ‘low’ flow environment to the east of the island.
Navigation buoys maintained by the NLB are serviced annually at approximately the same time each year (Sean Rathbone, NLB, pers. comm.). However, due to many external influences, such as responding to faulty navigation aids, scheduled servicing varies greatly. Maintenance includes: checking the mooring system and navigational aids, in addition to cleaning with scrapers and pressurised water. The parts of the buoy that are cleaned on an annual basis include; the top of the buoy, the mooring chain connectors and the top wall (Figure 4). The slope and the skirt are not cleaned annually and communities develop over the deployment time of the buoy. Approximately every 6 years, the buoys are returned to the main base in Oban where they are painted with a non-biocidal, anti-corrosion paint (International paint, Intersheen) (Sean Rathbone, pers. comm.).
12
Figure 5: A navigation buoy used by the Northern Lighthouse Board showing distinct sections; top wall, slope, and skirt.
The photographic method involved taking five digital photos at the highest point on the skirt (0.7 m depth) (Figure 5) using a camera (Olympus Tough 8010, 5.0x optical zoom, 14 mega pixel) mounted on a 20 cm x 20 cm quadrate frame and at a distance of 35 cm from the surface of the buoy. Each photo was equally spaced around the circumference of the buoy (approx. 1.2 m apart). A total of 227 photographs were taken from 30 buoys sampled. Images were
processed
using
the
programme
Vidana
1.0
(download
at
http://projects.exeter.ac.uk/msel/vidana, 31/01/2011) to calculate percentage area cover of each biofouling taxa (Figure 6). Where organisms were identified down to species level (Hayward and Ryland, 1990) their percentage cover was calculated. However, those flora and fauna that were not easily identified to species level were grouped at a higher taxonomic level.
13
Figure 6: A screen shot of the program Vidana used to analyse photographic samples. Sample taken from the Bo Quidam buoy (Sound of Harris) which had been deployed for 3 years.
A destructive scrape sampling method was used to remove all biofouling biomass from the buoy into a specially designed container (2000 cm3), which had a 20 cm x 20 cm quadrate attached for reference. Material was transferred to labelled containers, which were filled with a solution of 10% (v/v) buffered formalin for fixation and transportation back to the laboratory. Subsequently, material was rinsed thoroughly using a one millimetre sieve (Hartley, 1982). Biofouling material was drained thoroughly and the wet weight (0.01g) of the entire sample was measured using an electronic scale (Sartorius, Universal U6100S). This material was then added to a large a large water filled plastic pipe and the displacement of water was used to measure the volume (0.01 mL) of material in the sample (Figure 7).
14
B A
C
Figure 7: The apparatus for measuring displacement. A) Electronic scales, B) clear water filled tube, C) sample.
5.2 Statistical analysis The percentage cover of different functional groups was estimated through the addition of similar taxonomic groups [hard fouling species (examples; barnacles, bivalves and shelled gastropods) and soft fouling species (examples; hydroids, algae and crustacea)]. Predicting density of biofouling material from percentage cover of functional groups was achieved by firstly measuring the weight and calculating the corresponding volume for each sample. Trends in density and the percentage cover of functional groups and flow rate (ms-1) were investigated. Data were fitted to a generalised linear model with a quasipoisson distribution to determine whether increased representation functional groups influenced the density of biofouling biomass (R Development Core Team, 2012). Model fit and selection were performed using ANOVA
(Zuur et al., 2009). Model validation included checking a
histogram of Cook's distances for influential data points and comparing calculated leverage threshold and estimating model leverage. Variance inflation factors were calculated to check
15
for co-linearity between the predictor variables (Quinn and Keough, 2002), and none of these indicated obvious influential cases or outliers.
5.3 Result The proportion of soft fouling (functional group 2) was dropped from the model during optimisation. The process reduced the number of key predictor variables to aid interpretation of the results. The final model used the proportion of hard fouling (functional group 1) to predict the density of biofouling. The proportion of hard fouling had a highly significant (P<0.0001) influence on the density of biofouling measured (Table 3 and Figure 8). Table 3: Summary of the primary model output.
