Methodology for Assessing Meteorological Observation Data to Account
for Wind Potential in The Design of a Wind Power Plant
NATALIA MAMEDOVA1
1Associate professor, Basic Department of digital economy, Higher School of Cyber Technologies,
Mathematics and Statistics
1Plekhanov Russian University of Economics
1Stremyanny lane, 36, Moscow, 117997
1RUSSIA
Abstract: The development of clean renewable energy sources is a strategic task to ensure the balance of energy
supply to territories. When implementing a policy of reducing dependence on or abandoning fossil fuels, the use
of renewable energy sources is an obvious competitive solution. And for territories remote from power supply
networks, the development of renewable energy sources is generally the only alternative. Wind energy is
increasingly being used to generate electricity. In this sense, accurate accounting of the influence of wind
potential on the energy balance is the basis of energy-saving architecture. From a thermodynamic point of view,
wind is a high-quality source of energy. Its high efficiency makes it possible in principle to convert into other
types of energy. However, the wind energy flow is unstable – the performance of wind power plants is due to
their extremely high sensitivity to the conditions of their location. In this situation, the reliability of the initial
data on wind energy resources is a criterion of paramount importance. Therefore, the development of a
methodology for evaluating data from long-term meteorological observations of wind speed and direction is of
important empirical importance. To design a wind power plant, it is not enough to enter ready–made data on the
value of specific power and specific wind energy in the territory into economic calculations - the data deviation
is too large. It is necessary to calculate the technical potential of the wind power plant for each prospective
location option. Both the approach to accounting for wind potential and the approach to scaling the data of the
observation station to remote territories ensure the reliability of the initial data for the design of a wind power
plant. The proposed methodology highlights all these aspects and offers an algorithm for evaluating the data of
long-term ground-based meteorological observations on the territory of Russia.
Key-Words: Renewable Energy Source (RES), Wind Energy, Meteorological Observations, Wind Potential, Wind Power
Plant (WPP), Wind Generator, Data Model for Wind Potential Assessment
Received: May 17, 2021. Revised: May 13, 2022. Accepted: June 16, 2022. Published: July 4, 2022.
1 Introduction
Russia has a significant wind potential, wind energy
resources are estimated at 10.7 GW, and the technical
potential of wind power plants is estimated at 2,469.4
billion kW per year. At the same time, based on the
energy balance data, these opportunities are
implemented insignificantly [1].
The area for which the technical potential of wind
power is calculated is determined by the formula:
, (1)
where ST - the area on which the technical
potential is calculated, m2; q is the area as a
percentage of the total area of the territory of Russia,
where the required wind speed prevails, %; S - the
total area of the territory of Russia – 17,125191x109
m2.
Energy wind zones in Russia are located mainly
on the coast and islands of the Arctic Ocean from the
Kola Peninsula to Kamchatka, in the areas of the
Lower and Middle Volga, Don, on the coast of the
Caspian, Okhotsk, Barents, Baltic, Black and Azov
Seas. Separate wind zones are located in Karelia,
Altai, Tuva, and Baikal [2].
According to JSC "SO UES" (the System
Operator of the Unified Energy System) - a
specialized organization that single-handedly carries
out centralized operational dispatch management in
the Unified Energy System of Russia - the installed
capacity of wind power plants for 2022 is 0.83% (in
2020 – 0.4%) of the total capacity of all power plants
of the Unified Energy System of Russia) [3]. Despite
the fact that now the share of wind power plants in
the UES (Unified Energy System) Russia does not
exceed 2%. For comparison, the installed capacity of
solar power plants is 0.8%, and hydroelectric power
plants – 20.25% of the total capacity of all UES
power plants.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
196
Volume 17, 2022
Despite the impressive resources of the wind
potential, investments in wind farms need state
support and patronage [1]. And we are talking about
direct investments and benefits not only at the
construction stage of wind farms. At the operational
stage, such support measures are applied as
compensation for losses of grid organizations in
power grid facilities and competitive selection,
following which the investor receives the right to
build renewable energy facilities of any kind with a
guaranteed return on investment. Similar measures
are being taken by the governments of the USA and
China [4, 5]. The Russian Government has
determined the limits of capital and operating costs
for renewable energy facilities [6]. This indicates the
presence of state interest in the development of
renewable energy. But the context of the support
measures themselves indicates that the development
of renewable energy as a defining vector of the
country's energy is not considered. Today, the
interest in renewable energy is no longer just a tribute
to the fashion of "green" technologies, but it is also
not comparable to the amount of financing of
traditional energy.
