Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK

A NOVEL DESIGN FOR AN OFFSHORE WIND FARM VESSEL: APPLICATION OF

THE AERODYNAMICALLY ALLEVIATED MARINE VEHICLE (AAMV)
D James and M Collu, Cranfield University, UK

SUMMARY

With the recent third round of site allocations for offshore wind farms in extended UK waters, new challenges for
efficient operation and maintenance require new solutions to be provided for technician and equipment transfer out to
200 nm from shore. Based on the ongoing work at Cranfield University, a representative methodology for the design of
an innovative Aerodynamically Alleviated Marine Vehicle (AAMV) is demonstrated. This process builds upon previous
work including theoretical and experimental models, culminating with the summary of a preliminary design for a vessel
of similar capability.

Utilising aerodynamic efficiencies of wing-in-ground effect (WIG) craft, it is shown how a vessel can be equipped with
lifting surfaces in order to alleviate the weight of the vehicle, leading to a lower effective displacement, drag and
required power. The design spiral of conventional marine craft is modified to include the relevant considerations to
equilibrate the aerodynamic forces and moments. Some areas of current and future work are discussed, with
experimental results presented.

1. INTRODUCTION and safety risk assessments for the large number of

It is well-known that an aerodynamic lifting surface in
close proximity to a ground plane theoretically offers an
opportunity to achieve more efficient flight than is
possible in free flight conditions [25]. This paper
outlines the some challenges in the development of a
hybrid class of vessel that is able to bridge the maritime-
to-air domain, comfortably operating in the water surface
yet still delivering the speed of aircraft during an
airborne cruise phase. A methodology for sizing an
appropriate platform is presented, as well as
experimental results of a high speed hullform with
characteristics suitable for this intended application.
This concept has been termed an Aerodynamically
Alleviated Marine Vehicle (AAMV), as it employs
aerodynamic lifting surfaces to counteract the
gravitational weight of the vehicle, resulting in a lower
effective displacement, wetted surface, and ultimately
resistance for a given speed. The resulting increase in
transport efficiency using this type of configuration
promises the ability to move people and cargo over water
with much greater comfort, speed and efficiency than is
currently achievable by any other method.
By equipping a vessel with aerodynamic lifting surfaces
in order to alleviate its weight, a lower effective
displacement can be achieved. This concept has been
proven to function successfully, although there has been
little mainstream interest in these craft.
1.1 CONTEXT
The greatly increased distances from shore planned for
the Round 3 wind farm developments raise a number of
operational concerns, including the requirement for
cheaper and more cost-efficient ways of transporting
technicians and associated equipment, as well as health
anticipated helicopter journeys [1].
The current method for surface transfer of personnel is
by specially-designed small aluminium catamarans, with
typical speeds around 20 kt, with a standard payload of
12 technicians plus limited equipment. If a greater speed
or shorter transit duration is required the choice is then
for helicopter transfer. This option is very expensive
compared to the marine craft, payload is generally
limited to about five technicians, and fatal crashes and
other dangerous events have raised health and safety
concerns.
Wing-in-ground effect (WIG) marine vehicles can
provide a third option for wind farm operators. It has
been stated that innovation is needed to minimise lost
time of not being able to get personnel onto a rig [1] and
WIG craft fit the concept of operations for offshore wind
farm service. These vessels would be faster and possibly
more comfortable than existing high speed crew transfer
vessels, and significantly cheaper than helicopters whilst
offering greater payload capacity and superior station
keeping characteristics.
1.2 PROBLEM STATEMENT
WIG craft aim to exploit the enhanced lift and reduced
drag due to the ground effect, having the potential to be
more efficient than an equivalent aircraft configuration
flying out of ground effect, and at speed much greater
than high speed marine vehicles. However, WIG craft
designed to date have not been able to fully exploit this
theoretical advantage, with reduced efficiencies often due
to the inclusion of additional lifting and/or control
surfaces and other required design compromises [8, 20].
Additionally, WIG vessels so far have not demonstrated
sufficient robustness for uninterrupted commercial

