1

Lightning Protection of Wind Turbines A

Comparison Of Real Lightning Strike Data
And Finite Element Lightning Attachment
Analysis
Vidyadhar Peesapati and Ian Cotton, Senior Member, IEEE

there is also uncertainty regarding the likely point of lightning

Abstract--Wind turbine lightning protection systems have been attachment. Current turbine lightning protection systems have
developed to the point where lightning damage is relatively rare. been shown reasonably effective for downward lightning but
However, with windfarms moving offshore, manufacturers are there is an underlying risk in assuming the same systems will
striving to continuously improve lightning protection systems work for upward lightning.
while ensuring that they comply with relevant IEC standards. This paper aims to examine the likely points on a wind turbine
The case of offshore wind farms is particularly important due to
the difficulties faced in accessing a wind turbine should this be
structure where upward propagating lightning may form. It
required owing to lightning damage. The paper details work also compares these results with real lightning strike data, to
done to model upward propagating lightning strike attachment see if the results correlate. In the introduction, a background of
on a wind turbine. A 3D electrostatic model of a full scale wind the problems wind turbines face from lightning is given and
turbine has been modelled using available Finite Element the phenomenology of lightning is more fully described. Later
Analysis software. This full scale model is subjected to high sections introduce the finite element models used in this work
electric fields comparable to those created by a charged cloud. and the corresponding results.
Results from these simulations are then compared with those
found from analysis of real lightning strike data taken from wind
II. CHARACTERISTICS OF LIGHTNING
turbines and windfarms across the world.

Index Terms--Lightning, wind turbine, finite element, Lightning is an atmospheric discharge of current. The highest
modeling, electric field. recorded value of lightning current is around 250kA [1][2].
However, this value is very rarely seen, the median (for a
I. INTRODUCTION downward negative stroke) being about 30kA with the median
values of charge transfer and specific energy being 5.2C and
W ind turbines being developed offshore are increasing in
size and this trend is not likely to abate in the near
future. The lightning strike frequency to onshore wind turbines
55kJ/ respectively [1][2]. The visible part of the lightning
strike process, whether lightning strikes the ground or not, is
termed as a lightning flash. The individual components of
is usually calculated based on the likelihood of downward this lightning flash are defined as strokes. Lightning can be
lightning attachment. However, taller wind turbines such as classified into two main types with respect to wind turbines,
those being used offshore will be increasingly subject to upward and downward initiated. These are also known by the
upward lightning attachment. This form of a lightning (to be names, cloud-to-ground and ground-to-cloud lightning. These
discussed in more detail in a later section) is triggered by the two forms of lightning can be further subdivided into positive
wind turbine itself in the presence of a thundercloud and has and negative polarity respectively, the polarity being that of
the ability to significantly increase the number of lightning
the charge transferred from the cloud to the ground.
strikes to a wind turbine per year. As the attachment processes
are fundamentally different to those of downward lightning, A. Downward Initiated Lightning
Downward initiated lightning starts from the cloud with a
stepped leader moving towards the earth. The end of the
leader, the leader tip, is in excess of 10MV with respect to the
This work has been carried out under the Supergen V Wind Energy
Theme, funded by the Engineering and Physical Sciences Research Council, earth [3]. As the tip descends, it raises the electric field
UK. Further information about Supergen Wind is available at strength at the surface of the earth. Where this field is elevated
http://www.supergen-wind.org.uk/ significantly, typically around sharp and/or tall objects, local
V Peesapati is with the Department of Electrical and Electronics ionization of the air takes place and answering leaders are
Engineering, Power Systems Group, University of Manchester(e-mail:
[email protected]) emitted and travel towards the downward propagating leader.
I Cotton is with the Department of Electrical and Electronics Engineering, When an answering leader and stepped leader meet, this
Power Systems Group, University of Manchester completes the channel or path from the cloud to earth, thus
(e-mail:[email protected]).
 2