In terms of the development of "green" generation
in the period from 2021 to 2027, it is planned to put
into operation 2863.1 MW of wind and solar power
plants as part of the first stage of the renewable
energy development support program. Currently, the
Russian Government has approved a new program to
support the development of renewable energy until
2035, the cost of which can amount to about 360
billion rubles for the period from 2025 to 2035 and
ensure the commissioning of about 6.7 GW of
renewable energy capacity [7].
What prevents an increase in the share of wind
energy in the installed capacity of UES power plants?
In our opinion, technical and technological features
play a more significant role here than the immaturity
of state support.
The technical condition of the equipment (its
structural defects, wear), non-compliance with the
performance of the installed capacity equipment
create significant risks for investments in wind
energy.
It is also necessary to consider the increased
environmental restrictions on the conditions of
protection of air basins.
In addition, we should not forget about the factor
of reducing the use of installed capacity of power
plants. The maximum value of the Capacity
Utilization Factor (CUF) is 1. But in life and for
traditional power plants, it ranges from 0.4 to 0.8. The
highest CUF is for nuclear and geothermal power
plants (0.7-0.8), the lowest is for hydroelectric power
plants, since they are charged with removing load
peaks (4-5 hours a day). As for wind farms, their CUF
in Europe and Russia averages 0.2-0.3. But it
depends mainly on wind conditions. There are
examples of wind farms where CUF is 0.4 and higher.
Due to technical features for wind and solar power
plants, there is no guarantee of using power per hour
of maximum power consumption. After all, the
available capacity of wind and solar power plants
during the passage of the maximum power
consumption is assumed to be zero.
All of these are stress factors of the environment,
the study of which is a primary task for the
development of wind energy and investment in the
design of wind power plants.
It is not enough to be guided by the wind potential
data – they have many deviations. The main reason is
that the wind potential is determined according to
meteorological observations, and the distance
between observation stations in many regions of
Russia is 300 km or more. Yes, according to the
recommendations [8], in order to track every change
in the weather, the distance between observation
stations should be 50 km in rural areas, forests,
steppes, deserts; in the suburbs of megacities - 20 km;
in cities - 5 km (it is possible and more often, but it
will be necessary not to clarify the forecast, but to
study the microclimate of the city). But the current
situation with the number and distribution of
observation stations is very far from the
recommended one.
Since there are no other reliable data to assess the
wind potential, the methodology for assessing
meteorological observations of wind speed and
direction is a popular solution. On its basis, it is
possible to objectively consider the wind potential of
a separate territory for deciding on the design of a
wind power plant.
The methodology will be based on data from long-
term ground-based meteorological observations in
Russia. That is, the technique will use pure data
accumulated by observation stations. In the absence
of open databases and widely available algorithms
for assessing wind potential, the methodology will be
a source of reliable data for an informed decision on
the design of a wind power plant.
The initial parameters for the design of a wind
generator are presented in Table 1.
Table 1 – Specification of the wind generator
Wind
turbine
class
Power range,
kW
Range of
diameters of the
wind wheel, m
Range of rotation
speeds of the wind
wheel, rpm
0,025
1
0,5
2,5
2000
500
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
197
Volume 17, 2022
Very
small
1,5
10
3,0
9,0
500
200
Small
20
60
10
15
140
92
75
150
18
24
60
40
Medium
200
300
26
30
40
40
400
500
35
40
35
30
Large
600
750
43
48
30
30
900
1300
50
64
32
20
Very large
1500
3000
70
90
20
15
4000
6000
105
124
15
13
The need to compile a methodology for evaluating
data based on ground-based meteorological
observations is also determined by the fact that the
reliability of the NASA SSE1 (Surface Meteorology
and Solar Energy) data array in Russia is minimal.
This is due to the fact that more than half of the
territory of Russia is located in latitudes above 60°.
This conclusion was made based on the results of
work with the RETScreen International2
climatological information database containing data
from NASA SSE and ground observations.
2 Development of Methodology for
Evaluating Meteorological
Observations of Wind Speed and
Direction
The totality of wind characteristics in terms of its use
for the production of mechanical or electrical energy
is called a wind energy cadaster.
The main components of the cadaster are:
(1) Average annual wind speed. The annual and
daily course of the wind, i.e. its changes by
day and month of the year.
1https://power.larc.nasa.gov/data-access-viewer/
(2) Speed repeatability, types and parameters of
speed distribution functions, i.e. how long a
certain wind speed lasts during the year.
(3) Maximum wind speed.
(4) Distribution of wind periods and periods of
calm.
(5) Specific power and specific wind energy.
(6) Wind energy resources of the region, i.e. how
much energy can be generated from a certain
area.