2015: The Royal Institution of Naval Architects

 Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
operations, with frequent structural failures or
manoeuvrability concerns. These issues may be
rectified, or at least reduced, by designing a vehicle with
inherently different characteristics, minimising their
vulnerability in the marine environment.
Novel craft without a long history must be designed by
physics-based methods from the beginning [16],
therefore the aim of the present work was to develop a
suitable method to estimate the structural weight and
centre of gravity position, and to design and test a
suitable hullform with favourable characteristics in order
to contribute to the overall conceptual preliminary
design.
2. WIG STATE OF THE ART
A general review of the latest research on wing-in-
ground effect vehicles has been undertaken, highlighting
some of the main technological challenges. The promise
of increased aerodynamic efficiencies demonstrated at
the theoretical level have not been easily achieved, often
due to matters of stability, hydrodynamics, structural
design and operational practicalities. Longitudinal pitch
stability during transient and airborne modes has been
one of the main challenges to be resolved. Additionally,
overcoming the hydrodynamic hump drag has proven
problematic, often requiring significantly higher power
during the take-off phase than at any other time in the
operational profile. Whilst several general methods have
evolved to address these issues, the limitations imposed
by various configurations remain impediments to more
efficient and effective designs.
2.1 AERODYNAMIC EFFICIENCY
It is well known that aerodynamic performance
characteristics are strongly influenced by wing profile,
aspect ratio, and the presence of other surfaces and
geometric features. Increasing the aspect ratio is known
to increase aerodynamic efficiency, although there are
various factors limiting wing span in practice. Certain
manoeuvres applicable to all WIGs, such as roll or
banking in a turn, are limited by wing span and height
above surface. Some types of WIG rely on a dynamic air
cushion under the main wing, where a more compact
platform has been shown to be more effective [27].
Additionally, the span of the wing may be limited by the
port facilities where the vehicle is intended to operate.
Nevertheless, aspect ratio may be more influential upon
aerodynamic efficiencies than other geometric
parameters [14].
Many existing WIG craft have adopted fixed wing
aircraft configurations, with identifiable fuselage, wing
and empennage structures. Alternative configurations
have been investigated to address a loss of aerodynamic
efficiency, trying to utilise geometries closer to a wing-
body approach. For example, a multihull superstructure
can be shaped as an aerodynamic profile, in order to
exploit the large area available between the hulls and
increase the ground effect due to the hulls acting as end
plates. Such an approach was introduced with the
Ekranocat concept [6], where the effect of having
aerodynamic lift sustaining a fraction of the vehicle
weight was defined as aerodynamic alleviation.
Matveev and Dubrovsky [15] presented a hybrid 1000 t
trimaran that comprises three wave-piercing planing
hulls and a wing-shaped superstructure: based on
numerical simulations and aerodynamic experimental
data, this configuration seemed to be characterised by a
high overall efficiency and good seaworthiness, at speeds
about twice those of contemporary fast ferries and
combat ships.
Due to stability requirements (described in section 2.2),
many WIG configurations use a large horizontal surface
as tail, and this additional element decreases the overall
aerodynamic efficiency of the vehicle. In order to
minimise the size of the tail, S-shaped mean camber
profile families have been designed (e.g. the DHMTU
series) specifically for WIG configurations. Another
solution has been to adopt a reverse delta configuration
for the wing (i.e. Lippisch-type configuration). Despite
the fact that a reflexed aerodynamic profile and a
reverse-delta wing plan area have a lower efficiency than
a clean profile and plan area, they allow a substantial
reduction of the tail horizontal area required, leading to
an increased overall vehicle aerodynamic efficiency.
2.2 STABILITY & CONTROL
Longitudinal pitch stability has been identified as one of
the key design parameters that are critical for success of
WIG vehicles. The necessary proximity of the ground
plane that offers the promise of increased aerodynamic
efficiencies at flying heights not more than 10% of the
wing chord, measured from the horizontal ground to the
trailing edge of the foil, demands sufficient stability and
control to avoid unintended physical contact with the
ground surface [12].
This proximity also complicates the equilibrium states
necessary for stable operation, introducing the concept of
an aerodynamic centre in height in addition to the more
recognised and understood aerodynamic centre in pitch.
The mathematical approach for the determination of
pitch and height static stability for ground effect wings
was begun by Kumar [13] in his research at Cranfield
College of Aeronautics, where the basic problem was
framed and investigated. It was shown how the
rearwards shift of the wing centre of pressure and the
resulting change in pitching moment as the wing
approaches the ground necessitates a centre of gravity
position upstream of the aerodynamic centre to maintain
pitching stability.
For any wing to be stable in flight, the response to any
disturbance from an equilibrium steady state should be to
return to the original position. This can be shown as:
2015: The Royal Institution of Naval Architects
Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
dC M
0 (1)
d
where dCM is the change in pitching moment and d is
the change in angle of attack, or pitch angle.
The phenomenon of ground effect, by definition, states
that there is an increase in lift as the distance to the
ground plane decreases. This can be expressed as:
dC L
0 (2)
dh
where dCL is the change in lift coefficient and dh is the
change in height above ground.
Irodov in the U.S.S.R. and Staufenbiel and Kleinedam in
Germany [9, 24] independently further developed the
mathematical framework to investigate static and
dynamic longitudinal stability of WIG vehicles. It was
shown that there is a change of pitching moment about
the vessel centre of gravity with a change of height above
ground, and what is termed a change in the position of
the aerodynamic centre with height above ground. To
ensure static stability, the relationships must follow:
dC M dC M
d dh 0 (3)
dC L dC L
d dh
Equation (3) can be simplified to show the position of the
pitch and height centres must be located so that:
x p xh 0 (4)
where x indicates the position of the aerodynamic centres
in pitch p, and height h, as measured from the leading
edge and normalised against chord length.
Whilst the requirements for static stability do not dictate
a position of the centre of gravity, it is required to be
located in between the centres in pitch and height in
order to avoid dynamic instability [9, 24].
Figure 1 shows a graphical representation of the
necessary positions of the aerodynamic centres in pitch