allowing the charge in the cloud to travel through the ionised An upward initiated discharge often starts with a continuing
channel. The location from which these answering leaders current on which may be superimposed short duration high
form is critical in determining which point of the wind turbine magnitude current pulses. Although the current values are
will be hit by lightning. The first transfer of significant current quite low at around 10kA [3] as compared to downward
is the first return stroke, which has a peak value of upto a few lightning, the charge transfer associated with the continuing
hundred kiloamps and a typical duration of a few hundred current phase can be quite high. The initial continuing current
microseconds. After a certain time interval, further strokes may be followed by a number of return strokes that are similar
may follow the already ionised path, these are known as to those observed in a negative downward lightning flash.
subsequent return strokes (Fig 1). On average, a negative
downward lighting flash may contain 2 to 3 subsequent return III. LIGHTNING AND WIND TURBINES
strokes. Positive downward flashes (only 10% of those
observed worldwide) are higher in magnitude but typically The high peak currents carried by lightning strikes are a
contain no subsequent strokes.
source of significant energy. If the wind turbine lightning
protection system does not divert this lightning current safely
to ground through a low impedance path, significant damage
can result. If a component of a wind turbine is damaged there
are two things that need to be considered; the resulting repair
costs and the associated loss in production caused due to
downtime. Given the move of wind turbines to offshore
locations and the need to mobilise special floating cranes,
these costs and repair times will only have increased in recent
years. Due to the risk of damage and possible downtime due to
lightning strikes, all new wind turbines are normally equipped
with lightning protection systems. Guidance on how to
Fig. 1. Profile Downward Initiated Lightning
achieve this is given in [4].

B. Upward Initiated Lightning The main functions of the lightning protection installed on a
The presence of tall structures and objects brings rise to wind turbine are:
another form of lightning, which is upward initiated. Tall
structures enhance the electric field produced by a Successful attachment/formation of the lightning
thundercloud and this can give rise to upward propagating strike to a preferred attachment point such as the air
leaders that move towards the cloud and which can then termination system on the blade
develop into a lightning flash. The attachment process is Facilitating the passage of the lightning current
therefore somewhat different to downward propagating through the system into the earth without causing
lightning and this issue will be dealt with in a later section. damage to systems including that damage that would
This phenomenon is particularly common where the cloud result from high levels of electric and magnetic field
height is quite low (often during winter months in coastal Minimising the levels of voltages and voltage
areas or in mountainous regions). The profile of an upward gradients observed in and around the wind turbine
discharge is different as compared to that of a downward
initiated discharge (Fig 2). These functions can be achieved by a number of different
methods which have changed and improved with the
development of new wind turbines. As they are the highest
part of the turbine and therefore particularly prone to lightning
attachment, the blades are usually considered an extremely
important part of the wind turbine in terms of lightning
protection. The different types of lightning protection installed
in wind turbines blades are [3]:

Air termination systems on the blade surfaces

High resistive tapes and diverters
Down conductors placed inside the blade
Conducting materials for the blade surface
Fig. 2. Profile Upward Initiated Lightning
 3

Fig. 3. Lightning Protection Methods for Rotor Blades

In all types of LPS, the metallic air terminations, strips and consists of two antennas which are placed on opposite sides of
diverters and down conductors should be of sufficient cross the tower [10] [11]. In the event of a lightning strike to the
sections that they safely conduct the lightning current without wind turbine, the lightning current flow induces a voltage into
any physical damage. the antennae system as a result of the rate of change of
magnetic field around the tower. Lightning strikes away from
A system that is widely used is the internal lightning the wind turbine are discriminated against by adding the
protection system consisting of a internal lightning down outputs of the antennae together. The Jomitek system has a
conductor capable of carrying the lightning current. Metal simple analog trigger level which at the offshore wind farm is
receptors (Fig 3) which act as air terminations penetrate the set to 1 kA peak. The output of this measurement system is
blade surface and are then connected to the down conductor.
connected to the turbine SCADA system. In the event of the
This system of external receptors connected to an internal
system being triggered, an alarm is raised but no data
down conductor is being widely used for blades upto 60m [3]
regarding the possible magnitude of the lightning current is
and the system is not likely to change for blades of larger
lengths. obtained. For this reason, peak current sensor (PCS) cards
manufactured by OBO Bettermann are placed on the down-
According to [8], rotor blades with built in lightning conductor of each of the blades and of the air terminals
protection are far less likely to experience extensive damage protecting the aviation lights and the wind-vane. These cards
as compared to those without. Current lightning protection have a magnetic strip imprinted with a pre-defined signal.
systems for rotor blades are designed to withstand 98% of When placed near a down conductor, the magnetic field
lightning strikes [5] but there is still a risk of damage, resulting from the flow of lightning current erases a portion of
particularly at the attachment point. More information this magnetic strip and by the use of a card reader, the current
regarding the types of lightning damage that have been that the card has observed can be found. Once the Jomitek
observed in wind turbines and the types lab tests that can be system has been triggered, it is possible to go to the wind
used to verify/understand lightning protection performance is turbine to remove the PCS cards and therefore gain knowledge
available in [4] [5] [9]. of the lightning strike magnitude along with the component
struck. The PCS cards are capable of only recording one peak
IV. LIGHTNING DATA SOURCES current (the highest observed). During multiple strikes it is not
possible to determine the number of strikes. Also the PCS
cards have a range of 3-120 kA. Peak currents lower than 3 kA
The lightning data used in this paper comes from an Offshore
are not registered and peak currents above 120 will register
Windfarm. The data has been collected over a period of over
120kA.
three years, the oldest lightning strike record being in June
2003 with the last reading analysed being from in October
The data from the offshore windfarm is firstly examined by
2006. This is an equivalent of 216 turbine years.
showing the components of the wind turbine on which the
PCS cards indicated current readings following an alarm.
Data at this windfarm is collected in two ways. The first Multiple registrations were present on a number of occasions.
measurement system is the active Jomitek system. This The data in Fig 4 shows that the wind vane and aviation lights
 4