The last two components of the wind energy
cadaster are, in theory, a data source for solving the
problem of the feasibility of designing a wind
generator. It is these data that can be used as the basis
for the calculation part of the project (Table 2). They
are a guideline when choosing a site for the
construction of a wind farm of nominal capacity.
Table 2 - Dependence of the specific power of a
wind generator on wind speed
Class
number
Class
characteristics
Specific
power,
W/m2 at a
height of 50
m
Average
annual speed,
m/s at an
altitude of 50
m
1
Poor
0-200
0,0-5,6
2
Marginal
200-300
5,6-6,4
3
Fair
300-400
6,4-7,0
4
Good
400-500
7,0-7,5
5
Excellent
500-600
7,5-8,0
6
Outstanding
600-800
8,0-8,8
7
Superb
> 800
> 8,8
The current procedure for making decisions on the
construction of a wind farm involves the following
mandatory steps:
(1) Selection of the location of the wind farm
(wind generator park) depending on the type
and market of electricity (wholesale/retail).
(2) Determination of the wind energy cadaster at
a pre-selected location based on data from
the nearest observation station.
(3) Determination of preliminary main technical
and economic indicators of a wind farm.
(4) Installation of a weather mast at the site of a
future wind farm and continuous observation
2https://www.nrcan.gc.ca/maps-tools-and-publications/tools/modelling-
tools/retscreen/7465
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
198
Volume 17, 2022
of wind speed in the directions of parts of the
world for at least a year.
(5) Determination on the basis of measurements
of normative technical and economic
indicators of a wind power plant and
deciding on its construction.
Despite the existing logic of the presented
procedure, the risks associated with the reliability of
the source data are significant.
In particular, a risk factor that can negate all
efforts is the determination of the wind energy
cadaster in a pre–selected location based on the data
of the observation station closest to this location. And
since the distance between them can be very
significant, the deviation of the solution is large. Of
course, the results of continuous observations for at
least a year will answer the question whether it is
worth placing a wind farm in this place. That's just a
negative answer leads to the starting point, from
which it is again necessary to determine the wind
energy cadaster for the newly selected location and
take meteorological readings again for at least a year.
That is, considering the fact that the reliability of
the initial data is a determining factor of economic
and energy efficiency, the choice of the location of
the weather mast for continuous observations should
be made on the basis of an assessment of data from
long-term observations of wind speed and direction.
This leads to the following question and the
following risk factor – what initial meteorological
data should be used to decide about the location of
the weather mast? Meteorological data requires
verification. For example, the data of the NASA SSE
database, when compared with the measurement data
of ground observation stations, show a deviation of
no more than 15%. However, this is an acceptable
deviation only if the projected wind farm is located
in the immediate vicinity of the meteorological
observation station. In other cases, when calculating
the wind potential of a particular location, such a
deviation is significant.
In this regard, the idea of symbiotic use of satellite
and ground-based observations should be abandoned.
The logical solution would be to focus only on the
use of long-term ground-based observations.
Accordingly, the data of the weather station for the
wind energy cadaster in the pre-selected location and
the zones of neighboring observation stations will be
used to carry out optimization calculations and
substantiate the efficiency of the projected wind
power plant.
Having determined the source of meteorological
data for the assessment of wind potential, it is
necessary to answer the following question, which
also includes a risk factor. Namely, for what period
is it advisable to accumulate data? A non-stationary
time series of data should have a sufficient and
acceptable duration so that its probabilistic
characteristic contributes to the identification of a
trend and deterministic periodicity.
We proceed from the dependence of the wind
potential on actinometric data, since the very
existence of wind is due to solar origin [9]. Therefore,
it is proposed to use a binding to the cycles of solar
activity based on the direct dependence of wind, as a
natural phenomenon, and its speed, as a consequence
of the effects of temperature changes, on solar
activity. The application of the Schwabe-Wolf cycle
[10] will provide the necessary duration of the time
series of meteorological data and will make it
possible not only to link the cyclicity of actinometric
data and wind potential, but also to compare data
between Schwabe-Wolf cycles to identify
dependencies. As a result of the assumptions made,
the developed methodology is based on actinometric
data of the 23rd cycle (May 1996 - December 2008)
and the 24th cycle (December 2008 – December
2019).
The next risk factor is associated with the
determination of the main technical and economic
indicators of the projected wind power plant. Unlike
other types of power plants, the generation of
electricity by a wind turbine cannot be regulated. It is
possible to stop the turbine and stop the production,
but to increase it, and in general it is impossible to
precisely adjust. Therefore, the CUF of wind power
depends on the characteristics of the unit itself and its
location. In this regard, the decision on the location
of the wind farm becomes even more responsible.