and height, and body centre of gravity for airborne
stability near the ground. The implications of these
relationships permeate every aspect of the WIG craft and
must be thoroughly understood before any meaningful
progress can be made toward its development.
2015: The Royal Institution of Naval Architects
Figure 1: Critical positional relationship of aerodynamic and
weight centres
Aerodynamic stability is clearly fundamental to the
success of such a vehicle but is not limited to a singular
cruise-only design condition. Although aerodynamic
forces may be negligible at lower, conventional marine
craft speeds, analysis of the transitional regimes leading
to and from the airborne condition indicate
hydrodynamic forces can be of the same order of
magnitude as aerodynamic forces.
Collu, et al [4] investigated the longitudinal stability of a
generic AAMV configuration in a combined aero-hydro
regime, and documented the third neutral point for this
condition, called the hydrodynamic centre in heave, that
should be considered in AAMV design. In order to have
a static stability condition, this position should be
downstream of the aerodynamic centre in height.
2.3 WATERBORNE PERFORMANCE
As all marine WIG vessels operate directly from the
water surface, and certainly cannot attain flight without
lifting off from this surface, hydrodynamic performance
characteristics play a critical role for overall vehicle
success.
The high power required for take-off has proven to be
one of the greatest impediments to the development of
this type of vehicle [20] as hydrodynamic resistance is
typically the largest contributor to overall vessel drag
during the transitional phases. This limits operation in
increasing sea states and has often led to installed power
in excess of cruise mode requirements. High power
requirements and peaks in the resistance curve as a vessel
attempts to climb out of the water are not new
phenomena, as naval architects have been successfully
designing to overcome this hump speed for more than
half a century.
There have been a number of configurations particular to
WIG craft that have originated with the intent of
reducing hump drag. Examples include vessels with
stepped planing hulls incorporated into the underside of
their fuselage, hydrofoil appendages, power augmented
ram (PAR) employing forward mounted thruster engines
that direct their exhaust under the main wing, and
dynamic air cushion (DAC) that use similar techniques
as conventional surface effect ships to encapsulate the
oncoming air and maintain a high pressure zone on the
underside of a lifting surface as the vessel moves
forward.
 Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
Figure 2: Archetypal Russian design, the KM, skimming above
the sea surface
The most prolific WIG test programme to date, organised
in the former Soviet Union beginning in the 1960s,
typically favoured the PARWIG design. The iconic
images of the Caspian Sea Monster (KM), amongst
others, clearly displays the bow thrusters mounted far
forward, just abaft the cockpit. These thrusters were
throttled back to idle during airborne cruise as they were
only required during the take-off phase.
Figure 3 shows representative resistance curves for
aeroplane-derivative WIG designs, and the effective drag
reduction when using the PAR technique. This data was
compiled from recorded experiences with the Russian
Navy. In spite of the disadvantages of installing these
booster engines for the relatively short time spent in the
take-off phase, the characteristic bow thrusters remained
a hallmark of this type throughout their development
[11].
Figure 3. Ekranoplan-type WIG typical resistance curve with
and without PAR [8]
The exact differences between PARWIG and DACWIG
remain opaque at times, but the fundamental concept of
dynamic air injection is shared by both types. DAC is
often understood to be more suitable for very small craft
due to their ability to operate proportionately closer to
the water surface. Derivatives of this type include craft
that generate a static air cushion that permit hovering
above the surface at a zero speed condition.
Hydrofoil assisted take-off has been tried a number of
times but appears to have inherent limitations such as the
onset of cavitation on the upper surface, additional
mechanical complexity, and safety concerns at low
altitude and negative angles of attack, rendering the
technique less than ideal.
The majority of existing WIG designs utilise planing
surfaces to reach the speeds required to initiate
aerodynamic lift. Whilst this may be the simplest
arrangement to incorporate into a prospective design, it
has contributed to the difficulty of take-off in waves,
often exacerbated by unfavourable running trim (see also
section 3.4). The transitional phases of take-off and
landing are frequently categorised as the most difficult
part of WIG design, with sea state and environmental
conditions given as primary factors [27]; therefore
performance improvements in this area would be of
foremost importance for a new design concept.
2.4 SUMMARY OF WIG
There are clearly technical challenges that remain to be
resolved in order to utilise the aerodynamic forces
generated by very high speed vessel operation.
Longitudinal instability has often proved problematic in
this speed regime, but there remains a promise to harness
these forces to safely sustain the weight of the craft.
Historically, WIG have physically resembled low flying
aircraft, with their associated geometrical features. There
are several main families of geometric configurations,
although they have often ultimately proved unfavourable
in a marine environment.
The AAMV stands apart from these WIG craft, as it can
operate successfully in a number of modes: low speed
displacement and so-called semi-displacement
waterborne modes; a transient condition defined by
hydrodynamic and aerodynamic forces of the same order
of magnitude [2]; and finally completely free of the water
surface for high speed transit, fully airborne although not
beyond the influence of the ground effect zone.
3. AAMV DESIGN CONCEPTS
The goal of AAMV design is the complementary
combination of a suitable aerodynamic configuration
with hydrostatic and hydrodynamic support through a
lightweight and sufficiently strong structure to permit
safe and reliable operation. The specific aerodynamic
configurations being researched, consisting of aerofoil
profiles and potentially unique planforms tailored
especially for use in ground effect, are not covered in this
paper. Determination of the vessel structure, including
hulls, cross-deck structure and superstructure will be
outlined in the following section.
2015: The Royal Institution of Naval Architects
Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
Figure 4. Multi-Disciplinary Design iterative cycle
3.1 MULTI-DISCIPLINARY OPTIMISATION
In general, many existing WIG craft have been designed
considering the airborne phase primarily and therefore
the capability of such vehicles while in waterborne phase
are very limited: no waterborne cruise speed considered,
very high drag at low to medium waterborne speeds, very
limited manoeuvring capabilities.
According to well-known naval architecture design