PCS cards (actually fitted onto the lightning protection rods of the peak current read from the each PCS cards due to an event
these components) indicated lightning current flowing through detected by the active alarm system by date. In total, 33 out of
these components regularly. This result is in contrast to the 51 lightning events occurred during the winter months. This
work by a number of other researchers who suggest that the shows that winter lightning is a significant phenomena even in
blades are the parts of a wind turbine that are most likely to be the country where the windfarm is installed, which is not
hit by lightning [3] [5] [8] [12] [13]. However as most wind particularly prone to winter lightning.
turbines are not equipped with PCS cards and those which are
have them mainly fitted in the blades, there are for most wind In other areas prone to winter lightning, the following has
turbines simply no effective registration of lightning striking been shown through observations of lightning activity to
elsewhere but the blades. From Fig 4, there is likelihood of towers:
increased lightning strikes to the windvane. The increase in The incidence of lightning strikes to tall structures
the number of strikes to the windvane could be explained by during winter is greater relative to that during
an increase in upward lightning (as will be demonstrated by summer [15].
the attachment modelling in the next section). The hypothesis
of upward lightning being a major contributor to the lightning
strike rate is strengthened when the distribution of lightning
strikes by month is examined. The data shown in Fig. 5 gives

Fig. 4. Number of PCS Cards Showing a Lightning Current Reading Following an Alarm Even

Fig. 5. Lightning Strikes Given By Their Peak Current versus Season of Occurrence
A large percentage of these lightning strikes, up to significantly tall structure in any season is likely to
98% are upward propagating (it is noted that any receive mainly upward propagating lightning). Even
 5

on flat terrain and for objects of moderate height, area of interest within the model are accurate. The model is
there is a large number of upward strikes [15][16]. shown in Fig 6.
Winter lightning simultaneously strikes more than
one tower within a lightning flash. The probability of
occurrence of these simultaneous strokes to multiple
points was up to 20% in the studies detailed in [16].

The data that has been presented matches with the findings of
the first bullet point, i.e. more lightning in winter than
summer. The second bullet deals with the number of upward
lightning strikes.

