FULLTEXT01
FULLTEXT01
FULLTEXT01
Oscar Larsson
1. Acronyms: ..................................................................................................................... 2
2. Introduction: .................................................................................................................. 3
2.1 Assumptions for this project: .................................................................................. 4
2.2 Purpose: ................................................................................................................... 4
3. Theory:........................................................................................................................... 5
3.1 Solar Geometry and Irradiance: .............................................................................. 5
3.2 Total solar module area to cover annual energy consumption: .............................. 7
3.3 Evaluating the total required area of the solar plant: .............................................. 8
3.4 Electrical energy consumed over a time period: ................................................... 10
3.5 One line diagram: .................................................................................................. 12
3.6 Solar cell production: ............................................................................................ 13
3.6.1 Simulating the optimal angle: ......................................................................... 13
4. Results and data: .......................................................................................................... 14
4.1 Results from optimal tilt angle simulation: ........................................................... 14
4.1.1 Weather data: .................................................................................................. 14
4.2 Load profile of Karlstad airport: ........................................................................... 15
4.2.1 Estimated energy load after the addition of chargers: .................................... 16
4.3 Facility grid construction: ..................................................................................... 19
4.4 Required size for solar park: ................................................................................. 19
4.4.1 Simulation of the PV systems: ........................................................................ 21
4.5 Storage potential through excess energy generation: ........................................... 22
4.6 Current grid structure: ........................................................................................... 23
4.7 Power through the current grid: ............................................................................ 24
4.7.1 Addition of chargers into the system. ............................................................. 26
4.8 Embedding LVP into the system: ......................................................................... 27
4.9 Connecting PV system into the grid: .................................................................... 28
5. Discussions: ................................................................................................................. 30
5.1 Evaluating possible placements of the future solar plant: .................................... 30
5.2 Evaluating the daily load peaks: ........................................................................... 32
5.3 SAM: ..................................................................................................................... 33
5.4 Future grid design: ................................................................................................ 34
5.4.1 Design of the PV system:................................................................................ 34
5.4.2 Design of the storage system: ......................................................................... 36
5.4.3 Storage system: ............................................................................................... 39
5.5 Microgrid or conventional interconnection with the current grid ......................... 39
5.6 Future system: ....................................................................................................... 40
6. Conclusion: .................................................................................................................. 41
7. Appendix: .................................................................................................................... 43
8. References .................................................................................................................. 46
1
1. Acronyms:
Acronym Word
LFV Luftfartsverket
2
2. Introduction:
The airport sector accounts for around 5% of the global greenhouse gas emissions and is
expected to increase in the future with flights steadily increasing from the early 2000s until
2020 with the SARS-covid pandemic decreasing the number of flights from 40.3 million
in early 2020 to 16.9 million later during 2020 [1]. Global annual flights are on the increase
since the sharp decrease in 2020 which is leading to the increase of greenhouse emissions
from the flight sector. Reducing greenhouse emissions is crucial to reduce the effect of
global climate change and decarbonizing the flight sector is a crucial step to a more
sustainable industry. Decarbonisation of the energy supply chain is accomplished by
introducing sustainable and environmentally friendly energy production and consumption
processes. It is of primary importance to sustain the energy balance between production
and demand for the stability and the security of an electrical grid. That is to maintain the
frequency and the voltage levels in about the rated values aiming to avoid black-outs.
Making the energy production secure and balanced is difficult when integrating solar and
wind energy because their energy production is non-plannable which in turn makes the task
of integrating more loads into the system more difficult [2] [3]. To avoid grid failures and
unforeseen interactions due to differences between power demand and production
optimized power management solutions are required for stable operability of the electricity
system [4] [5] [6] [7]. Due to great advancements in AI, big data analytics and IoT energy
prediction and planning could be supported using these technologies instead of using grey
and white-box prediction methods when performing system performance studies [8] [9]
[10] [11]. Due to individual consumers having different consumption profiles load
predictions are difficult, but by IoT integration into intelligent grids power predictions
could become more accurate, combining this with ESS it could have a prominent role in
relieving the power grid and also used in energy exchange [12] [13].
No snowfall over the solar park during winter: Due to that the most recent hourly snow
fall data from NSDAP comes from 1986 and because of switching trends in weather due
to climate change, no snow data will be included in the simulation of the solar plant.