practice, a holistic approach represented by the design
spiral of Figure 4 has been carried out in order to fully
integrate all mission requirements and design constraints
into a final AAMV configuration. The main idea was to
define the vehicle operational requirements, leading to
the identification of any necessary design constraints.
Having an external geometry defined (wing/s and hull/s),
it is possible to estimate the equilibrium attitude of the
AAMV configuration for a range of speeds, and then to
estimate the power-speed curve. It should be noted that
the equilibrium attitude approach adopted has been
developed specifically for AAMV configurations, so it
takes the instantaneous hydrostatic, hydrodynamic and
aerodynamic forces to calculate the attitude. It is based
on the long-form Savitsky method [5], modified to take
into account the influence of the aerodynamic lift, drag
and moment generated by the aerodynamic surfaces.
Since they are operating in ground effect, the effect of
the distance from the ground (as well as the angle of
attack) is considered in evaluating the aerodynamic
forces [3].
Then, based on the power and propulsion system type
(e.g. Turboprop), it is possible to choose from a database
of existing power and propulsion systems to select the
2015: The Royal Institution of Naval Architects
most suitable one. At this point it is possible to perform
the scantling design of the structure (section 3.3). The
previously mentioned steps are repeated iteratively, until
the convergence criteria based on the total estimated
mass is satisfied.
3.2 NUMERICAL MODELLING OF AERO-
HYDRODYNAMIC FORCE COUPLING
In 2006, a research program aiming at investigating
AAMV configurations was initiated at Cranfield
University, focusing on the estimation of the loads and
the vehicle dynamics during the high speed waterborne
phases, when hydrodynamic and aerodynamic forces of
the same order of magnitude can occur [2]. Among the
results, a concept configuration characterised by very
high lift-to-drag ratios has been defined [26], and a
model of dynamics specific for this configuration has
been developed, deriving a novel static stability criteria
[4].
The model of dynamics consists of two parts: a system of
equations of equilibrium, used to estimate the
equilibrium attitude of the AAMV configuration, and a
system of equations of motion, used to estimate the
dynamic response (modes of oscillation) of the vehicle.
Regarding the equilibrium model, the aerodynamic
forces are calculated estimating the angle of attack and
distance from the ground of the aerodynamic surface,
then interpolating the values of the provided
aerodynamic coefficient database to obtain the lift, drag
and pitching moment at that angle of attack and height
above surface.
Hydrostatic and hydrodynamic forces are obtained using
a Savitsky approach for a prismatic planing surface [21],
modified to take into account the influence of the
aerodynamic forces and moments on the equilibrium [3].
 Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
Due to the limitations of the Savitsky approach, it would
be difficult to estimate the hydrodynamic forces for the
present non-planing configuration, therefore it was
decided to determine the required coefficients
experimentally.
The net forces can be estimated by calculating the
alleviated displacement (equal to the vehicle
displacement minus the aerodynamic lift) and the trim
angle, and then interpolating between the measured
hydrodynamic resistance and moment coefficients for a
particular equilibrium state.
For a given hull, it is then necessary to build a database
of drag and moment coefficients, having as input the
alleviated displacement and the trim angle, in function of
the Froude number. This approach is similar to the one
adopted by NACA to experimentally test seaplane hulls
[18].
Figure 5. Design spiral for AAMV with emphasis on structure
3.3 STRUCTURAL DESIGN
As the structural weight constitutes the greatest part of
the overall vessel displacement, the following section
outlines the methods used to estimate appropriate
characteristics for this weight group.
3.3 (a) Hull Scantlings
The definition of High Speed Marine Vehicles is found
in the HSC Code adopted by IMO as: a craft capable of
maximum speed, in metres per second (m/s), equal to or
exceeding: 3.70.1667, where is the volume of
displacement corresponding to the design waterline (m3),
excluding craft the hull of which is supported completely
clear above the water surface in non-displacement mode
by aerodynamic forces generated by ground effect.
Such a definition is not entirely applicable to a
conventional WIG vehicle nor an AAMV. However, it
constitutes a reference for the naval architectural
performance of these vehicles.
In order to obtain a preliminary scantling design for the
hull structure, the HSC Code was consulted. Actually,
WIG guidelines were issued by IMO (MSC/Circ.1054)
but no prescriptive provisions are reported as far as
scantling design is concerned. Rather, the HSC Code
provides useful parametric formulations, suitable to
obtain a preliminary scantling design of the hull
structures and allowing estimation of the hull weight and
its distribution.
Scantling checks are based on first principles and
formulations can be easily adapted/modified as deemed
necessary. Hull plating and stiffeners were verified
adopting prescriptive formulations reported in HSC
Code.
Hull motions and loads are defined as per HSC Code.
Static, dynamic and impulsive loads are considered.
Even if WIG specific loads (e.g. landing impact) are not
defined, in lieu slamming loads are considered sufficient
for the intended aim.
In order to explore the design space of the structural
design, the ordinary frame spacing was initially selected
according to the RINA Rules for pleasure craft:
s_r = 1.2*(0.35+0.005*L) (5)
where s_r is the ordinary frame spacing and L the hull
length.
A MATLAB routine was developed based on the HSC
rules. The main data needed to start the calculations with
the HSC approach are:
Main dimensions of the vehicle (Length, Beam,
Draft, Displacement etc)
Service type (Passenger/Ferry/Cargo, Supply,
Pilot/Patrol, Rescue as per applied rules)
Environmental conditions (sea state)
Construction material
The main outputs are:
Structure system total weight
Structure system centre of gravity position
A database of stiffeners, usually used for standard
constructions, was implemented in the code, the
drawback of using this database is that the estimated
structure could be heavier compared with the one
designed that makes use of bespoke members. To tackle
this heavier weight coming from the use of the above
database, an optimization based on the stiffeners spacing
was implemented in the code. The stiffeners spacing is
changed within an interval of 20% of the reference
spacing given by the RINA Rules and the spacing
resulting in the lightest hull structure is then selected.
In order to account for the use of composite materials,
useful guidance was found through researching other
high speed marine vehicles. In particular, information
was found in the report published by the LASS project
2015: The Royal Institution of Naval Architects
Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
[23] about the influence of lightweight design using both