The application of a methodology for evaluating
meteorological observations can improve the
characteristics of deciding on the positioning of a
wind farm. This will increase the CUF.
As a conclusion, recognizing these risk factors as
significant, it is impractical to decide only on the
basis of open data from observation stations, since we
have no idea about the deviations inherent in the
calculations, nor about the method of scaling the data
to the territory of the entire region. It is important to
understand for what period to evaluate the data, by
what parameters to form a sample, how to consider
meteorological data at a distance from the location of
the observation station.
We propose a methodology that reproduces the
calculation of the wind potential step by step with
reference to a separate territory for making an
informed decision on the design of a wind power
plant. Such a technique will make it possible to
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
199
Volume 17, 2022
decide based on objective primary data, having a
holistic view of all possible options and alternatives
for considering wind potential in the selected
territory.
The methodology is based on a sequence of several
stages. At the first stage, the initial (primary) data of
meteorological observations are accumulated and a
database is developed. At the second stage, a data
model is built to assess the wind potential. At the
third stage, the model is interpreted and the
boundaries for deciding on the design of a wind farm
are determined.
2.1. Development of a Database of
Meteorological Observations of Wind Speed
and Direction
The sources of obtaining the initial information for
the meteorological observation data model are:
(1) Weather stations that measure all climatic
parameters, including wind speed, usually
four times a day. At modern weather stations,
measurements are carried out on 8 points, i.e.
directions relative to parts of the world:
north, south, east, west (4 directions) and
between them: northeast, etc. (4 directions).
(2) Continuous observation weather stations, as
a rule, constructed at the proposed sites for
the installation of wind power plants.
(3) Probes and balloons launched periodically to
different heights from certain stations, called
areological.
Since the decision on the sufficiency of the wind
potential should be based on data from long-term
observations, then, first of all, data from the network
of observation stations are in demand. Operating with
regular observation data will ensure the objectivity of
the data model. Continuous observation weather
stations are not a data source for the model, since
their network is much smaller. But their data helps to
calculate the magnitude of the error of the initial
series of daily values of wind speed and direction.
After all, 4 standard measurements of weather
metrics are made during the day.
The data obtained during continuous registration
are compared with the data of regular observations
and help to determine the deviation of daily
measurements. When comparing, the margin of
deviation and the confidence coefficient are
established. For the presented methodology, the error
value of the initial series of daily values for regular
measurements is 8-12%. The confidence coefficient
is 0.8.
The request for meteorological data is sent to an
organization that performs research and operational
and methodological functions in the field of
hydrometeorological forecasts. In Russia, these
functions are performed by the Federal State
Budgetary Institution "Hydrometeorological Center
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
200
Volume 17, 2022
of Russia". The request is formed considering the
following features.
For further processing, the data is requested in
symbolic form (in a table). The data structure
corresponds to the SYNOP (surface synoptic
observations) observation methodology. The sample
is based on observations over the past 30 years (from
1990 to 2020). The network of observation stations
includes all weather stations on the territory of Russia
located within meteorological zones 1 to 5. The
Hydrometeorological Center of Russia provides data
from 600 observation stations, at the moment the
number of operating stations is 585, 15 are closed or
mothballed.
Figure 1 shows a fragment of the list of
observation stations, which includes stations located
on the Arctic coast from the Kola Peninsula to
Chukotka.
Figure 1 – A list of observation stations of the Arctic
coast, the data of which are used to build the model
Recall that promising energy wind zones in
Russia, according to experts, are located mainly on
the coast and islands of the Arctic Ocean from the
Kola Peninsula to Kamchatka. For example, the
estimated capacity of the Kola wind farm under
construction (Enel Green Power) is 210 MWh, and
the project is considered highly profitable.
Having formed a data model of meteorological
observations of wind speed and direction, we will be
able to verify the correctness of the conclusions.
The request must include a list of climatic
parameters for data sampling:
(1) Average wind speed in m/s at an altitude of
50 m. above the earth's surface (during the
observation period - a month, a year),
regardless of the method of its determination.
(2) Maximum wind speed at gusts in m/s at an
altitude of 50 m. above the earth's surface
(during the observation period - a month, a
year).
(3) The average wind direction at an altitude of
50 m. above the earth's surface (during the
observation period - a month, a year).
(4) The repeatability (percentage of time) of
wind directions and calms having a speed in
the intervals of 0-2 m/s, 19-25 m/s (January,
July, year).
(5) Wind speed, the probability of exceeding
which is 5%.
(6) The highest wind speeds of varying
probability (wind speed possible 1 time in 5
years, 1 time in 10 years, 1 time in 15 years,
1 time in 20 years).