For any system the three phases will be equally loaded: Because of insufficient data
over the loads of each of the phases, a detailed calculation and analysis of the load on each
phase is not possible. In this project, all loads will be treated as a balanced three phase
system with no phase shifting occurring on any of them.
2.2 Purpose:
The purpose of this study project is to investigate some possible way to implement a green
airport completely powered by solar power supported by an ESS. To do this, the project
will look at the following:
4
- Using previously investigated information to construct a suggested implementation
for the future system.
3. Theory:
3.1 Solar Geometry and Irradiance:
Solar declination is the tilt of the polar line relative to the perpendicular line from the sun.
This declination can be estimated by the following equation:
180 (1)
𝛿=( ) ∗ (0.006918 − 0.399912 ∗ cos(𝐵) + 0.070257 ∗ sin(𝐵)
𝜋
− 0.006758 ∗ cos(2𝐵) + 0.000907 ∗ sin(2𝐵) − 0.002697
∗ cos(3𝐵) + 0.00148 ∗ sin(3𝐵))
360
𝐵 = (𝑛 − 1) ∗ ( )
365
Where 𝛿 is the solar declination and ϕ is the latitude of the location 𝜑𝑠𝑠𝑟 is the angle where
sunrise and sunset occur.
The hour angle is the angle to the sun relative to the local meridian that the location is.
Because the earth rotates once per day, the hour angle can be described by the equation:
𝜑 = 15 ∗ (12 − ℎ) (3)
5
Where h is the hour of the day.
There are several models for estimating the total solar irradiance that falls on a tilted
surface. Some of the more widely used models are the isotropic model, the HDKR model
and the Perez model. The first two models, the isotropic and HDKR model are simpler to
implement and can be used with the information provided about the sites for evaluating the
expected power production. In this project the focus will be on the HDKR model because
it provides more accurate simulations while still being easier to implement into code.
1 + 𝑐𝑜𝑠𝛽 β (4)
λT = (𝜆𝑏 + 𝜆𝑑 ∗ 𝐴𝑖 ) ∗ 𝑅𝑏 + 𝜆𝑑 ∗ (1 − 𝐴𝑖 ) ( ) ∗ (1 + 𝑓 ∗ sin3 ( )) + 𝜆
2 2
1 − 𝑐𝑜𝑠𝛽
∗ 𝑃𝑔 ∗ ( )
2
Where 𝜆T is the total solar irradiance per hour [J/mm2 *h], 𝜆b is the beam irradiance also
referred to as DNI per hour [J/mm2 *h], 𝜆d is the diffuse horizontal irradiance also referred
to as DHI. β is the tilt angle of the surface, 𝜆 is the total solar irradiance that falls on a
surface horizontal to the ground per hour which also is referred to as GHI [J/mm2 *h], I0 is
the total extra terrestrial irradiance that falls on a horizontal surface every hour [J/mm2 *h]
and it can be estimated using the following equation:
𝑛
𝜆0 = 𝐺𝑠𝑐 ∗ (1 + 0.033 ∗ cos (360 ∗ ) ∗ (𝑐𝑜𝑠𝜙 ∗ cosδ ∗ cos𝜑 + sinϕ ∗ sinδ) (5)
365
Where Gsc is the solar constant [W/mm2], n is the nth day of the year, ϕ is the latitude of
the surface, δ is the solar declination and 𝜑 is the hour angle
6
Pg is the diffuse reflectance of the surroundings which also is referred to as albedo. 𝑓 is a
modulating factor to correct for cloudiness which is describes by the equation:
𝜆𝑏 (6)
𝑓 = √( )
𝜆
Rb is the geometric factor which when the surface is situated in the northern hemisphere
can be described by the equation:
𝜆𝑏 (8)
𝐴𝑖 =
𝜆0
Where EAn is the annual energy consumption of the facility expressed in kWh and EPVAn is
the annual power production of the solar power facility expressed in kWh. EPVAn can be
expressed as the following:
7
Where GAn is the annual solar irradiance that falls on the surface expressed as kWh/m2, APV
is the solar panel area expressed in m2 and η𝑡 is the efficiency of the system which is
unitless. η𝑡 Is expressed as:
Where η𝑃𝑉 is the efficiency of the solar module, η𝐼𝑛𝑣 is the efficiency of the power inverter
and η𝑀𝑃𝑃𝑇 is the efficiency of the maximum power point tracker (MPPT).