composite and aluminium materials.
The AAMV scantlings are determined assuming that a
metallic alloy is used (aluminium or steel), for which the
MATLAB code has been properly developed and
validated. Then, depending on the selected composite
material, the total structural weight is reduced by a
certain percentage depending on composite
reinforcement type (glass or carbon fibre).
3.3 (b) Wing Scantlings
Three different approaches were followed and compared
for the scantling design of the wing.
Firstly, and in a manner similar to the approach used for
the hull/fuselage, Roskam [19] presents formulae to
estimate the structural weight of the wing. It is worth
remembering that the wing configuration considered for
the AAMV is quite uncommon (e.g. aspect ratio is close
to one, sweep angle is zero), as well as the load
distribution being quite different. Nevertheless, the
values estimated using the Roskam approach are close to
those derived by the direct approach illustrated in the
following paragraph.
In the next approach, the design of the wing structure is
performed using a direct method, i.e. the size of each
structural member (plating, stiffener, primary member) is
calculated depending on the loads acting on the wing.
The spars are designed taking into account the global
transverse bending moment, while the other elements
have been checked by direct calculations based on beam
theory taking into account local loads. It is worth
highlighting that no loads coming from water impact on
the bottom of the wing were considered.
Averaging the weight obtained by the Roskam approach
and the direct approach was considered a good
compromise.
The third approach is based on the average hull deck
structural weight per unit of area, and so it is available
only if the HSC code approach was applied for the hull
weight estimation. The deck weight per unit area is
multiplied by the wing area to obtain the wings
structural weight estimation. As expected, this approach
gives very high values as the deck is verified considering
higher acting loads.
2015: The Royal Institution of Naval Architects
Figure 6. Representative structure in way of the main wing
3.3 (c) Structural Summary
The baseline configuration initially had both the hull and
the wing made of fibre-reinforced plastic (FRP). Some
preliminary assumptions about this material were made,
as described below.
Often FRP manufacturing processes are more expensive
than those for aluminium and steel. FRP may be cost-
effective for mass production, as the cost of the required
moulds can be spread over a number of vehicles.
However, the FRP manufacturing process leads to a
product with higher margins of uncertainty (its
mechanical features depend greatly on the direction in
which it is stressed), and because of that during the
scantling process some safety margins have to be
considered, leading to a heavier vehicle. Within the
structural scantling code, it was not possible to take into
account these aspects, and because of that the resulting
weight calculation is less accurate in comparison to the
weight calculation for aluminium and steel.
The total mass of the vehicle is strongly influenced by
the choice of the hull material. Steel has been considered
only for comparison and completeness, but in general
this material is considered not suitable for an AAMV
application.
As expected, a heavier configuration requires a larger
wing area in order to generate the lift necessary with the
same take-off speed. Since the length of the vehicle (and
therefore the chord of the wing) does not change, the
wing span increases if using aluminium, even more so
for the steel configuration. If a maximum AAMV width
were imposed, it may have a substantial impact on the
choice of the material for the hulls.
Similarly, the draught of the vehicle at rest follows an
expected trend. The draught for a steel configuration and
an aluminium configuration is typically 35% and 15%
higher than the draught for a FRP configuration.
The best trade-off material seems to be aluminium, also
considering its better performance against corrosion and
its ability to withstand impact loads, which are supposed
 Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
to be quite frequent especially during take-off and
landing.
These results are well expected, and qualitatively it
would have been possible to predict them without
developing a dedicated spiral design approach.
Nonetheless, the important contribution of the present
method is that it allows conducting sensitivity analyses
and estimating quantitatively the impact of each
parameter on the overall design; steps that would not
have been possible without the development and
implementation of this approach.
3.4 HYDRODYNAMIC SURFACES
The attainment of sufficiently high speeds for waterborne
take-off has typically been achieved through the use of
recognisably standard hard-chine, hydrodynamically
planing hullforms faired into the underside of the main
fuselage. There are several significant disadvantages of
this method, briefly described as follows.
The overall dimensions of the planing surface
must conform to the vessel geometry, most
notably the longitudinal aspect ratio (length over
breadth) must lie within the constraints of the
fuselage dimensions, which may be sized
according to other criteria of the design
envelope, e.g. minimisation of aerodynamic
parasitic drag.
The longitudinal centre of mass (LCG) must be
precisely located for stable WIG operation,
although that position is typically much further
forward than the location most suitable for
planing craft. This could result in excessively
high hydrodynamic resistance as the vessel fails
to adopt the attitude required to achieve planing
efficiencies realised by marine-only vessels.
Hydrodynamic resistance of beamier geometries
can be very high prior to reaching the planing
condition, often requiring higher powering
configurations to be installed than would be
necessary during an efficient cruise condition or
any other point of the operational profile. This
power is needed to overcome the so-called
hump drag.
Extreme accelerations resulting from slamming
and wave impact of the planing hull has been
well documented [7]. This has led to structural
failures, increase of scantling and structural
weights, debilitating injuries to personnel and
chronic fatigue to operators and crew. Increase
of deadrise is the standard method of reducing
these slam-induced pressures, although at the
cost of reduced efficiency.
The planing hullform is also very inefficient at
lower speeds of the displacement and semi-
displacement regimes, with resultant powering
requirements far in excess of specifically
designed vessels operating at similar speeds.
In addition to resistance penalties, a coupled
longitudinal-transverse instability phenomenon has been
reported for planing hullforms with unusually far forward
centres of gravity. The vessel may attain stable
equilibrium on-plane in a normal manner, before
experiencing a sudden decrease in the running trim angle,
whereby the fullness of the bow sections tend to initiate a
roll and/or yaw response. This is not presently a well
understood problem [22].
These factors indicate an alternative hullform would be