(7) Especially dangerous phenomena associated
with physical processes in the atmosphere.
Figure 2 shows a fragment of the map layer, on
which the locations of the observation stations of the
Arctic coast are plotted with geometrics. The
geometries contain the following data: the name, the
coordinates of the station, the height of the weather
site (the height of taking weather readings).
Figure 2 – Fragment of the map layer with the applied
geometries of observation stations
The monitoring data of all stations are combined
into a single SQL Server database. Forming a query
to the database, the analyst can set the boundary
values of the average and maximum speed, wind
direction, can select data on the repeatability of wind
direction.
With the help of a query, the analyst has the
opportunity to form map layers and visualize the
geometries of observation stations corresponding to
the sampling parameters. In addition, layers with
geodata can be loaded into the map, which will help
to correct the decision about the location of the wind
farm. For example, to load a layer on which the
locations of energy-deficient areas, or territorial
zones with a special economic status, or critical
infrastructure facilities are located.
2.2. Data Model for Estimating wind potential
The data model is by its nature a simulation model,
since it reproduces the work with parameters for
deciding on the design of a wind farm. To build the
data model, we proceeded from the basic conditions
for designing a wind farm based on the results of the
wind potential assessment.
First, the data model considers the following wind
parameters:
(1) The starting wind speed at which the wind
generator starts rotating is in the range from
2.5 to 4.0 m/s.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
201
Volume 17, 2022
(2) The nominal wind speed at which the power
of the wind generator reaches the nominal
value is from 10 to 14 m/s.
(3) The maximum wind speed at which the wind
generator is disconnected from the grid and
stops is in the range from 20 to 25 m/s.
(4) The wind speed of the storm (the speed at
which the stopped wind generator should not
collapse) is from 60 to 80 m/s.
These are the boundary parameters of wind speed,
according to which a request will be made to the
database and a list of observation stations will be
displayed, the location of which provides the starting
conditions for the design of a wind power plant.
The wind energy flow is very unstable. At a wind
speed of 10m/s, the specific power of the wind flow
is about 100 watts per 1 m2 of the area swept by the
blades of the wind generator, and at a speed of 5 m/s,
this power is 8 times less [9]. If this parameter is
applied in the data model to estimate the wind
potential, then the sample of locations recommended
for the design of a wind power plant will include
those for which the average wind speed is at least 8
m/s (ideally, at least 10 m/s).
Further, the data model considers that there is no
constant wind direction in the interior of the
continent. Since different parts of the land are heated
differently at different times of the year, the
seasonality factor is the determining factor for the
"wind direction" parameter. In addition, the data
model considers that the wind behaves differently in
the interior of the continent at different heights. And,
since yawing flows are typical for heights up to 50
meters, the solution generated by the data model
mainly relies on climatic readings recorded at an
altitude of 50 meters and above. For observation
stations located on and near the coast (up to 40
kilometers), the factors of seasonality and the height
of the weather site do not have a determining value in
the data model.
Secondly, the data model considers such a
parameter as the rated power of the wind generator.
In general, the power range is from 0.025 kW to
6,000 kW, based on the classification of wind
generators [11]. When designing a wind farm, the
data model considers that the wind generator operates
with rated power only if the wind speed is equal to or
greater than the nominal. The rest of the time, the
wind generator operates with less than nominal
power. Therefore, in the data model for estimating
wind potential, the calculation is based on the
nominal average annual wind speed and nominal
power.
When designing a wind farm, the wind generator
is calculated for a certain power, for example 800
kW. With an average annual wind speed of 6 m/s, the
wind generator will produce 1,500,000 kW/hours of
electricity per year, with an average annual wind
speed of 5 m/s – 1,100,000 kW/ hours of electricity.
A 2,000 kW wind generator with an average annual
wind speed of 6 m/s will produce 3,700,000
kW/hours of electricity per year, with an average
annual wind speed of 5 m/s - 2,300,000 kW/hours of
electricity.
If it is necessary to increase energy production, for
example, by 1.5 times, then in addition to the option
of changing the location of the wind farm, the data
model also considers the option of increasing the
height of the mast to 22-25 meters. This makes it
possible to increase the average annual wind speed at
the axis height by 20-30%.
The data model will form the same solution when
the average annual wind speed is less than 4 m/s. It
should only be noted that the model has a parameter
for bridging the gap in the distribution of wind speed
frequency. Such a gap occurs when data on zero and
low speeds are entered into the model. And the
solution is to use not the Weibull distribution, but a
polynomial regression model as a probability density
function for the wind speed frequency [12].