Where AMReq is the required solar module area to cover the facilities annual power
consumption. With this AMReq can be expressed as the following:
𝐸𝐴𝑛 (15)
𝐴𝑀𝑅𝑒𝑞 =
𝐺𝐴𝑛 ∗ η𝑃𝑉 ∗ η𝐼𝑛𝑣 ∗ η𝑀𝑃𝑃𝑇
This equation can be used to estimate the required module area to cover the facility with
an annual energy consumption using the parameters specified in the equation.
𝐻𝑃𝑉 (16)
𝐺𝐶𝑅 =
𝐿𝑠𝑝𝑎𝑐𝑖𝑛𝑔
8
Where Lspacing is the space length between each row of arrays of solar panels and HPV is the
height of the solar panel. The GCR can have a value ranging from 0 < GCR < 1.
The total modular area of the solar park can be expressed as:
And the total area for the entire facility can be expressed as:
Where Atot is the total required area for the entire solar facility, Lb is the Length of a single
solar cell, nb is the number of solar cells per solar panel, Lh is the height of a single solar
cell, nh is the number of solar cells per solar panel, nrow is the number of rows of solar panel
and nstring is the amount of strings of solar panels.
Which means that the total area required can be expressed as:
9
3.4 Electrical energy consumed over a time period:
When presented with a graph like the following:
Where P(t) is the instantaneous power over the load expressed in Joules per second and t
is the time expressed in seconds the total energy consumption of the load between t2 and
t1 can be expressed as the following.
𝑡1
𝐸 = ∫ 𝑃(𝑡) ∗ 𝑑𝑡 (20)
𝑡2
Where E is the total energy consumed over the period t1 to t2 expressed in joules.
𝑛2
𝐸 = 𝑇 ∗ ∑ 𝑃̅ [𝑛]
(21)
𝑛=𝑛1
Where T is the sampling time in seconds. 𝑃̅ is the average power consumption over
sampling interval n and is supposed constant during whole time T and E is the energy
consumed over the time period n1 and n2.
10
In this project, the facility does not have a continuous power monitoring system but instead
uses an energy monitoring system. This system monitors the total hourly energy
consumption of the facility which leads to a graph like this:
120
100
(kWh/h)
80
60
40
20
0
00:00 00:00 00:00 00:00 00:00
Hour
Where each dot is a measuring point summarizing the total energy consumed over that
particular hour. Because the energy is measured in kWh/h the alternative graph can be
expressed as follows:
120
Power consmuption
100
80
(kW)
60
40
20
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
This shows that the energy consumption in kWh/h can instead be expressed as an average
power consumption of the facility in kW over the entire hour. Because of the limited load
11
data over the continuous load profile of the facility this method will be used when
evaluating the load profile of the facility.
The annual energy consumption EAN can be determined using the following equation
Where Ph is the average hourly power consumption expressed in [kWh/h], Eh is the hourly
consumed energy of the facility expressed in [kWh], n is the nth hour of the year and EAN
is the annual energy consumption expressed in [kWh]
𝑆3𝑝ℎ𝑎𝑠𝑒
𝐼= (23)
√3 ∗ 𝑈3𝑝ℎ𝑎𝑠𝑒
Where the one line diagram will have aggregated loads where multiple loads on a shared
transmission line will be aggregated into one larger load. This is due to the fact that the
power cables between each load could have the same power rating which means that it can
be viewed as a long shared cable.
12
Figure 4: An example of a one line diagram
The picture above demonstrates an example load, because of this all lines connecting each
of the nodes have the same rating. These four separate nodes going into separate loads can
be aggregated into one larger load:
This simplification works under the assumption that all of the phases in each load are
equally loaded.
Figure 6: A curve showing the annual solar energy that falls on a surface depending on the tilt angle of the surface
As shown in Figure 6, at a tilt angle of 37.3 the annual energy that falls upon the surface is
846.173845kWh/m2.
14
4.2 Load profile of Karlstad airport:
At Karlstad airport the energy consumption of the facility is monitored by Ellevio for the
purpose of billing the current energy consumption and additionally bill the highest hourly
energy that the facility has been importing from the grid. The facility does not monitor the
continuous load that the facility imports from the grid but instead collects the total energy
consumption of the facility over the sampling time. The time interval between each energy
sampling is once every hour and the information about the energy consumption during the
sampled interval is transmitted directly to Ellevio where the information is logged to
Ellevio´s website. It is from this website that the following energy load profile data has
been extracted.