Fixed effects (Intercept) Propotion of hard fouling
Slope Standard Error t value P value 0.0886 0.2133
0.0132 0.0279
6.731 7.654
<0.0001 <0.0001
Figure 8: The modelled fit between the proportion of hard fouling and density. This can be used to estimate corresponding density from photographs.
16
6. Discussion The above data can be used in a predictive manor to apply knowledge of the percentage cover of hard fouling species to predict the density of the material on the structure. Once informed about the density of the biofouling material, information relating to the predicted volume for example that produced by the SfM technique, can be used to inform engineers on a real time basis. As more data become available the reliability of predicted estimates will also be enhanced.
7. The impact of biofouling characteristics on hydrodynamic loading In addition to determining biofouling characteristics on navigation aids in/or adjacent to areas currently under development, this project aimed to investigate the impact of different biofouling characteristics on hydrodynamic loading. This was achieved by working with the Exeter University Renewable Energy Research group who specialise in off-shore mooring engineering for wave energy converters (WECs). Using data generated from the navigation buoy survey alongside other published figures a series of models were run using the OrcaFlex software (Orcina Ltd). This software is arguably one of the world’s leading packages for the dynamic analysis of off-shore marine systems. Its key features include adjusting the environmental conditions that the system is exposed too as well as the characteristics of the mechanical system (Table 4). The mooring systems for WECs must be designed to withstand the environmental loadings experienced whilst maintaining the device on station (Harris et al., 2004). If the drift is too large some part of the chains sitting on the seabed will be lifted on the mooring lines on the leading edge of the buoy (i.e. the line exposed to wind, waves and current). The consequences of this are that mooring loads may dramatically increase in the front mooring lines and also vertical loads may occur on the anchor if the chains are fully lifted (Harris et al., 2004). Secondly, the power cable exporting the electrical energy produced by the wave energy converter may become taut, which should be avoided (Harris et al., 2004). Lastly, in an array configuration, several devices may be installed close to each other and contact between devices should be avoided (Harris et al., 2004). Properly designed mooring systems must also maintain the device in the proper orientation relative to incoming waves, and must interact with the device to optimise the control system for the specific bandwidth of the WEC (Harris et al., 2004). Finally each mooring system must be designed with redundancy built to safeguard against mechanical failure (Harris et al., 2004). Therefore, the impact of biofouling on off-shore renewable energy devices was investigated by
17
focussing on the mooring system which represents an integral part of the proper function of all off-shore WECs. This was achieved by focusing on a system currently being developed and exploring the impact of altered biofouling characteristics on the function of the device.
The OrcFlex model (Figure 8), based on the south west mooring test facility consists of a buoyant object similar to a navigation buoy with a tri-axial mooring lines produced from 44mm nylon rope. Outer diameter, mass, density, drag coefficient and added mass coefficient were altered to include the full variation in published values (Table 4). The model was run for a total of 300 seconds to resolve relevant fluctuations over time with 990 unique combinations of biofouling characteristics. The mean effective tension (kN) was investigated over each time series on the 2nd mooring line as this line was extended into the principal current direction. Moreover, the mean drift of the buoy (horizontal direction) was investigated as this is an important parameter for the design of a mooring system. Under extreme loading situations may of the buoys sunk (i.e. the mean position z was <-1m). These were removed from the analysis. The OrcaFlex program provided a valuable approach to resolving a large number of calculations simultaneously. This method had clear advantages over more complex Computational Fluid Dynamics (CFD) calculations that would have proven too inefficient and costly to execute.
18
Table 4: An extract of the environmental and mechanical features of the Orcaflex model. Bold values represent default configuration. Outer diameter, mass, density, drag coefficient and added mass coefficient were altered.