Thirdly, the data model calculates for the
projected wind farm and the diameter of the rotor of
the wind generator. It is selected based on the average
annual wind speed. The rated power of the wind
generator is determined by the diameter of the rotor
squared. Here the data model manifests itself as
follows. With winds up to 6-7 m/s, the data model
shows that the output of a rotor with a diameter of 5
meters is higher than that of a rotor with a diameter
of 4.2 meters. At average annual wind speeds of more
than 10 m/s, the output is leveled.
In addition to the parameter of the nominal power
of the wind generator, the data model also considers
technical limitations. For example, it is considered
that at a wind speed of 12-13 m/s, the power of the
wind generator reaches a nominal value of 1 MW and
remains constant in the range of 13-25 m/s. It turns
out that a significant power of the wind flow is not
used. But no other solution is possible, since it is
impossible to overload the wind generator above its
rated capacity.
The expansion of the operating range in the data
model is considered impractical for the following
reasons. Firstly, the average annual wind speeds of
more than 25 m/s were not recorded by observation
stations during the study period. Secondly, the wind
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
202
Volume 17, 2022
pressure on the wind wheel during its rotation is
proportional to the area of the swept surface. And, so
that the pressure force does not overturn the wind
farm, it is necessary to strengthen the foundation and
its attachment to the tower. Such a decision would be
economically wrong.
Combining all the above parameters considered
when designing a wind farm, the wind potential
assessment data model generates a solution. The
solution can be generated based on the nominal
average annual wind speed, and offer options for the
nominal power of the wind generator of the projected
wind power plant. Or the opposite situation is
possible, when, based on the nominal power of the
wind generator available in the project, a decision is
generated on the options for the location of the wind
power plant, considering the range of values of the
nominal average annual wind speed.
The rated power of the wind generator (RWPPs)
depends on the measured in m/s, air density
(p=1.225kg/m3), wind energy utilization coefficient
(Avg=0.45), gearbox efficiency coefficient (ηred),
diameter of the wind wheel squared (D2), wind speed
in cube (V3), coefficient the efficiency of an electric
generator (ηgen), or in other words, the coefficient of
conversion of mechanical energy into electrical
energy.
The following formula is used [13]:
(2)
The total efficiency of the mechanical and
electrical elements of the power path of the wind
turbine is 0.9.
In Russia, the average air density at the earth's
surface is 1258 g/m3; at an altitude of 5 km - 735 g/
m3, 10 km - 411 g/ m3, 20 km - 87 g/ m3. At the
equator, the density values in the troposphere are less,
and in the stratosphere - more than in Europe. In
winter, the air density is greater than in summer. In
the proposed model, the air density is assumed to be
1.225 kg/m3, considering the average values of the
height of the location of meteorological sites at
observation stations.
The capacity factor (CF) of wind power depends
on many design features, but ultimately on the profile
of the blade and the degree of its roughness, as well
as on the ratio between the speed of rotation of the
blades and wind speed. This coefficient ultimately
determines the efficiency of the wind generator. The
maximum value of the CF coefficient is 0.593 [14,
15]. On land, the throughput coefficients range from
0.26 to 0.52 [16]. In the proposed model, the value of
the CF coefficient is assumed to be 0.45. This
corresponds to the characteristics of high-speed wind
turbines with streamlined aerodynamic blades.
2.3. The Order of Interpretation of the Data
Model for the Assessment of Wind Potential
and Scaling of Meteorological Observations
Even with meteorological observations from the
stations at your disposal, you need to answer one
important question. Namely, whether it is possible to
use these data to assess meteorological conditions in
territories remote from the location of the observation
station. There are also related questions. At what
distance from the station and under what conditions
does the data remain valid? What approach should I
use to scale the data?
According to the results of the research of the
leading Russian scientific center in the field of
actinometry of the A.I. Voyekov, measurement data
with an acceptable deviation can be extrapolated to a
distance of no more than 130 km from the weather
station [17]. Accordingly, as one of the parameters of
the data model for estimating the wind potential,
scaling of data to a radius from 80 to130 km from the
observation station is accepted. The exact value
within the interval is determined considering the
terrain – for homogeneous terrain, the value is
assumed to be 130 km. The inhomogeneity of the
terrain reduces the data scaling radius, as the
accuracy of wind speed and direction readings
decreases. The greatest volatility of data
characterizes wind resources in areas with
mountainous and foothill terrain [18].
The order of scaling in the methodology for
evaluating meteorological observation data is as
follows.
The calculation is based on a number of
parameters and using several approaches. The
physical parameter is comparable terrain conditions.
The second parameter is the distance between the
observation stations. The third parameter is the
distance from the station location. And then three
approaches are used: calculation of average
indicators of weather stations connected to a
network; data extrapolation method; data
interpolation method. And a combination of
approaches is also used, for example, the method of
extrapolation of averaged values.