V. FEA MODELLING OF A CLOUD AND WIND TURBINE

FEA Modelling of lightning attachment to wind turbines has

been carried out by other researchers, [14] [17] [18] being
examples of some of the work carried out. However, with few
exceptions, most of these models were focused on downward
propagating lightning and therefore involved the modelling of Fig. 6. Cloud Model (charge sizes not to scale)
a stepped leader to examine the resultant electric field
environment on/around the wind turbine. One example where Fig 7 gives the vertical component of the electric field at the
upward propagating lightning is considered is [14] where the ground due to this vertical tri-pole. This plot agrees with that
authors have looked specifically at upward propagating shown in literature [15] and has a maximum ambient field of
lightning in terms of comparing the likely lightning just over 5kV/m (the E component in the graph must be scaled
attachment points when the blades are in motion and when by a factor of 1000 owing to the operation of the model).
they are stopped. The authors analyze the electric field According to the convention adopted in electrostatic field
enhancement for different angular positions of the blades analysis software, a downward directed field is taken to be
when they are stationary. It was concluded in this paper that positive in polarity [15].
the wind turbine was prone to higher field enhancement and B. The Wind Turbine and Blade Model
that leaders would form in more locations when the blades are
in motion. It was advised that the halting of the blade The model used in the simulations is based on a 2MW present
movement therefore reduces the risk of lightning strike day wind turbine. The blade model used in the simulations
attachment. presented is 40 meters long and is a simplification of a real
blade in that it is taken to be made entirely of fibre glass with
From the lightning strike data that has been analysed in the a relative permittivity of 4. The chord length of the airfoil used
previous section, it can be seen that upward initiated lightning for designing the blade is 2.75m, the chamber length (upper +
might contribute to a large portion of the lightning strikes that lower) is 0.9m, and a thickness of 10cm is maintained
attach to a wind turbine. Thus while evaluating likely throughout the blade. Due to the complicated shape and
attachment points from a wind turbine in this paper, only difficulty of meshing, these dimensions are maintained along
upward lightning is taken into consideration. In comparison to most of the length of the model, although several aerofoil
previous wind turbine models [14] [17], the model used profiles are normally used in real blades. The chord length
includes the windvane at the back of the nacelle and the risk of decreases from 3.5 meters at the blade root to a thickness of
lightning attachment to the same is analysed along side with less than 5mm at the blade tip.
the blades.
Integrated into the blade design is the lightning protection
A. The Cloud Model system, in this case an internal downconductor with tip
The models that are usually used to replicate the electric fields receptors. The tip receptor is placed 2.5 meters from the tip of
found under a thundercloud consist of a vertical tri-pole made the blade (37.5 meters from the root of the blade) and the
from two positive charges and a negative charge. A summary remaining two receptors are placed evenly along the length of
of various models that have been produced can be found in the blade. The tip receptor is usually a lot closer to the tip of
[15]. The cloud model that has been simulated here is taken the blade, but due to meshing difficulties in the FEA software,
the position has been changed further down the blade. The
from this source and consists of three point charges suspended
receptors are 10mm in diameter and are connected to a down
in air at different heights from the ground. In the simulation,
conductor that runs inside the hollow blade. The diameter of
the ground has been modeled as a perfect conductor. The three
the down conductor is set to 10mm. The down conductor and
charges of 3C, -40C and +40C are placed at heights of 2, 7 and the receptors are made of copper.
12 km from the ground respectively. The charges are modeled
as spheres of radii 900m for the 40C charges and 150m for the
3C charge. The size of the spheres is picked in such a way that
when the meshing process takes place, the electric fields in the
 6

Fig. 7. Electric Field (Ey) Plot above Ground

The wind turbine tower and nacelle are conductive and are set
to ground potential. The height from the bottom of the tower
Fig. 8. Wind Turbine Model
to the centre of the hub is approximately 70m (Fig 8). The
nacelle is 6m long, 4.5m wide and 6m high. The wind
instruments are placed at the rear of the nacelle. Wind Condition (i): The inception of a streamer discharge at the tip
instruments on the windvane are protected from lightning of the object by the creation of an electric field in the order of
using lighting rods. The lightning rods protecting the wind 3MV/m
instrument end in a hemispherical tip with a radius of
curvature of 0.02m. Condition (ii): The availability of the critical background
C. Modelling Of Lightning Attachment field necessary for the transition of a streamer into a leader
and the stable propagation of that leader (the magnitude of this
In both upward and downward propagating lightning, the
critical background field depending on the height of the
formation of upward propagating leaders from the wind
structure)
turbine is critical in terms of defining the location on the
structure that will be struck. For downward propagating
Equation (1) is used in determining the required background
lightning, the downward stepped leader provides a non-
field necessary for the propagation of the leader. According to
uniform electric field in the air around the wind turbine. This
[16] (a simplified theory of Lalandes theory), for stable
high electric field, when intensified by elements of the wind
propagation of the leader, the atmospheric electric field Eatm
turbine lightning protection system, will allow the formation
must be greater than the minimum value required for stable
of upward propagating leaders should specific conditions be
propagation of the leader Estab which is given by the following:
met.
For upward propagating lightning, the mechanism for the
formation of upward propagating leaders is nearly identical
but in this case the electric field in the air is near uniform 240
being generated by the cloud and not by a stepped leader. Estab ( H ) = + 12 . , with 1 (1)
H
1 +
This means that the strong influence of the stepped leader 10
position is removed from the consideration of upward leader
formation and it is likely that the likelihood of formation will Where H, is the height of the structure.
be dominated by the geometry of the wind turbine structure
itself. It must be noted that though these are the conditions required
for the successful attachment, they do not need to be fulfilled
To consider this issue further, it is important to consider the in the same order. For example, the background field might be
conditions necessary for the propagation of a leader. These are enough for the streamers to develop into leaders, before the
as follows and have been used in other papers relating to inception of streamers itself. Also, if one condition is fulfilled
upward / downward lightning (even though these conditions it is not necessarily the case that the other is readily available.
have been derived from laboratory experiments for the For example, a streamer might not develop into a leader if the
streamer inception from Frankin rods) [19]: background field necessary for the propagation is not
available. In such a case, according to [19], the streamers will
be incepted again and again until there is enough background
field for them to develop into leaders.
 7