600
Hourly energy consumption
500
(kWh/h)
400
300
200
100
0
Jun-2017
Dec-2017
Jun-2018
Jun-2019
Jun-2020
Jun-2021
Apr-2017
Apr-2018
Apr-2020
Aug-2017
Aug-2018
Dec-2018
Apr-2019
Aug-2019
Dec-2019
Aug-2020
Dec-2020
Apr-2021
Aug-2021
Dec-2021
Feb-2017
Feb-2018
Feb-2019
Feb-2020
Feb-2021
Oct-2017
Oct-2018
Oct-2019
Oct-2020
Oct-2021
Date
Figure 7: Hourly energy consumption from Karlstad airport from Feb-2017 to Dec-2021
Because the energy consumption of the facility is readily available this project will use this
information as a guide to the specifications of the future solar plant. Mainly this will look
at the annual energy requirements of the facility. Looking closer at Figure 7, the most
energy intense periods are during the colder periods of the year. Since the most energy
intense periods also have the highest average power consumption using these periods when
15
evaluating the stress on the existing grid would be the most appropriate, these energy
intense periods occur between November and February.
Using the load profile provided by Karlstad airport prior to any added chargers the
following values are given for each of the years.
16
Using the daily flight schedule provided from Karlstad airport:
Monday – Saturday
Route Departure Return
Karlstad -> Oslo -> Karlstad 06:00 08:00
Karlstad -> Gothenburg -> 06:30 08:30
Karlstad
Karlstad -> Oslo -> Karlstad 17:00 19:00
Karlstad -> Gothenburg -> 17:30 19:30
Karlstad
Table 2: Suggested flight schedule
Because the time between each departure is around 30 minutes and the fastest possible
charging time for an individual plane is 40 minutes, the airport would at least need two
airplanes available to be able to meet the demand of departures.
Monday – Saturday
Route Departure Return Plane in charge
Karlstad -> Oslo -> Karlstad 06:00 08:00 Plane 1
Karlstad -> Gothenburg -> Karlstad 06:30 08:30 Plane 2
Karlstad -> Oslo -> Karlstad 17:00 19:00 Plane 1
Karlstad -> Gothenburg -> Karlstad 17:30 19:30 Plane 2
Table 3: Flight schedule with assigned planes
While one charger could potentially cover all the charging for both planes while still being
able to keep to the flight schedule, having multiple chargers has the following advantages:
Due to the limited information about the ES-19 charger, it is unclear whether the charger
could charge the plane at lower power level, this would be advantageous since battery
lifetime is negatively affected by higher charge rates. Having a lower power rating also has
the advantage of decreasing the stress on both the chargers and on the transmission cables
in the system. In this situation, it would be more advantageous to charge the planes during
a period of multiple hours instead of the optimal charging time of 40 minutes.
Looking at the flight schedule plane 1 and plane 2 have a period of 9 hours between the
next departure which could be used to charge the plane. The recommended charging time
would be around 8 hours which would give an hour to prepare for the departure or any
temporary changes to the flight schedule.
2
𝐸𝐹𝐵𝑎𝑡𝑡𝑒𝑟𝑦 = ℎ ∗ 𝑃𝑐ℎ𝑎𝑟𝑔𝑒𝑟 = ∗ 1000kWh
3
(24)
𝐸𝐹𝐵𝑎𝑡𝑡𝑒𝑟𝑦
𝑃𝑐ℎ𝑎𝑟𝑔𝑒𝑟 =
ℎ
(25)
Where EFBattery is the energy required to fully charge the battery from an uncharged battery,
h is the hours required for charging and Pcharger is the power delivered to the charger.
1000 2
𝑃𝑐ℎ𝑎𝑟𝑔𝑒𝑟 = ∗ ≈ 83,33𝑘𝑊
8 3
18
Comparing the two power modes and their daily energy consumptions:
1000
Average power consumption
800
(kW)
600
Charging 8 hour
Charging 40 minutes
400
No charging
200
0
0 10 20 30 40 50 60
Half-hour
Figure 8: Average half-hour power consumption of facility. Grey curve shows the facility consumption without chanrging, blue curve
shows power consumption with charging over 8 hours and orange curve shows power consumption with charging over 40 minutes.
