Aug 4-Factors Affecting Performance of Wind Farm in Kenya
Aug 4-Factors Affecting Performance of Wind Farm in Kenya
Aug 4-Factors Affecting Performance of Wind Farm in Kenya
JULY 2015
i
CHAPTER ONE
INTRODUCTION
1.1 Background of the Study
Over the last two decades, energy is becoming a basic factor for industrial production and
the level of electricity consumption is used to measure the progress and economic
development of nations. Globally, growing population, industrialization and rising living
standards have substantially increased dependence on energy. However, renewable
energy production has experienced a rapid growth especially in the United States and
Europe, with significant growth in emergent countries such as China and India (Wiser &
Bolinger, 2012). Wind power can simply be defined as electricity generated from sources
that naturally exist and are not subject to depletion due to its consumption such as windmills
or turbines.
Wind energy production contributed a significant share to this expansion and has
attracted institutional investors; because of wind energy is safe, clean and abundant.
However, some factors such a limited transmission capacity and higher capital start-up
costs of wind power, as compared to fossil fuels, make it difficult for the expansion of
wind power in some markets (Thollander, 2007). Wind power is playing a major role in
power supply in an increasing number of countries. According to REN21 (2014)
renewable energy consumption achieved high levels of penetration in several countries.
By the end of 2013, at least 85 countries had seen commercial wind activity, while at
least 71 had more than 10 MW of reported capacity by year’s end and 24 had more than 1
GW in operation.
Annual growth rates of cumulative wind power capacity have averaged 21.4% since the
end of 2008, and global capacity has increased eightfold over the past decade. In the EU,
capacity operating at year’s end was enough to cover nearly 8% of electricity
consumption in a normal wind year (up from 7% in 2012), and several EU countries met
higher shares of their demand with wind. 68 Wind was the top power source in Spain
(20.9%, up from 16.3%) during 2013, and met 33.2% of electricity demand in Denmark
(up from 30%). According to United Nations Industrial Development Organization
1
(UNIDO) (2011) four German states had enough wind capacity at year’s end to meet over
50% of their electricity needs.
In the United States, wind power represented 4.1% of total electricity generation (up from
3.5% in 2012) and met more than 12% of demand in nine states (up from 10% in nine
states in 2012), with Iowa at over 27% (up from 25%) and South Dakota at 26% (up from
24%). Wind power accounted for 2.6% of China’s electricity generation. Globally, wind
power capacity by the end of 2013 was enough to meet an estimated 2.9% of total
electricity consumption (IEA, 2014). Asia remained the largest market for the sixth
consecutive year, accounting for almost 52% of added capacity, followed by the EU
(about 32%) and North America (less than 8%). Non-OECD countries were responsible
for the majority of installations, and, for the first time, Latin America had a substantial
share (more than 4.5% (Wolsink, 2012).
China led the market, followed distantly by Germany, the United Kingdom, India and
Canada. Others in the top 10 were the United States, Brazil, Poland, Sweden and
Romania and new markets continued to emerge in Africa, Asia, and Latin America. The
leading countries for wind power capacity per inhabitant were Denmark (863 W per
person), Sweden (487.6), Spain (420.5), Portugal (412) and Ireland (381). Africa and the
Middle East saw little new operating capacity beyond Morocco (0.2 GW) and Ethiopia,
which completed Africa’s largest individual wind farm (120 MW), with the aim of
mitigating the impact of dry seasons on national hydropower output (Watts, Schluter &
Whiting 2014). However, other countries in the region moved ahead with new projects,
and several announced longterm plans.
2
Frankfurt School of Finance and Management gGmbH (FSFM 2013) in terms of
investments in wind power have been continuously growing since 2004.
More importantly, new players have called for attention as a considerable increase in their
investments was observed and not expected. Among these countries are: South Africa,
Morocco, Mexico, Chile and Kenya. According to FSFM (2013) in all these cases
governments are taking leading roles in encouraging RE such as wind power usage to boost
power supply, improve energy security and reduce concentration dependence from other
sources. These actions have taken a more important role in places where it was noted that
electricity access is limited, which is the case of many African countries.