Environmental variable
Mechanical system variable (44mm nylon rope)
Water salinity (%)
3.5
Outer diameter (m)
0.037, 0.037, 0.057, 0.077, 0.097, 0.117, 0.137, 0.157, 0.177, 0.197, 0.217, 0.237
Water temperature (˚C)
10
Mass (Te/m)
0.0013 + corresponding mass for all outer diameter values at densities of 1.0, 1.3 and 1.6 Te/m3
Water density (Te/m3)
1.025
Axial stiffness (kN)
228.448
Water depth (m)
28
Torsional stiffness (kN.m2)
80.000
Seabed type
smooth
Drag coefficient (CDx)
0.8,1.0,1.2,1.4,1.6,1.8
Current speed (ms )
1
Drag coefficient (CDz)
0.0080
Wave height (m)
1
Added mass coefficient (CMx)
0.0,0.6,0.8,1.0,1.2,1.6
Wave period (s)
7.5
Wave type
JONSWAP
-1
19
Figure 8: A screenshot of an OrcaFlex simulation showing the mooring configuration.
20
7.1 Results The impact of each biofouling characteristic was investigated qualitatively by graphing the mean effective tension and mean buoy horizontal movement over the 300 second simulation (R Development Core Team, 2012). The environmental conditions that the mechanical systems were exposed to was kept constant (Table 4). The key features of the bioufouling community that increased the effective tension were the mass of biofouling, the outer diameter of the line and the density of the biofouling (Figure 9). These features of the biofouling community are heavily co-dependent. The thickness of biofouling present results in the mooring system to have increased mass which is modified by the density of material present. The key inferences that can be made are that the additional mass added to the mooring system has the direct effect of increasing the mean effective tension on the mooring system. In the most extreme cases the tension was so great that it overcame the buoyant force of the structure resulting in its sinking. Notably, the added mass coefficient and the drag coefficient had little impact on the effective tension experience by the system.
As the biofouling thickness and density increased a general reduction in buoy horizontal movements was observed as additional mass was added to the mooring system (Figure 10). A clear trend in increased horizontal movement was observed as the drag coefficient increased. The added mass coefficient had little impact on the buoys horizontal movement (Figure 10).
21
Figure 9: The influence of different biofouling characteristics on the effective tension of the 2nd mooring line.
22
Figure 10: The influence of different biofouling characteristics on the buoy movement.
23
7.2 Discussion Biofouling was clearly observed to have a negative impact on the function of this mooring system under simulated conditions. The environmental conditions chosen for these simulations are very common and do not represent extreme conditions experienced during storm events. In the most extreme cases of biofouling the additional mass lead to the buoyancy force of the buoy to be overwhelmed. These results were excluded from the analysis as it is assumed that engineers would anticipate this occurrence and engineer more buoyant structures. However, even if the device does not sink the impact of the lowered position of device may be to reduce the energy capture efficiency of the device (Harris et al., 2004).
As minimum and maximum thickness and density estimates were made from observations and taken from published data it is not unreasonable to anticipate given enough time biofouling may have a massive impact on the functionality of WECs.
In this particular system it was the mass of biofouling which resulted in increased tension in the mooring line and not the inertia coefficients. Therefore, practical means of estimating inertia coefficient appear less important than making estimates of the mass added to the mechanical system. The methods detailed above provide a means to estimate the mass added. However, other methods may prove as reliable.
When considering the horizontal drift of the buoy added mass had a beneficial impact on the system through reducing the drift. This is perhaps unsurprising as a heavier system may be more inclined to resist environmental forces resulting in reduced surface movement in a horizontal direction. The drag coefficient tended to increase the buoys horizontal movement. This suggests that lines with lower drag coefficients and greater mass will result in the least horizontal movement for WECs. The greatest values chosen for the drag coefficient (1.6 and 1.8) represent the most extreme cases (i.e. 100% covering of long kelp on lines) with typical values likely to be between 0.8 and 1.4 (Theophanatos and Wolfram, 1989, Gudmestad and Moe, 1996). If model simulations are accurate important inferences can be made for the management of WECs. Firstly, if large roughness elements exist in the biofouling community, for example, a large cover of kelp species, then drag coefficients are likely to be considerable and should be considered with care. However, if relatively small roughness
24
element exist (i.e. typical k/D values) then attempts to estimate variable drag coefficients will be less valuable overall.