The determination of the average values of the
climatic characteristic is reduced to the calculation of
the average values recorded by the network of
observation stations. The selection for subsequent
calculations depends on the number of stations – a
small number of weather stations leads to the fact that
all available stations are included in the network. It is
clear that the greater the number of monitoring
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
203
Volume 17, 2022
stations in the network, the more accurate the
calculations of the average values.
The average long-term values reflect the patterns
of wind propagation in the territory. It remains to
make sure that they can be trusted. As a positive
factor, we have the homogeneity of the available
observations and an unchanged instrumental base
installed at the observation stations. The error of the
initial series of daily values of wind strength and
direction obtained during continuous recording is 8-
12%. Based on a confidence probability of 0.8, the
required length of the time series is 18-25 years.
Calculations have shown that the length of the time
series in 20 years gives an error of 5.5% - that is, it
does not go beyond the error of the original series.
This allows us to state the accuracy of the average
long-term ground observations for the selected
averaging period of 22 years (two Schwabe-Wolf
cycles).
To improve the quality of data in the processing
of meteorological observations, methods of
extrapolation and interpolation of data from
observation stations are used.
The extrapolation method based on the average
value of the time series is used if the distance between
the observation stations in a straight line is more than
260 km – the square of the maximum allowable
extrapolation distance with an acceptable deviation.
With a smaller distance between observation stations,
the extrapolation method based on the average value
of the time series is used if the relief between the
stations is not uniform and the difference between
wind directions is more than two points (recall that
the wind direction measurement is carried out by 8
points).
Extrapolation formula based on the average value
of the time series:
, where (3)
– the average value of the time series levels in the
past;
- extrapolated level value; L – lead time;
– the level taken as the extrapolation base.
The confidence bounds for the mean are defined
as follows:
, where (4)
– tabular value of Student's t-statistics with n-1
degrees and probability level p; – the average
square error of the average value.
The total variance associated with both the
fluctuations of the sample average and the variation
of individual values around the average will be
. Thus, the confidence intervals for predictive
evaluation are equal to:
(5)
The data interpolation method is applied
according to Newton's formula. This formula is
convenient when interpolating functions for values of
x close to x0. Newton's formula has an advantage over
Lagrange's interpolation formula – when using it, the
number of nodes can be increased without repeating
all calculations again. This corresponds to the
requirements of the methodology for estimating
meteorological observation data for cases when a
decision is made to interpolate data across several
locations to scale the data.
Newton 's interpolation formula:
(6)
At the same time, it is necessary to consider the
calculation deviation. The deviation affects the
accuracy of the design of a wind farm, the choice of
equipment, the choice of operating mode and the
forecast of the amount of energy received. The
deviation increases as the radius of the distance from
the observation station increases. Extrapolation or
interpolation of averaged values carries a smaller
deviation, which speaks in favor of using these
approaches for data scaling.
When extrapolating the daily values of wind
strength for 100 kilometers, the deviation is 11%, for
200 kilometers – 17%, for 300 kilometers – 25%, for
500 kilometers – 42%.
When interpolating by two points of the location
of observation stations, the value of interpolation to
the middle of the distance between the points (up to
500 kilometers) reduces the error by a factor of 1.5 in
comparison with extrapolation. The interpolation
error is comparable to the averaging error for an
observation station at a distance of 100-120
kilometers between stations. If the distance between
the points exceeds the specified value, interpolation
is performed on three or more points. At a distance
greater than or equal to 500 kilometers, the
interpolation deviation is such that it makes it
impractical to use the approach to scale the data of
meteorological observations of wind speed and
direction.
The interpretation of the data model for the
assessment of wind potential includes a set of
indicators reflecting the average values of long-term
meteorological wind measurements by parameters.
The complex includes the following indicators:
(1) Coefficient of variation of seasonal (annual)
values of wind speed (wind direction), which
reflects the degree of variability of the
regime.
(2) The number of hours in a day with an average
(acceptable) wind speed (wind direction) in a
set of indicators for the month (year) to
consider the features in the daily data.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
204
Volume 17, 2022
(3) Type of distribution of daily wind speed data
(wind direction) - prioritization order –
negative asymmetric island-top distribution;
negative asymmetric distribution; normal
distribution; bimodal distribution; positive
asymmetric distribution; positive
asymmetric island-top distribution.
3 Conclusions
The proposed methodology for evaluating
meteorological observations data to account for wind
potential can be used in the design of a wind power
plant of any type and any rated power of a wind
generator.