Sub-Saharan Africa has been advancing rapidly since 2000, with the region’s economy
doubling in size by 2013. Energy demand in the region grew by around 45 percent over
the same period (IEA, 2014). Even so, some 630 million people, or two-thirds of the
population, still lack access to electricity. Large wind projects like the new one in Kenya,
Lake Turkana Wind Power project will help fill this gap over the coming decades. A
recent report by the International Energy Agency determined that renewable energy will
make up almost half of sub-Saharan Africa’s power generation growth by 2040. Kenya
has stepped up its efforts to transform 40,000 acres of land into a wind farm, in a bid to
3
meet growing demand for electricity. The 300 MW Lake Turkana Wind Power Project,
which is being developed in the country's North-East, hopes to produce 20% of the
country's current installed electricity generating capacity when it comes online in 2016.
In Kenya, wind power accounts for about 1% of the country’s total power generation, a
number that is expected to rise above 10% by 2017. The new wind project “Lake Turkana
Wind Power project” is expected to bring change gradually in the social-economical life
of the citizens. Kenya hopes to add 5,000 megawatts of power to its grid, an ambitious
4
target that will take installed capacity up from 1,700 megawatts in 2013 to 6,700
megawatts by 2017. Aside from wind and solar power capacity, Kenya is rich in
hydropower and geothermal energy sources. The country recently installed a 280-
megawatt geothermal plant one of the Africa largest. Geothermal recently overtook
hydropower as the single largest electricity provider in the country. The demand for
electricity grows, that issue is not going away. Electricity consumption grew by 9%
between 2010 and 2011, according to the Institute of Economic Affairs (2013) demand
for electricity will grow by a further 12% by 2030. In terms of wind projects Kenya
introduced feed in tariff policy to “drive down prices and as a means to price discovery”
therefore leaving all the market forces interact freely. Despite increasing investment in
power energy the overall picture reflects that the growing opportunities wind power are no
longer represented by the traditional power generating industry but its relevance in terms of
volume are still significant. As such the current study aims to bridge this knowledge gap by
investigating factors affecting performance of wind farm in Kenya with focus to Ngong
Wind Farm.
ii. To establish the influence economic situation on the performance of wind farm in
Kenya with focus to Ngong Wind Farm.
iii. To examine how technology development on the performance of wind farm in Kenya
with focus to Ngong Wind Farm.
5
iv. To determine the influence of government policies on the performance of wind farm in
Kenya with focus to Ngong Wind Farm.
From the specific objectives the following research questions will be answered:
ii. How does economic situation influence performance of wind farm in Kenya with
focus to Ngong Wind Farm?
iii. Does technology development affect performance of wind farm in Kenya with
focus to Ngong Wind Farm?
iv. How do government policies influence performance of wind farm in Kenya with
focus to Ngong Wind Farm?
6
Asia. In the case of Africa, the electricity access has been estimated at around the 58% out of
the one billion people that live in that continent. To put these numbers into perspective, an
attempt to achieve universal electricity access will demand investments of 641 billion US
dollars between the years of 2010 and 2030 which translates into roughly 32 billion per year.
This investment level is nothing compared to the observed in the fossil fuel (262 billion)
industry and even less than what was made during 2012 in developing countries for the RE
industry (112 billion) (FSFM 2013). As such only those countries that aims to improve their
productivity will achieve their MDG which will only be achieved through enhancing power
production. Thus the current study thus aims to investigate factors affecting performance of
wind farm in Kenya with focus to Ngong Wind Farm.
7
CHAPTER TWO
LITERATURE REVIEW
2.1 Introduction
The chapter provides an extensive literature and research related to factors affecting
performance of wind farm in Kenya. The chapter is thus structured into theoretical,
conceptual and empirical review. The study also presents the knowledge gap the chapter
seeks to fulfill.
Extrapolating wind speeds from the height of measurement stations to the much higher
hub height of wind turbines is probably one of the most critical uncertainty factors
affecting the wind power performance at a site [25]. The change in horizontal wind
speeds with height (known as wind shear or the wind profile) is generated by friction
from the earth’s surface and so is highly dependent on the specific site conditions: the
surface roughness of the terrain, air temperature, season, atmospheric stratification and
the wind speed itself.
Many simplifications for extrapolating wind speeds exist, most notably the empirically
derived power law and the theoretically derived log law (Equation).
wx= A log(hx-d/z0)
The simplified log law assumes that wind speed, w, is related to the logarithm of its
height, h, under the assumption of neutral atmospheric stability. The logarithmic wind
profile tends to zero at a height of the roughness length, z 0, plus the surface displacement
height, d, and is scaled by a constant, A, which equals the friction velocity (u*) divided
8
by 0.4 (von Karman’s constant). The log law provides robust extrapolations of wind
speeds over a range o fwind speeds, locations and altitudes [25,27].