desirable, one that could accommodate a centre of
gravity further forward than typically found, but also one
that allows that vessel to achieve the required waterborne
speeds.
4. HULLFORM VALIDATION
As the numerical modelling routine requires
hydrodynamic force and moment coefficients for
coupling to the aerodynamic surfaces, an alternative hull
design and test programme was carried out to
experimentally determine these coefficients for the new
hullform.
4.1 HULL DESIGN
Due to the issues set out in section 3.4, a design brief
with the following characteristics was followed:
1. Must be able to carry a centre of mass (and
therefore centre of buoyancy) around amidships,
defined as LCG/LOA = 0.5.
2. Must be suitably shaped to allow high speed
operation without excessive powering or
undesirable behavioural characteristics, e.g.
submarining, yaw instability, excessive
slamming, etc.
3. Would ideally offer low speed resistance, as
defined by FnL < 1.0, similar to conventional
efficient semi-displacement hullforms.
Additionally, the hull would have to retain its desirable
characteristics at various draughts, as the vessel will
experience vertical translation (heave) as the
aerodynamic lift forces generated by the wing increase.
In contrast to conventional planing hulls, where it is
common practice to maintain the longitudinal sections
straight and flat throughout the after part, significant
rocker was introduced along the entire length of the keel.
It was thought that a bow up trimming moment caused
2015: The Royal Institution of Naval Architects
Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
by hydrodynamic pressures aft may well be
advantageous aerodynamically to increase angle of attack
on the wing and initiate the weight unloading process,
albeit at the risk of inducing downward hydrodynamic
suction force.
Once a trim angle of approximately 3 is reached with
the bow clear of the water, the buttock lines become
effectively flat. Instead of the deeply veed forward
sections of conventional planing hulls, the enclosed
volumes are inverted, i.e. somewhat flat-bottomed with
finer angles topside on the reverse stem, with cutaway
forefoot and foredeck in profile. Figure 7 shows a
representative lines plan for a hull with these
characteristics.
Figure 7. Representative lines plan
The LCB was situated at 0.5 LWL for the static design
draft condition, with the transverse sections configured
so this centroid moves forward gradually as the vessels
rises bodily. As the location of the centre of mass is not
expected to shift significantly during take-off or landing
phases, this misalignment of longitudinal centres should
increase the bow-up trimming moment, likely to be
favourable for the aerodynamic configuration.
Design ratios such as length over volume and length to
beam ratio were selected to be representative of suitable
values, without particular optimisation for a specific
design or other criteria.
4.2 EXPERIMENTAL TESTING
A model of 3 m LOA was constructed in the Ocean
Laboratory workshop at Cranfield University, of glass
reinforced plastic (GRP) over a CNC milled foam core;
hand laid, vacuum bagged, and resin infused. The
laminate schedule consisted of 2-4 plies (dependent on
location) of 300 gsm E-glass in a 2/2 twill weave.
Before lamination, a mounting plate was recessed into
the deck amidships, bonded in and glassed over. The
surface was finished to an approximate average
roughness of 0.4 micron (240 grit sandpaper) and
finished with a high gloss enamel coating.
For the initial tests it was decided to use a fully
constrained towed rig, with the model positioned
precisely to replicate specific displacement and trim
combinations. The resulting forces and moments
imposed on the model were measured to a calibrated
accuracy of 0.75 N and 0.05 Nm respectively, using a
six-axis force/torque transducer. Test runs were carried
out in calm water at carriage speeds ranging from 1 m/s
to 12 m/s.
2015: The Royal Institution of Naval Architects
The testing programme matrix included four
displacement conditions (with corresponding heights
above datum) and three trim angles: 0, 3, 6 bow up.
Multiple speed runs were carried out to assess the
conditions most likely to be seen by each of the
combinations. Ten unique configurations were tested in
total. The model was designed to be suitable for scale
factors between 5 and 10. Depending on the scale
selected, the corresponding maximum full-size speeds
tested would be between 52 kt and 74 kt.
For this paper, the results of the trimmed conditions have
no particular importance as these would not be natural
angles developed without the interaction of the
aerodynamic wing. The actual attitude, or running trim,
of the completed vehicle would be a function of wing
size, orientation and location, in addition to the
hydrodynamic components.
The tests were undertaken in April 2014 in the towing
tank in Qinetiq-Haslar, as this facility features one of the
highest towing carriage speeds currently available. The
tank dimensions are 271 m long, 12.2 m wide and 5.5 m
deep. The sampling frequency of 100 Hz allowed
sufficient data to be collected at the maximum towing
velocity and permissible acceleration/deceleration.
Prediction of total free body resistance for the model
configuration in the tested speed regimes was estimated
prior to testing to be used as a form of comparative
benchmark. The estimate was based upon slender body
theory, characterised as a first principles potential flow
approach to predict the far-field wave pattern. The
contribution of viscous effects was estimated using a
form factor multiplier on the frictional resistance of the
static wetted surface area, as documented in the ITTC
1957 method. Although this form factor can be shown to
vary with increasing speed, careful selection of a single
average value for the entire range was expected to have a
maximum error of +/- 5% [17].
Figure 8. Preliminary results of the even keel performance at
four displacements
 Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
4.3 DISCUSSION
Preliminary results of the level trim resistance tests is
shown in figure 8, along with the nominal prediction line
for the full displacement condition. The model appears
to perform slightly better than expected at lower speeds
before a rapid drag increase around 9 m/s, however as the
bodily sinkage was fully constrained it is likely a free
body test would demonstrate a heavier effective
displacement and higher drag values.
There was no attempt to install spray rails or any other
method for the reduction of side wetting. Images and
video from the tests show considerable rise of the bow
wave up the topsides at many of the higher speeds. The
control of the side wetting in future test would be
expected to reduce skin friction drag and provide
additional lift.
The curves represent the four tested displacement
conditions, in order to represent stages of weight
unloading experienced during a take-off procedure. The
lower part of the speed range is where the vessel would
be entirely hydrostatically and then hydrodynamically
supported. As forward speed increases, aerodynamic
alleviation of the vessel weight would occur, effectively
reducing the displacement.
The upper part of the speed range would correspond to