Looking at the graph, the shorter loading time will have a much greater impact on the
system during its loading time than the longer charging time. It would be highly advised to
try to extend the loading time of the charging to avoid unnecessary stress on both the
chargers and the grid.
By referencing Table 1 and equation (15) using the value from the most energy intense
year 2019 with an annual energy consumption of 2.214GWh as Ean, the value 846.174
[kWh/m^2*year] from the code as GAN and using efficiency values of the Power
Electronics: FS3150M power inverter of around 97.011%, and efficiency of around
20.052% for the polycrystalline solar panel CSI Solar co. Ltd. CS3Y-470P. the following
equation can be made:
The modular area required to provide the facility with solar energy would be around 1,68
hectar.
With the addition of electrical airplanes, the annual energy requirement would increase to
around 2,65GWh, This would be an energy increase of around 19% which would mean
that to cover the additional energy requirement the solar plants modular area would need
to be 19% larger, which amounts to around 20000m2 or around 2 hectar.
20
4.4.1 Simulation of the PV systems:
The system specification are selected as the following:
Solar cell CSI Solar co. Ltd. CS3Y-470P CSI Solar co. Ltd. CS3Y-470P
Table 4: Table over the specifications of two solar power plants suggestions.
As seen in Table 4 both systems are oversized to the inverter rating. The reason for this is
that oversized systems have a higher safety margin meaning that this system can deliver
power even in less ideal circumstance. The other reason is that oversizing also reduces the
mismatch power loss.
21
The choice of having a fixed solar panel installation is based on safety because of glare.
While tracking solar panels have a higher energy output than a fixed installation having
tracked solar panels located around the airstrip increases the risk of glare occurring to the
pilots when they come in to land the airplanes.
Comparing the power production of the solar farm with the consumption during the year
2018:
120
100
80 System power generated
60 Energy Consumption
40
20
0
0 20 40 60
Week
Figure 9: Weekly energy consumption of the facility versus the solar power plant production.
This shows that an excess of energy is produced during the summer and an energy
deficiency occurs during the darker parts of the year. During the colder months to
accommodate the increased energy consumption and reduced production a total of
1233,86MWh of energy needs to be imported from the grid even when supported by the
PV production. During the summer an excess of 1783,91MWh of energy is produced. This
leaves an excess of 550,8MWh of energy which could either be stored or exported directly
to the grid.
22
Figure 9 also shows that the energy consumption of the facility exceeds the available
energy production during the colder months. This shows that if the requirement of having
a green airport where the energy used in the airport is 100% green, storing an excess of
1233,86MWh of energy and using that during the colder months is required.
The grid is represented as a slack generator on the HV side, T1, T2 and T3 are the installed
power transformers, each of them have a power rating of 1600kVA, AL are multiple loads
aggregated into an aggregated load and the generator at Backup LVP are the generators
installed in the facility that are used during LVP. T1 and T2 are the primary power
transformers in the system while T3 is exclusively used during the LVP protocol.
23
4.7 Power through the current grid:
Inspecting the load profiles over the facility in Figure 7, it shows that the heaviest load
periods occur between the months of November and February. This general trend occurs
because during the colder periods of the year the systems that rely on electricity for heating
are used more frequently. Because a measured continuous load over the facility is not
available, estimating the load profile from the energy profile of the facility is necessary
using the theory in 3.4 with the energy graph between 2017-2022.
Because the colder seasons experience a significantly higher peak energy loading these
periods will be the ones that are investigated for the current status of the grid. By using the
highest hourly energy consumption that occur each day between November and February
from 2017 to 2022 coupled with the delivered VArh for that particular hour a peak power
graph can be made:
600
500
[kVA]
400
300
200
100
0
Nov 2017
Nov 2018
Dec 2019
Jan 2019
Nov 2019
Nov 2020
Nov 2021
Dec 2017
Jan 2018
Feb 2018
Dec 2018
Feb 2019
Jan 2020
Feb 2020
Dec 2020
Jan 2021
Feb 2021
Dec 2021
Jan 2022
Figure 11: This graph shows the peak mean power consumption during the colder months between november and febuary.
Here the highest recorded average power consumption occurred in Dec 2021 with a value
of approximately 674kVA.