The mechanical wind power that can be extracted from the wind depends heavily on the
wind speed, and for each wind speed there is always an optimum turbine speed at which
the wind power extracted at the shaft of the turbine is maximum, at any other speed apart
from this optimum speed we get sub-standard operation of the system. This is basically
area specific, because the wind speeds would vary from place to place. Now the
mechanical power transmitted at the shaft is:
P=0.5Cp AρV∞3
Cp is a function of the TSR & the pitch angle. For a wind turbine with radius R, the
above formula can be written as,
P=0.5Cp πR2ρV∞3
The maximum value of the shaft power output for any wind speed can be expressed as:
Pm α ω3
The coefficient of power of a wind turbine basically signifies the conversion efficiency of
the wind energy of the wind into mechanical energy, which in turn is used to drive the
generators. It differs from the overall system efficiency as it doesn’t include the losses in
transmission (mechanical) and in electrical power generation. In horizontal axis machines
the theoretical limit is known as Betz limit, which is around .593(16/27 or 59.3%). For
good turbines it is in the range of 35-45%.
Combining the observed ideal load factor data allows one to calculate a weather-
corrected load factor (WCLFf,t) for every observation, giving a time series for each farm
that should not be affected by wind conditions changing from month to month. The actual
9
load factor (LFf,t) for a given farm (f) and month (t) is divided by the ratio of its ideal load
factor (ILFf,t) for that month to the farm’s mean ideal load factor over a given period
(whether or not the farm existed throughout the whole period), as in Equation below. This
can be simplified to the performance ratio (PR) for each month multiplied by the farm’s
average ILF.
The WCLF represents what a particular farm would have produced each month if wind
conditions followed their long-term mean distribution. As with the uncorrected load
factors, the absolute value encompasses the available wind resource, the quality of the
local site conditions, and the turbine model installed, but the variation over time is no
longer dominated by seasonal weather patterns.
Hughes (2007) demonstrated that it is possible to separate the impact of turbine ageing
from these other factors by using an error components model with fixed effects. The
observed load factor, LFf,t, of wind farm f in month t is estimated by least squares
regression against the ideal load factor, ILFf,t, with fixed effects for each site, sf, and age of
the turbine, Af,t, minimizing the sum of the squares of the error component, εf,t, as in
demonstrated in the equation below:-
The fixed effects (sf and Af,t) are the freeform equivalent of a linear trend.
Several factors can confound observed output in a fleet of wind farms, given that a
turbine’s output is dependent on wind speeds at its site and the efficiency with which it
captures the energy in that wind. For example, if wind speeds have fallen slightly over
time, farms would have lower load factors in upcoming months, when they are at their
oldest, giving a spurious correlation between used duration and their performance. If
improvements in design can be increased a turbine’s output relative to capacity (its power
coefficient) then newer turbines (of the improved design) will have higher load factors
than old turbines, so that turbine output appears to decline with age, when really it
improves with newer generations. On the other hand, if the best (windiest) sites can be
10
occupied first, then old farms could have higher load factors than new ones built on
inferior sites, so that turbines would appear to improve with duration used.
Independent Variables
Dependent Variable
Organization characteristic
Economical characteristic
Performance of wind farm
Technology advancement
Government support
Loring (2007) stated that higher levels of citizen participation increases the support
towards these projects at the local level, helps develop trust and creates the necessary
networks of support and acceptance for these developments. It also gives the opportunity
to control the project locally, meaning that the projects can be tailored according to the
local context (Walker, 2008). Citizens working together and sharing a community identity
is often seen as a way to overcome obstacles met in the development path, such as a
negative perception towards renewable energy technology (Walker, 2007).
Moreover, projects that are partly or fully owned by the community have higher chances of
being accepted at the local level, to receive easier planning permissions or to obtain the
necessary funding for their development. Also citizens can procure equipment and materials
collectively, which can lower the purchase costs (Walker, 2008). Looking at the size of the
initiative, large-scale projects are found to have an advantage in terms of cost saving (Warren
and McFadyen, 2010) although a smaller size is preferred since it is easier to manage or to
diffuse the necessary knowledge (Walker, 2010). Another reason is that if projects are easy to
control, they can be replicated a lot easier. Therefore, managing multiple small-scale projects
often poses fewer difficulties than managing a single, large-scale complex project. Moving
fast and being successful on small-scale projects is seen as a better way to attract the interest
of more people.