conditions where aerodynamic lift is significant, with the
result that the predicted drag would effectively skip to
the next curve, representative of the newly unloaded
condition. For a given power, the ongoing lowering of
hydrodynamic resistance should allow the craft to
accelerate rapidly, further generating aerodynamic lift
until water contact diminishes entirely at lift-off.
As the vessel was constrained in pitch and heave, the
results cannot be taken as a definitive representation of
the vessel free body performance, but would form the
basis of a future testing programme.
5. CASE STUDY
5.1 REQUIREMENTS
To illustrate an application of the preliminary design
methodology, a representative AAMV configuration is
presented. The vehicle has been designed to satisfy the
following main requirements:
Payload: 2000 kg
Nominal range: 500 nm (926 km)
Airborne cruise speed: 100 kt (185 km/h)
The payload is representative of 12 technicians with
relevant tools, plus three crew members. It can also be
representative of fewer technicians with an additional
weight of equipment.
If the Round 3 Dogger Bank development is considered,
the distance from shore varies from 125 to 290 km,
allowing for a maintenance trip to the site (back and
forth), with around 300km available for intra-wind farm
operations.
It will be possible to reach the closest wind turbines in
around an hour considering the taxi time to exit from
port, the take-off phase, the cruise phase at 100 kt, the
landing phase and the taxi phase to reach the wind
turbine; or the furthest one in around two hours, that is
about 20-30% of the time needed by the current available
vessels.
5.2 OUTPUT CONFIGURATION
A MATLAB implementation of the illustrated
methodology has been utilised for preliminary sizing of
an AAMV configuration, considering the aforementioned
requirements. The resultant vehicle, with a general
arrangement similar to the one illustrated in Figure 9,
consists of two hulls, a central wing structure connecting
the two hulls, representing the main lifting surface of the
vehicle, and a high tail operating out of ground effect to
ensure longitudinal static stability. The power and
propulsion system is represented by two turbo-propellers.
Reminding the reader that the AAMV configuration is
quite novel, and therefore there is a lack of experimental
data, the estimates presented have to be considered as
conceptual design level estimates, with a sensible range
being 25% around them.
The estimated total displacement is around 10.2 t, and the
total power installed is around 1.3 MW. The length
overall of the vehicle is around 15 m, and a total width of
10 m. The draught, at rest, is less than 1 m. The weight
and the power installed are comparable to the WIG
vehicle HoverWing HW20 [10], with its displacement
also around 10 t and power installed around 1.1 MW,
designed to transport a similar payload.
Figure 9. AAMV configuration considered in the case study
2015: The Royal Institution of Naval Architects
Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK
6. CONCLUSIONS
Current options for personnel transfer to offshore wind
farms are generally limited to either relatively slow and
often uncomfortable crew boats, or expensive and often
dangerous helicopter transit. This paper has proposed a
third option, one that combines the speed and comfort of
an aircraft above the water surface, with the waterborne
performance and operational costs associated with high
speed marine vehicles.
The present work focused on specific considerations in
the design of a suitable hullform for an AAMV
configuration. In many current and past WIG craft,
aerodynamic aspects have governed the design, centred
on the airborne cruise condition, leading to poor
resistance and manoeuvrability characteristics during
waterborne operation. A true AAMV needs to be fully
operational as a marine vehicle, something that existing
aero-derivative designs have not been able to achieve.
An experimental hullform has been developed and
initially tested for suitability in this challenging
environment. A thorough testing programme would be
required, including both calm water and in waves, before
definite conclusions could be presented about the
suitability of this hull shape.
Structural weight has been estimated using various
methodologies to determine their suitability for usage in
this application. A representative case study has been
shown, with a conceptual rendering of a proposed
arrangement.
7. ACKNOWLEDGEMENTS
The authors would like to thank the staff at the test
facility in Qinetiq-Haslar for their help and advice before
and during the test campaign.
8. REFERENCES
1. BVG ASSOCIATES, Towards Round 3:
Building the Offshore Wind Supply Chain, The
Crown Estate, 2009.
2. COLLU, M., Dynamics of Marine Vehicles
with Aerodynamic Surfaces. PhD Thesis,
Cranfield University, 2008.
3. COLLU, M., PATEL, M.H. & TRARIEUX, F.,
A Unified Mathematical Model for High Speed
Hybrid (Air and Water-borne) Vehicles, 2nd
International Conference on Marine Research
and Transportation (ICMRT07). Ischia, Naples,
Italy, pp. 8998, 2007.
4. COLLU, M., PATEL, M.H. & TRARIEUX, F.,
The longitudinal static stability of an
aerodynamically alleviated marine vehicle, a
mathematical model. Proceedings of the Royal
Society A: Mathematical, Physical and
2015: The Royal Institution of Naval Architects
Engineering Sciences, 466(2116), pp.1055
1075, 2009.
5. DOCTORS, L., Hydrodynamics of High-Speed
Small Craft, University of Michigan, pp. 137
205,269,270, 1985.
6. DOCTORS, L., Analysis of the efficiency of an
ekranocat: a very high speed catamaran with
aerodynamic alleviation. Int. Conf. on Wing in
Ground Effect Craft, London, UK, 1997.
7. FALTINSEN, O.M., Hydrodynamics of High-
Speed Marine Vehicles. Cambridge University
Press, 2005.
8. HALLORAN, M. & O'MEARA, S., Wing in
Ground Effect Craft Review, DSTO
Aeronautical and Maritime Research
Laboratory, Report DSTO-GD-0201, 1999.
9. IRODOV, R.D., Criteria of the longitudinal
stability of the ekranoplan, Ohio, 1974.
10. http://hoverwing.webs.com/hoverwing20.htm
11. KOMISSAROV, S. & GORDON, Y., Soviet
and Russian Ekranoplans, Midland Publishing,
2010.
12. KOROLYOV, V.I., Longitudinal Stability of
Ekranoplans and Hydrofoil Ships, RTO-AVT
Symposium on Fluid Dynamics Problems of
Vehicles Operating Near or In the Air-Sea
Interface, Amsterdam, 5-8 October 1998.
13. KUMAR, P.E., Stability of Ground Effect
Wings, Cranfield College of Aeronautics,
Report Aero No. 196, 1967.
14. KOLYZAEV B., ZHUKOV V. & MASKALIK
A., Ekranoplan, Peculiarity of the Theory and
Design, Saint Petersburg Sudostroyyeniye,
2000.
15. MATVEEV, K.I. & DUBROVSKY, V.,
Aerodynamic characteristics of a hybrid
trimaran model, Ocean Engineering, 34(3-4),
pp.616620, 2007.
16. MCKESSON, C.B., The Practical Design of
Advanced Marine Vehicles, University of New
Orleans, 2009.
17. MOLLAND, A.F., WELLICOME, J.F. &
COUSER, P.R., Resistance Experiments on a
Systematic Series of High Speed Displacement
Catamaran Forms: Variation of Length-
Displacement Ratio and Breadth-Draught
Ratio. Ship Science Report No. 71, University
of Southampton, 1994.
18. PARKINSON, J. B. & HOUSE, R. O.,
Hydrodynamic and Aerodynamic Tests of
 Design & Operation of Wind Farm Support Vessels, 28-29 January 2015, London, UK