8. Conclusions The project attempted to investigate the needs of engineers when considering the impact of biofouling on off-shore renewable energy devices. At present, biofouling appears to fall short of the principal engineering concerns. However, this is likely to change as devices become more developed and the success of larger commercial projects depend on longer deployment periods. The primary focus at present for engineers is to ensure that biofouling considerations are met through following recommendations such as the Stiftelsen Det Norske Veritas (DNV) classification society (Gudmestad and Moe, 1996). These recommendations are designed to protect property and ensure safety. However, many MREDs may have to consider biofouling more closely to optimise the performance of the device.
This project considered the characteristics of biofouling that are important for engineers when estimating the hydrodynamic performance of MREDs by gathering information from engineers first hand and through reading the published literature. An attempt was made to add to an existing data set and predict the characteristics of biofouling by geographical area, tidal strength and deployment length. This approach proved very challenging as remodelling surfaces from stereo-paired images was very time consuming and many possessed anomalies which had to be edited further.
In response to this, a low cost method to investigate
biofouling characteristics was investigated by applying a developmental software package (VisualSfM). This produced very encouraging results and provided a tool which will allow for all key characteristics of the biofouling community to me estimated in situ. This project also produced a tool to estimate the density of the biofouling material by calculating the cover of hard fouling species.
One major economic consideration for the renewables industry is the maintenance and survivability of devices (Mueller and Wallace, 2008). This work provides relevant methods for engineers, operators and developers to make more accurate predictions on loading of their devices. This may be of particular relevance to the marine renewable energy industry as maintenance of devices in highly energetic environment will be a costly and difficult task.
25
This work also generated an increased understanding of how biofouling of mooring systems affects the loadings and drift of WECs. Continued investigation in this area will ultimately lead to new MRED design criteria that reduce operational down time, reduce H&S concerns during
deployment
and
improve
operational
efficiencies
by
reducing
fouling.
26
27
An Assessment of the Impacts of: De-risking structural loading calculations for Marine Renewable Energy Devices NERC contribution to impact: Excellent science
Science outcome / breakthrough
a This three month Developed method for project aimed to underwater visual combine ecological sampling. data along with modern statistical tools to develop an Developed a tool for predicting approach for better density from predicting the key species physical composition. characteristics of biofouling communities useful Investigated the impact of when estimating biofouling structural loading on characteristics on off-shore renewable the function of a energy structures. To wave energy do this, a converter using multidisciplinary OrcaFlex approach was taken working with marine scientists, engineers and developers to assess the needs of the industry. The ability to better predict biofouling community characteristics will serve as a de-risking strategy at the design and development phase. Knowledge needed to improve the current state of practice with regards to managing biofouling on renewable energy devices is identified and included in the final report.
Impact: realised to date
Impact: potential scope
Ecological data gathered was used to better inform the loading models on the south west mooring test facility.
The monitoring of biofouling by engineers is a key element of ensuring proper functioning of underwater structures in addition to protecting assets and investment. Not only does this work address biofouling on off-shore renewables but it also has implications for the off-shore oil industry in particular the fuction and safeguarding of flexible riser systems.
Information was disseminated at the All energy conference to a large number of developers and associated companies. Discussion have taken already taken place regarding the outcomes of this KE activity with the following organisations: •Scottish Power Renewables •Sustainable Marine Energy Ltd •Alstom Power Ltd