12
information and training as a reason of unawareness of RE options, benefits or financing
opportunities.
Most of rural and remote areas are automatically at a disadvantage due to the high costs
associated with supplying electricity to meet its low electric demands, as well as the
prohibitive costs of building additional transmission lines to connect rural communities
(Brown and Escobar 2007). This is true in many rural areas of the world where low
population densities result in high capital, operation, and maintenance costs for electric
utilities.
Subsidy programs are found to be crucial for the development of wind power initiatives,
as sometimes the projects would not even have started without the existence of external
funding (EREC, 2008). However, many times getting access to funding can be a
13
complex process (Walker, 2007). Therefore it is important to have the right tools to attract
this funding, meaning a flexible organization and the necessary skilled people that put
strategies that attract the funding (Weerawardena & Mort, 2004). It can also be the case that
the necessary funding is inconsistent or can stop; therefore, finding other means that reduce
the dependency on external funding is seen as a step forward for the development of such
initiatives (Bomberg, McEwen, 2012).
Sometime, because the projects are small, the building of the necessary infrastructure to
support them can be considered too expensive only for this sole reason and therefore, will not
be built (Neuhoff, 2005). On the other hand, other authors argue that smaller scale projects
are more advantageous since they avoid making expensive upgrades to the grid or it avoids
voltage regulation challenges (Walker, 2008). Besides this, a powerful incentive for
choosing wind power technologies is the desire to be autonomous and disconnected from
large traditional energy companies. Working in autonomous small groups can provide the
liberty and the flexibility for a better and quicker organization (Bomberg and McEwen,
2012). Moreover, being autonomous from a centralized energetic system provides the
freedom to pursue the best interests of the citizens/members of the project and those
interests that fit the locality the best, instead of following the interests of a big company.
Another incentive is that through the adopted technology, there can be achieved an
improved management of the local energy demands (Walker, 2008). However, the
14
installations need to be easy to install and to maintain. It was found that if too much
maintenance is required, the technology is difficult to install or it does not add value to
the property, this could be turned into barriers towards adopting the renewable energy
technology (Faiers, Neame, 2006). Also, trained personnel are required since the
deployment of the technology does not only depend on the availability of the technology
itself, but on the available human capital as well.
It has been noted that the number of trained personnel in renewable technology needs to
grow, since the experience on the market with the embedded technology is already bigger
and exists for a longer period of time (Fressoli, Thomas, 2012). Seyfang (2010) stressed
on the importance of the end user in the development and diffusion of renewable energy
technology, since he might accept or not the new technology. In this sense the visibility of
the technology is often put into emphasizes.
According to European Commission (2005) it is evident that the countries which had the
highest average effectiveness during the last seven years such as Germany, Spain,
Portugal and Ireland had applied feed-in tariffs to promote electricity produced by wind
onshore power plants. Whilst Germany and Spain already effectively supported wind
onshore electricity before 2003, the wind onshore development in Ireland and Portugal
caught up after 2004. Regarding Ireland, the change from the tendering system to a feed-
in tariff, which took place in 2006, helped to speed-up the development of wind onshore
energy. Comparing the policy effectiveness of wind onshore electricity with previous
analysis, it becomes clear, that countries using quota obligations such as Italy, Sweden
15
and the United Kingdom have caught up in terms of policy effectiveness in particular in
2009 and 2010 (European Commission, 2008). However, their performance still lags
behind policy effectiveness in the group of effective feed-in tariff countries Spain,
Germany, Portugal and Ireland.
Another prominent issue is the availability of grants and subsidies, since the high
investments required for renewable energy is often seen as a limitation (Faiers & Neame,
2006). Therefore, projects are particularly advantageous not only when they can reduce
energy costs with the installed technology, but also when grants are obtained in order to
lower the investment costs (Walker, 2008). Moreover, it is important that the local
authorities balance the national targets with the local needs (Loring, 2007). This can be
done through the right institutional support that helps put actors together and with local
organizations that promote renewable energy production (Walker, Gordon, Evans, Bob,
Devine-Wright, Patrick, Hunter, Sue and Fay, Helen, 2007). Failing to put into places the
necessary learning mechanisms or integrating these initiatives into long-term
governmental programmes or goals can slow down or hinder the performance of wind
power (Walker, 2007).