As mentioned in the assumptions, because of the lack of information about either the energy
or power flow through the system all of the load points in the system will be treated as
equally loaded.
24
During the peak loading period if the assumption is that a 674KVA loading during this
period the system would look like this:
Figure 12: One line diagram of the airports grid without any additional chargers
Each of the transformers have a power rating of 1600kVA and all the connections between
the nodes have a current rating of 3200A.
The LV system runs on a 400V line to line RMS voltage then the equation becomes the
following:
This means that the cables have a power rating of 2216kVA in the LV system.
While the parts of this system can survive and function even when they are overloaded over
a shorter time period having the system heavily loaded over a longer period of time is
unwise as this decreases the lifetime of the transmission lines and transformers in the
system. In this project a power level of above 80% of the rated power is too heavy as it
would dramatically decrease the service time of the transformers. [18]
25
This means that the rating for the transformers and cables would decrease by 20% to be an
acceptable power load. This would mean that each of the transformers would have an
acceptable rating of 1280 kVA and the cables have a rating of 1773 kVA.
The total load that the transformers can handle would be 2560kVA and the average
apparent power load during the heaviest period is 674kVA. That would mean that the
system would have 1886kVA of extra power that could be added into the system.
Figure 13: One-line diagram of the airport with the addition of two chargers. Each of the chargers contribute 83,33kW each
Figure 13 shows that adding the chargers to the system under the condition that the chargers
would stay on a lower power rating would be feasible with no risk of overloading either
26
the transformers or the transmission lines. While the system can handle the lower power
level it is important to look at how the system would react if the charging mode of the
chargers were to switch to the highest power setting. This scenario could occur either
accidentally or deliberately during an unexpected change in the flight schedule or similar,
the system would need to be able to handle a sudden increase in required power.
Each of the chargers have a claimed charge power rating of 1MW, this would mean that a
total of 2MW would need to be able to be inserted into the system. Because of the large
load that each of these chargers provide, the placement of chargers will be on each end of
the one-line diagram. This is to try to spread out the chargers’ impact on each of the
transformers. Inserting the chargers into the system gives the following one-line diagram:
Figure 14: One-line diagram with additional chargers. Charger 1 is placed on bus 9 and charger 2 on bus 13. Each of the chargers here
consume 1MW.
Figure 14 show that inserting the chargers directly to the system could potentially overload
the power transformers in the system.
27
two hours. When LVP activates, the system disconnects T1 and islands parts of the system
to ensure that the crucial parts of the airport are energized. T3 is closed to support T2 for
the entire duration of the LVP. During LVP the system looks like the following:
During LVP AL1 needs to be supplied for a minimum of two hours. Because the LVP
disconnects all other loads and keeps the designated equipment on constantly during LVP
AL1 can be seen as a constant load. Because the load is constant under a duration of two
hours that would mean that an energy storage of at least 24kWh would be necessary to
supply the designated equipment.
28
Figure 16: Sattelite picture of the airport. The blue outline shows the size that the suggested solar power plant would occupy.
This area is suggested because the solar panels are not directly next to the airstrip which
might interfere with the safety for incoming flights. The rough distance cable required to
connect this into the grid is show below:
29
Figure 17: Sattelite picture of the facilities around the airport. The white lines are the suggested pathway for the transmission cables
between facility and solar production.
This path does not cross any airstrips which could potentially shut down parts of the
operation during installation. As shown in Figure 17 the cable required would need to be
roughly 600m. The cable required would be something like a 33 kV 3x70 XLPE cable.
Because the transmission line has a phase-to-phase voltage of 34,5kV a power transformer
would be necessary to connect the transmission line to the LV grid. This would require a
transformer with a turn ratio of 86,25. Because the grid cables have a rated power of
2217kVA and the total power generated from the PV plant is around 3,5MW connecting
the PV plant directly into the LV line would overload the cables. This is under the
assumption that all the excess power that is not used will be exported to the mains through
the LV network. A solution would be to connect the LV line for use in the LV system and
designate a new power transformer to export the excess power to the mains.
5. Discussions:
5.1 Evaluating possible placements of the future solar plant:
Placement of the solar power plant impact the efficiency, cost and complexity of the
network since it influences the distance that the power has to travel for delivery to the
30
consumption point. In general placement of any generator should be as close to the
consumption as possible to reduce losses in transmissions.