D. Upward Lightning Initiation locations winter thunderclouds is different. The authors highlight the
The model of a wind turbine must be combined with different vertical extensions of the clouds and the different
information from the cloud model in the finite element temperatures change the position of the charge centers.
analysis software to allow analysis of the likely points of
upward lightning formation. The electric field applied in the
model is a uniform produced by a plane electrode located
200m above the ground. The magnitude of the applied electric
field is based on the results found from Malans charged cloud
model in the FEA simulations already described. When the
wind turbine is inserted into this electric field, field
enhancement is seen at certain locations of the wind turbine.
As is shown in Fig 9 and 10, the level of electric field
intensification is particularly high around the lightning
protection system of the wind instruments and at the blade tip
receptor.

Fig. 10. Field Enhancement around the Windvane

Fig 11 shows the increase in the electric field 200 meters

above ground due the lowering of the cloud charges in 100m
steps.

Fig. 9. Field Enhancement around the blade receptors (the locations of the
receptors are circles). Light areas indicate high electric fields.

The results from the FEA simulation show a high probability

of upward streamer inception from the blades and the wind
vane. The production of upward initiated lightning initially
depends on a given level of electric field enhancement existing
around a component such as the windvane / blade and also
their position [14].These results coincide with those found
Fig. 11. Changes in Electric Field Due to Lowering of Charges
from analysing the lightning strike data of the offshore
windfarm. The actual lightning strike data presented in Fig 4
Winter clouds over the North Sea have a stronger vertical field
showed that the windvane at the back is as vulnerable to
in the lower atmosphere [20]. A large number of lighting
lightning attachment as the blades. This correlates with the
strikes on wind turbines are usually experienced during winter
results from the model which showed significant enhancement
[9]. The move of wind farms to offshore sites and the low
on the lightning protection system of the wind instruments.
cloud height in winter would seem likely to increase the risk
of upward lighting in the case of wind turbines. The
The impact of cloud height on ambient electric fields above
hypothesis of upward lightning being a major contributor to
ground can be assessed using the FEA model of the
the lightning strike rate is strengthened when the distribution
thunderstorm cloud (Fig 7).
of lightning strikes by month is examined in Fig 5.
The x-axis of this Fig 11 refers to the height of the lowest 3C
VI. CONCLUSION
charge (Fig 7). The ambient electric field above ground
dramatically increases as the cloud height is lowered as the
relative separation of the tripole charges become significant in 1. Based on the results from the cloud models it is shown that
comparison with the distance to ground. Fig 11 indirectly the ambient electric fields that the wind turbine is subjected
relates to the risk of higher frequency of lightning strikes to increase with a decrease in cloud height. Winter clouds
during winter. Winter cloud height is lower as compared to which develop at lower altitudes are therefore more capable
that in summer. According to [20], the charge distribution in of developing higher electric fields at ground than those
compared to summer clouds.
 8