McCormick (2007) defined socio-political issues, which include the process of how
actors (individuals and organizations) make decisions, resolve conflicts, form
partnerships, respond to government policies and engage with public issues. Such issues
correspond to a combined connection between the nature of related policies, power and
the public. Political issues regarding the interests of powerful actors in the form of
coalitions amongst strong interest actors contribute extensively to the adoption of policies
related to RE (Geller, 2003). This dimension of social acceptance can explain the
differences between countries with high and low RE implementation, in which political
will has already taken action. Within this element, acceptance of technologies and
policies, the public level of acceptance of key actors, and policy-makers are components
enhancing (or deterring) effective RE policy development.
16
power curve estimation. Ilinca (2011) reports estimated power losses due to icing
conditions up to 50% of total annual production. Hughes (2012) indicates declining
turbine performance due to increasing age in Denmark and the UK. Some further papers
apply nonparametric methods to estimate the wind energy production frontier. Kusiak et
al. (2012) use data envelopment analysis (DEA) to assess the performance of wind
turbines in presence of faults. They identify turbine downtime as the major reason for lot
of production. Ibarren et al. (2014) analyze the entire process of wind energy production
and include further production factors, such as land and investment cost in their DEA
model.
Moreover, Wolsink (2000) emphasizes the dynamic aspect of attitudes towards wind
farms and argues that they can be crucially affected by institutional factors, such as the
pattern in which decisions concerning the sitting of the turbines are made and how well-
informed local communities are with regard to wind energy issues (Jobert, 2007).
Wolsink (2007) argues that when local communities resist wind power installations, they
actually express their disapproval either of the top-down decision making procedures
used by the developers and policy makers or of the incentives of the prospective
developers. Further, resistance is also associated with the fact that emphasis is often laid
on large-scale development, rather than local concerns over wind power projects.
17
2.5 Critic of the Literature
Starting with an example briefly mentioned before, the study by Hansen & Wernerfelt
(1989) investigated common determinants that affect the performance (ROA) of
companies by grouping them into an economic and organizational perspective. Among
the elements examined were: the industry‟s profitability, competitors‟ concentration,
companies‟ size, the well-being of human resources and the goal emphasis from
employees. Although it was recognized by the authors that these determinants react and
perform differently according to the industry, they also notice that they are commonly
found in many previous studies, therefore signaling their importance as an initial ground
of determinants.
To examine the effect the asset size has in power plants, a study by Locatelli and Mancini
(2011) was reviewed. According to the authors, most of the research efforts in power
plants are focused in large projects examinations and his study tackles this issue by
including small and medium sized specifications. By comparing his analysis with
previous results and contrasting them with the Internal Rate of Return (IRR) generated
the authors claim that larger projects have a competitive edge against the smaller
counterparts. They also argue that this is in part explained due to the efficiency levels
achieved by larger operations although due to restrictions in design and locations having
a smaller installation is sometimes the only alternative for some projects. This is
something that applies as well to the RE industry as for the places with restricted
electricity access it was mostly convenient to install small and medium scale projects
rather than a utility scaled project.
A study by Jacobs et al. (2010) it can be said that announcements from environmental
related causes are mostly positively received in stock markets. Although mixed signals
were found the overall observed relationships were positive therefore giving the RE
industry a competitive edge due to the nature of its operations. This study also reflects
how aware markets are from the signals given by companies even if these are
announcements and not yet completed actions. Blacconiere et al. (2000) found that market
valuation changes due to the deregulation effects of electric utilities. Since this industry is
closely related to the RE sector it was opted to deeply review the elements included to derive
18
possible determinants. As expected, this study does report market level elements that are
responsive to variables as the Book value of Equity and stock returns. In this way it was
observed that population, income, Consumer Price Index (CPI), fuel costs, CAPEX Price
Index Costs and interest rates reflected mixed signals from the expected prior to their
examination.
19
CHAPTER THREE
RESEARCH METHODOLOGY
3.1 Introduction
This chapter presents the methodology, which will be used to carry out the study. It
further describes the type and source of data, the target population and sampling methods
and the techniques that will be used to select the sample size. It also describes how data
will be collected and analyzed. The suitable methodology in this study gives the
guidelines for information gathering and processing.
research. This research problem will be studied through the use of a descriptive research
design. According to Cooper and Schindler (2003), a descriptive study is concerned with
finding out the what, where and how of a phenomenon. This study therefore will be able
to generalize the findings to all staff involved in operation of Ngong wind power.