This project faces difficulties about the placement of the ESS and solar power plant due to
the fear of EMI to the airspace communication devices on the premise of the airport. This
concern come from a statement made by LFV where facilities with the risk of producing
EMI would need to be situated a distance away from the airport. LFV have strongly
recommended any solar installation to be situated 3km away from the airstrip. [19].
According to the airport the strong recommendation by LFV can be argued against by the
client but they would need to prove that no significant EMI could occur from the facility
which would mean that the facility would have to be constructed to be assessed. Situating
the solar power plant 3km away would require extra implementations such as:
- A 3km long cable to connect both the solar park and the airport
- Purchasing or leasing land for the construction of the solar park and acquiring the
land necessary to connect both facilities.
- An extra high-power transformer specifically to take down the high transmission
voltage down to the lower voltage levels.
𝑆3𝑝
∗ 𝐿 ∗ √𝑅2 + 𝑋 2 = ΔVLN
𝑉𝐿𝐿 ∗ √3 (26)
Where S3p is the total power transmitted through the cable, VLL is the line-to-line voltage,
L is the length of the cable in kilometres, R is the maximum AC resistivity of the cable per
kilometre, X is the maximum reactance of the cable per kilometre and ∆VLN Is the single
phase voltage drop in the cable.
Using data about the 33 kV 3x70 XLPE cable gives the following values comes out [20]:
106
3.5 ∗ ∗ 3 ∗ √0.062 + 0.1012 ≈ 21𝑉
34,5 ∗ 103 ∗ √(3)
This would mean that a 21V line voltage drop would equate to a phase-to-phase voltage
drop of around 36,4V. Using Elsäkerhetsverkets regulation about voltage differences its
stated that a voltage difference between ±5 nominal voltage is accepted, a 36,4V difference
31
on a 34,5kV would be approximately a 0.11% voltage difference which is acceptable.
The electrical losses due to the additional cable would be:
𝑃3Φ
𝐼𝐿 =
√3 ∗ 𝑉𝐿𝐿 (28)
3,5 ∗ 106
𝐼𝐿 = ≈ 58,571𝐴
√3 ∗ 34,5 ∗ 103
This would be a miniscule power loss when comparing the power transmitted which is
around 3.5MW.
While a 3km transmission line would be feasible it would increase the cost of the project
due to the increased cost of additional land and equipment. The recommended placement
of the solar plant would as close to the facility as is feasible because of the following
reasons:
- The area around the airport is owned by the airport itself which then requires no
extra steps to own the land.
- The area around the airport is already flattened which reduces the work necessary
to level the ground to be used in the solar plant which reduced the cost.
- Reduces the work and cost required for installing the transmission lines.
- Increases the ease of expanding the solar power plant for future load needs.
32
5.3 SAM:
SAM is a simulation tool developed by NREL for use in modelling future renewable energy
production and storage. This system has access to modelling for a variety of renewable
energy and storage systems such as photovoltaics, wind power, battery ESS and fuel cells.
[21] SAM allows for a variety of sizes of the systems ranging from small rooftop
installations used for residential load supplying to large scale power generation and storage
used for power quality and load shifting.
SAM allows for both evaluation on the economical as well as the technical parts about the
future system using the specification of the future system provided by the user.
The information provided through the simulation includes general information about the
systems performance and cost, this includes the annual energy production, the systems
capacity factor, energy yield, total cost, etc.
Alternatively, SAM allows for a modelling system where the economical aspect is absent
and instead looks at the performance of the system as a power generating facility.
This project will be using the non-financial photovoltaic simulation model as this project
is mainly focused on the technical requirements to be able to supply energy to the facility
and not a detailed combined techno-economical study.
33
5.4 Future grid design:
5.4.1 Design of the PV system:
In solar power plants the arrangement of the solar panels and power inverters is called
architecture. Architectures vary the connections between the solar panels and inverters and
the configurations of these.
There are 4 main architecture designs when designing solar plants: [22]
34
inverters and a loss of reliability because all power production goes through a single
inverter.
String:
In this configuration each string of solar panels is equipped with its own MPPT inverter.
Because each string has its own MPPT this leads to low losses due to voltage mismatch
between the strings, leading to high roundabout efficiencies and removing the necessity of
a string diode in each string. One of the drawbacks could potentially be a higher cost
because each string requires its own inverter.