2. The offshore windfarm data shows that the wind vane Sheffield, Sheffield, U.K., in 1995 and the Ph.D. degree in electrical
engineering from The University of Manchester Institute of Science and
instruments are at considerable risk of lightning attachment
Technology (UMIST), Manchester, U.K., in 1998.
and that most of these lightning strikes to the wind turbines He is currently a Senior Lecturer in the Electrical Energy and Power Systems
happen during winter months Group, School Of Electrical and Electronic Engineering, University of
3. Results from the FEA wind turbine model follow those of Manchester. His research interests include power systems transients, the use
of higher voltage systems in aerospace applications, and power-system-
the real lightning strike data, highlighting the windvane and
induced corrosion. His work is based in the University of Manchester HV
the blades to be the components at the highest risk of Laboratories, the largest of any UK University where a 2MV impulse
lightning attachment. generator, 800kV AC test set and 600kV DC test set is available.
Dr. Cotton is a Senior Member of the Institution of Electrical Engineers and a
Chartered Engineer.
VII. REFERENCE:
[1] Berger, K., Novel Observations on Lightning Discharges: Results of
Research on Mount San Salvatore. Journal of the Franklin Institute,
1967. 283(6): p. 478-525.
[2] Berger, K. Anderson, and R.B.a. Krninger, Parameters of Lightning
Flashes. Cigr, 1975(41): p. 23-37.
[3] IEC 61400-24, Wind turbine generator systems Part 24: Lightning
protection for wind turbines June 2000.
[4] Cotton, I., et al., Lightning Protection Of Wind Turbines - A Designers
Guide To Best Practice. 1999.
[5] Rademakers, L., et al., Lightning Damage of OWECS. 2002(Part 1:
Parameters Relevant for Cost Modelling).
[6] Project FINO 3 http://www.fino3.de/Joomla/index.php, accessed
January 2008.
[7] Sceibe, K., T. Axmann, and C. Birk, Lightning Research for Offshore
Wind Farms in the North Sea. 29th International Conference on
Lightning Protection, 2008.
[8] McNiff, B., Wind Turbine Lightning Protection Project NREL
Subcontractor Report, SR-500-31115. 2002.
[9] Naka, T., et al., Experimental Studies on Lightning Protection Design
for Wind Turbine Blades.
[10] http://www.jomitek.dk/, accessed January 2008.
[11] Srensen, T., et al., Lightning Strike Sensor for Power Producing
Wind Turbines. European Wind Energy Conference and Exhibition,
Nice, France, 1999..
[12] Naka, T., et al., Study on Lightning Protection Methods for Wind
Turbine Blades. IEEJ Transactions, 2005. 125(10).
[13] Durstewitz, M., Wind Energy Report Germany 2001; Annual
Evaluation of WMEP; ISET, Germany. 2001.
[14] Alonso, M.A. and D.C. Irastorza, Dynamic Wind Turbine Lightning
Protection Behaviour Under Strom Conditions. 29th International
Conference on Lightning Protection, 2008.
[15] Rakov, V.A. and M.A. Uman, Lightning - Physics and Effects. 2000.
[16] Miyake, K., et al., Winter lightning on Japan Sea coast-lightning
striking frequency totall structures IEEE Transactions on Power
Delivery, 1990. Volume: 5( Issue: 3): p. 1370-1376.
[17] Bertelsen, K., et al., Application of numerical models to determine
lightning attachment points on wind turbines. ICLP 2007, 2007.
[18] Alonso, M.A. and I.A. Larrion, Calculation of electric fields in a wind
mill due to a lightning discharge using finite elements method.
[19] Akyuz, M. and V. Cooray, The Franklin lightning conductor:
conditions necessary for the initiation of a connecting leader. Journal
of Electrostatics, 2001. 51: p. 319-325.
[20] Broc, A., et al., Lightning strike to helicopters during winter
thinderstorms over North Sea. Aerospace Science and Technology,
2005. 9: p. 686-691.

Vidyadhar Peesapati was born in Edinburgh,

U.K, in 1980. He received a Class I B.Eng. degree
in electrical and electronics engineering from the
University of Madras, Chennai, India in 2001 and
the M.Sc. degree in electrical power engineering
in 2006 from the University of Manchester, where
he is currently pursuing the Ph.D. degree. He is
pursuing his Ph.D. on Lightning Protection of
Wind Turbines a part of the Supergen V Wind
Energy Theme

Ian Cotton (M98) was born in the U.K. in 1974.

He received a Class I B.Eng. (Hons.) degree in
electrical engineering from the University of