The main focus of this study will be quantitative. However some qualitative approach
will be used in order to gain a better understanding and possibly enable a better and more
insightful interpretation of the results from the quantitative study. This method concerns
the intense investigation of problem solving situations in which problems are relevant to
the research problem. The underlining concept is to select several targeted cases where an
intensive analysis identified the possible alternatives for solving the research questions on
the basis of the existing solution applied in the selected case study. The researcher
attempts to describe and define a subject, often by creating a profile of group of problems
20
3.3 Target Population
Target population as described by Borg and Crall (2009) is a universal set of study of all
members of real or hypothetical set of people, events or objects to which an investigator
wishes to generalize the result. The target population of this study will be Ngong wind
farm while the study population will be all staff who deal directly with the day to day
Ngong wind farm. There are 60 staff working in the Ngong wind farm both in technical
and operation activities. Given that the target population fo the study is small the study
will adopt cesnsus where all staff will be involved in the study
The data instrument will address the four research objectives while it will be sub-divided
into two sections. The first section of the questionnaire will enquire general information
about the respondents, while the next sections answered the four objectives. The
researcher will involve two research assistances to help in distribution of question to the
targeted respondents. The questionnaires will be administered through drop and pick later
method. The quantitative section of the instrument to be employed will use both a
nominal and a Likert type scale format to determine each of the variables. A 5 point
Likert scale ranging from 1 to 5 was used as answers to statement like questions. The
Likert - type format is selected as the format yields equal - interval data, a fact that allows
for the use of more powerful statistical to be used to test hypotheses (Kiess and
Bloomquist, 2008).
21
3.5 Pilot Testing
The purpose of the pilot testing aims to establish the validity and reliability of the
research instruments and hence enhance face validity (Joppe, 2000). The pilot testing will
be conducted using the questionnaire on 10 staff. The pilot group will be done through
random sampling.
3.5.1 Validity
According to Mugenda and Mugenda (2003) validity is the accuracy and meaningfulness
of inferences, based on the research results. One of the main reasons for conducting the
pilot study is to ascertain the validity of the questionnaire. The study used both face and
content validity to ascertain the validity of the questionnaires. Content validity draws an
inference from test scores to a large domain of items similar to those on the test. Content
validity is concerned with sample-population representativeness. Gillham (2008) stated
that the knowledge and skills covered by the test items should be representative to the
larger domain of knowledge and skills.
3.5.2 Reliability
Reliability is concerned with the question of whether the results of a study are repeatable.
A construct composite reliability co-efficient (Cronbach alpha) of 0.6 or above, for all the
constructs, will be considered to be adequate for this study. The acceptable reliability
coefficient is 0.6 and above (Rousson, Gasser and Seifer, 2002). Cronbach Alpha was
used to test the reliability of the research instrument.
22
in prose. The study will also used inferential statistics to establish factors affecting
performance of wind farm in Kenya. Specifically, the study will use Spearman correlation
to establish this relationship.
23
REFERENCES
Carvalho, A., Gonzalez, M.C., Costa, P. & Martins, A. (2009). Issues on Performance of
Wind Systems derived from Exploitation Data. Industrial Electronics, 2009. IECON
'09. 35th Annual Conference of IEEE: 3599-3604.
Gunturu and C. A. Schlosser (2012): Characterization of wind power resource in the United
States. Atmospheric Chemistry and Physics, 12(20): 9687-9702.
Homola, M.C., Byström, J., Nicklasson, P.J., Sundsbo, P.A. (2009). An improved method
for wind power estimation. Working Paper, Narvik University College, Norway,
1-13.
Hughes, G. (2012): The Performance of Wind Farms in the United Kingdom and
Denmark. Renewable Energy Foundation, London.
Ibarren, D., Vazquez-Rowe, I., Rugani, B., Benetto, E. (2014). On the feasibility of using
emergy analysis as a source of benchmarking criteria through data envelopment
analysis: A case study for wind energy. Energy 67: 527-537.
Ilinca, A., (2011). Analysis and Mitigation of Icing Effects on Wind Turbines, Wind Turbines,
Dr. Ibrahim AlBahadly (Ed.), ISBN: 978-953-307-221-0.
Kneip, A., Simar, L., Wilson, P.W. (2014): When Bias Kills the Variance: Central Limit
Theorems for DEA and FDH Efficiency Scores. Econometric Theory, FirstView.
Kusiak, A., Verma, A., Wei, X. (2012): Wind turbine capacity data from SCADA. Wind
Systems Magazine. September 2012: 36-39.
24