Actually, the methodology itself is a ready-to-use
solution and can be developed by increasing the
number of parameters on the basis of which a
decision on the design of a wind farm is generated.
The methodology can be integrated into the structure
of the feasibility study of the project and be included
in the calculation part of the assessment of the
payback of the investment project and the energy
efficiency of the wind farm.
We emphasize that considering the wind potential
to determine its location is a mandatory and very
important stage of design. The proposed method not
only increases the reliability of decision-making, but
also significantly reduces the risks associated with
determining the location of the projected wind farm.
References:
[1] Energy strategy of the Russian Federation for the
period up to 2035. URL:
https://policy.asiapacificenergy.org/node/1240
[2] Decree of the Government of the Russian
Federation No. 1209-r dated 09.06.2017
(General layout of electric power facilities for the
period up to 2035). URL:
http://www.consultant.ru/document/cons_doc_L
AW_218239/
[3] Unified Energy System of Russia: Interim
results, Information Review, March 2022, URL:
https://www.so-
ups.ru/fileadmin/files/company/reports/ups-
review/2022/ups_review_0322.pdf .
[4] Center for Sustainable Systems, University of
Michigan, 2021, Wind Energy Factsheet, Pub.
No. CSS07-09.
[5] Jun Li, Decarbonising power generation in China
- Is the answer blowing in the wind?, Renewable
and Sustainable Energy Reviews, Volume 14,
Issue 4, 2010, Pages 1154-1171
[6] Decree of the Government of the Russian
Federation No. 1-r dated 08.01.2009 "On
approval of the Main directions of state policy in
the field of improving the energy efficiency of
the electric power industry based on the use of
renewable energy sources for the period up to
2035". URL: http://government.ru/docs/20503/
[7] Order of the Ministry of Energy of the Russian
Federation No. 88 dated 02/26/2021 "On
approval of the Scheme and Program for the
Development of the Unified Energy System of
Russia for 2021-2027"). URL:
http://www.consultant.ru/document/cons_doc_L
AW_384492/
[8] Laurencas Raslavičius, Vytautas Kučinskas,
Algirdas Jasinskas, Žilvinas Bazaras, Identifying
renewable energy and building renovation
solutions in the Baltic Sea region: The case of
Kaliningrad Oblast, Renewable and Sustainable
Energy Reviews, Volume 40, 2014, Pages 196-
203
[9] Hydrometeoizdat, Instruction to
hydrometeorological stations and posts, issue 3,
part 1, Leningrad, 1985.
[10] Climatic data for renewable energy in Russia
(climate database): Textbook. – M.: MIPT
Publishing House, 2009. - 58 p.
[11] R.M. Wilson, A Comparison of Wolf's
Reconstructed Record of Annual Sunspot
Number with Schwabe's Observed Record of
‘Clusters of Spots’ for the Interval of 1826–1868,
Sol Phys 182, 217–230 (1998).
https://doi.org/10.1023/A:1005046820210.
[12] Set of rules, SP 267.1325800.2016. High-rise
buildings and complexes. Design rules
https://docs.cntd.ru/document/456044284 .
[13] Lingzhi Wang, Jun Liu, Fucai Qian, Wind speed
frequency distribution modeling and wind energy
resource assessment based on polynomial
regression model, International Journal of
Electrical Power & Energy Systems, Volume
130, 2021, 106964
[14] P.P. Bezrukikh, Wind Energy: Reference and
methodological manual, Moscow, "ENERGY",
2010.
[15] A. Betz, Introduction to the Theory of Flow
Machines. (D. G. Randall, Trans.), Oxford:
Pergamon Press, 1966.
[16] N.E. Zhukovsky, NEZH type windmill,
Proceedings of TsAGI, 1920 (lithographed
edition).
[17] U.S. DOE, NREL (2015) “Transparent Cost
Database: Capacity Factor” Open Energy
Information.
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
205
Volume 17, 2022
[18] V.Y. Okorenkov, Methods and means of
verification of meteorological information and
measuring systems and measuring instruments,
Publishing House of the Central Research
Institute “Asterion”, St. Petersburg, 518 p.
[19] Juchuan Dai, Yayi Tan, Wenxian Yang, Li Wen,
Xiangbin Shen, Investigation of wind resource
characteristics in mountain wind farm using
multiple-unit SCADA data in Chenzhou: A case
study, Energy Conversion and Management,
Volume 148, 2017, Pages 378-393
Creative Commons Attribution
License 4.0 (Attribution 4.0
International , CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on POWER SYSTEMS
DOI: 10.37394/232016.2022.17.20
Ilirian Konomi, Valma Prifti, Andrin Kërpaçi
E-ISSN: 2224-350X
206
Volume 17, 2022