Tidal
This work will be made available to engineers requiring better methods for prediction hydrodynamic loading and managing the impact of biofouling on underwater structures
•Rolls-Royce •EMEC – European Marine Energy Centre •International Paint Ltd
PML applications, Exeter university and SAMS research services LTD
28
9. References BERR 2008. Atlas of UK Marine Renewable Energy Resources: Tidal. . www.renewables-atlas.info/ (Accessed 07 June 2012) ed.: Department for Business, Enterprise and Regulatory Reform. CHEN, S., WANG, Y. & CATTANI, C. 2011. Key issues in modeling of complex 3D structures from video sequences. Mathematical Problems in Engineering, 2012. COUTTS, A. D. M., PIOLA, R. F., HEWITT, C. L., CONNELL, S. D. & GARDNER, J. P. A. 2010. Effect of vessel voyage speed on survival of biofouling organisms: implications for translocation of non-indigenous marine species. Biofouling, 26, 1-13. DENNY, M., GAYLORD, B., HELMUTH, B. & DANIEL, T. 1998. The menace of momentum: Dynamic forces on flexible organisms. Limnology and Oceanography, 43, 955-968. DURR, S. & THOMASON, J. 2010. Biofouling, Chichester, Blackwell ERIKSSON, M., ISBERG, J. & LEIJON, M. 2005. Hydrodynamic modelling of a direct drive wave energy converter. International Journal of Engineering Science, 43, 1377-1387. GILL, A. B. 2005. Offshore renewable energy: ecological implications of generating electricity in the coastal zone. Journal of Applied Ecology, 42, 605-615. GUDMESTAD, O. T. & MOE, G. 1996. Hydrodynamic coefficients for calculation of hydrodynamic loads on offshore truss structures. Marine Structures, 9, 745758. HARRALD, M., AIRES, C. & DAVIES, I. 2010. Further Scottish Leasing Round (Saltire Prize Projects): Regional Locational Guidance. Scottish Marine and Freshwater Science. HARRIS, R. E., JOHANNING, L. & WOLFRAM, J. 2004. Mooring systems for wave energy converters: A review of design issues and choices. Marec2004. HARTLEY, J. P. 1982. Methods for monitoring offshore macrobenthos. Marine Pollution Bulletin, 13, 150-154. HAYWARD, P. J. & RYLAND, J. S. 1990. The marine fauna of the Brittish Isles and north-west Europe, Oxford, Clarendon Press. JUSOH, I. & WOLFRAM, J. 1996a. Effects of marine growth and hydrodynamic loading on offshore structures. Jurnal Mekanikal, 1, 77-96. JUSOH, I. & WOLFRAM, J. 1996b. Effects of marine growth and hydrodynamic loading on offshore structures. Jurnal Mehanikal, 1, 77-96. LEONARD, G. H., LEVINE, J. M., SCHMIDT, P. R. & BERTNESS, M. D. 1998. Flowdriven variation in intertidal community structure in a marine estuary. Ecology, 79, 1395-1411. MACLEOD, A. 2013. The role of Marine Renewable Energy structures and biofouling communities in promoting self-sustaining populations of Non-Native Species. PhD Thesis, Aberdeen. MOORE, J. R., BROWN, C. E. & COOPER, W. S. 2007. Quantification of exploitable tidal energy resources in UK waters. In: LTD, A. M. E. R. (ed.). Southampton. MUELLER, M. & WALLACE, R. 2008. Enabling science and technology for marine renewable energy. Energy Policy, 36, 4376-4382.
29
NOWELL, A. R. M. & JUMARS, P. A. 1984. Flow Environments of Aquatic Benthos. Annual Review of Ecology and Systematics, 15, 303-328. ORME, J. A. C., MASTERS, I. & GRIFFITHS, R. T. 2001. Investigation of the effect of biofouling on the efficency of marine current turbines. Institute of Marine Engineers, 91-99. QUINN, G. P. & KEOUGH, M. J. 2002. Experimental design and data analysis for biologists, Cambridge, UK ; New York :, Cambridge University Press. R DEVELOPMENT CORE TEAM 2012. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. SHI, W., PARK, H.-C., BAEK, J.-H., KIM, C.-W., KIM, Y.-C. & SHIN, H.-K. 2012. Study on the marine growth effect on the dynamic response of offshore wind turbines. International Journal of Precision Engineering and Manufacturing, 13, 1167-1176. SNAVELY, N., SEITZ, S. M. & SZELISKI, R. 2008. Modeling the world from internet photo collections. International Journal of Computer Vision, 80, 189-210. THEOPHANATOS, A. & WOLFRAM, J. 1989. Hydrodynamic loading on macroroughened cylinders of various aspect ratios. Journal of offshore mechanics and Arctic engineering, 111, 214-222. VOGEL, S. 1994. Life in moving fluids: The physical biology of flow, Second edition, Princeton University Press; Princeton University Press. ZUUR, A. F., IENO, E. N., WALKER, N. J., SAVELIEV, A. A. & SMITH, G. M. 2009. Mixed Effects Models and Extensions in Ecology with R, Springer, 233 Spring Street, New York, Ny 10013, United States.
30