Models of Floats for Single Float Seaplanes:
NACA Models 41-D, 41-E, 61-A, 73, and 73-
A, National Adivsory Committee for
Aeronautics, Technical Note No. 656, 1938.
19. ROSKAM, J., Airplane Design Part I:
Preliminary Sizing of Airplanes, Design
Analysis & Research, 1989.
led an industry funded project on the conceptual design
of an AAMV. He has also patented a novel trim control
mechanism for high speed marine vehicles. He is a
chartered engineer and member of the Royal Institution
of Naval Architects.

20. ROZHDESTVENSKY, K. V., Wing-in-ground

effect vehicles, Progress in Aerospace
Sciences, 42(3), pp.211283, 2006.

21. SAVITSKY, D., Hydrodynamic design of

planing hulls, Marine Technology, 1(1), pp.71
95, 1964.

22. SAVITSKY, D., Planing Craft, Modern Ships

and Craft. Naval Engineers Journal, Vol. 97,
Issue 2, American Society of Naval Engineers,
1985.

23. http://s-lass.com/en/Sidor/default.aspx

24. STAUFENBIEL, R., Longitudinal motion of

low-flying vehicles in nonlinear flowfields,
Proceedings of the Congress of the
International Council of the Aeronautical
Sciences. Munich, pp. 293308, 1980.

25. WIESELSBERGER, C. Wing Resistance Near

the Ground, NACA TM-77, 1922

26. WILLIAMS, A.G.W., COLLU, M. & PATEL,

M.H., Aerodynamic lift forces on multihulled
marine vehicles, International Journal of
Maritime Engineering, 152, pp.4150, 2010.

27. YUN, L., BLIAULT, A. & DOO, J., WIG Craft

and Ekranoplan, Springer, 2010.

9. AUTHORS BIOGRAPHY

Daniel James is the principal naval architect at the

Ocean Systems Test Laboratory at Cranfield University.
He is responsible for marine design/engineering services.
His previous experience includes several world-leading
high speed marine vessel designers and builders. He
holds a Professional Engineer license from the state of
Connecticut, USA.

Maurizio Collu is a lecturer in Offshore Renewable

Energy Engineering at Cranfield University. His two
main areas of research focus on the conceptual and
preliminary design of offshore support structures for the
emerging floating wind turbine industry, and of
aerodynamically alleviated marine vehicles (AAMV),
specialising in dynamics and stability. He has been
awarded the prestigious RINA Calder Prize, has
published in the Royal Society Proc A journal, and has

2015: The Royal Institution of Naval Architects