AC module:
In this configuration every solar panel is equipped with a MPPT inverter that is then
connected into a DC/AC converter.
This design lends itself to a high efficiency as there is no mismatch losses and each solar
panel can operate at its MPP at all times, has the highest reliability since failure of any solar
panel or its inverter does not disturb the production of any other panel and increased
flexibility for additional solar panels because installation just requires the addition of the
solar panel and its power inverter.
The highest drawback to this design is its high installation cost and losses due to high
voltage boosting which makes this configuration unsuitable for large scale power
applications.
Multi-string inverter:
This converter combines the design of the central inverter and string module architecture
by including a MPPT inverter on each string of solar panels and guiding it into a central
power inverter. This design combines the advantages of the central and string configuration
leading to high efficiency and low mismatch losses. This design does increase the
installation cost of the solar plants and lower reliability as in the case of the central inverter.
As mentioned in 5.1 one additional concern is the EMI emitted by inverters. This leads to
decreasing the number of inverters present in the system as a priority to avoid the risk
interference in the communication technology around the airport. For this project the
architecture of choice will be the central inverter type because of the minimal number of
35
inverters. If the danger of EMI were not to be considered the multi string inverter
configuration would be considered because of the increased efficiency.
36
between each of the systems inverters. This would look like the following:
Figure 19: Seperate ESS and power production. The systems could be linked together anywhere in the AC system.
This configuration lends the ability for the two systems to be allocated at different locations
and increases reliability since each of the systems works through separate inverters. The
main disadvantages might be an increased cost due to the additional components required
and it might decrease the roundabout efficiency of the system because of the additional
component required in the system.
Another way to connect the systems together is to link the DC busses of both the power
production and storage together and binding the shared DC bus to a shared inverter:
This system might lend itself to a higher efficiency and a decreased component cost. But
here the reliability of the system decreases because the production and storage are linked
37
to the same power inverter. This could be offset by having multiple inverters at the same
location, but this might increase the cost, maintenance and risks reducing the efficiency of
the system due to the potential of having the inverter operate at a low power loading which
in turn reduces its efficiency.
A centralized system would mean that independent of how the system is interconnected
with a power production most of the stored energy would occur in one central hub. This
system has the advantage that because less component is required the need for extra
equipment or maintenance decreases which in turn makes this a more economical option.
But this has an impact on the reliability of the system since a failure in one of the
components could lead to a complete shutdown of normal operation.
Figure 21: A single central ESS that supplies all parts of the AC system.
A decentralized system would mean that the energy storage systems are not centralize to
one location but spread out to several locations. This could mean that multiple smaller
energy systems are distributed to important loads while the bulk of the energy storage could
still be located at a larger energy system somewhere else. This system has higher reliability
that depends on the amount of distributed ESS since failure of a single ESS would not
38
completely stop any other ESS. The drawback of this system might be an increased cost
and maintenance required to keep the system operating.
Figure 22: Multiple ESS exists. In this configuration the DESS supply important loads during high power demand periods.
Microgrid:
Micro grids are defined as a group of interconnected loads and distributed energy resources
within clearly defined electrical boundaries that acts as a single controllable entity with
39
respect to the grid.
A microgrid can connect and disconnect from the grid to enable it to operate in both
grid-connected or island-mode [24]. This could effectively reduce the carbon emissions
from both the primary consumer but also lend to reduce the carbon footprint of a secondary
consumer as well. The idea of microgrid use for renewable energy resources have in recent
times become a popular suggestion for both increasing the efficiency of the renewable
energy reserve but also in use to assist the local electrical grid with power during peak load
hours. This could potentially help to stabilize power quality with the help of local load
shifting and use in voltage regulation. If this system incorporates energy storage systems
as well the grid owner could request a certain amount of energy for the highest peaks thus
leading to decreased stress in the power grid. This implementation would increase the
construction costs but could overall increase profits because of increased control over when
to export energy into the mains.
Because of this project aim to implement its own energy generation and storage this system
could benefit form integrating microgrid infrastructure and technology.
The chargers would be connected into the new DC network while the existing AC LV will
be bridged with a DC/AC converter.
Combining the points made in the sections above into a proposed system would yield the
following system:
40
Figure 23: The one line diagram of the future system. The generator with a sun is the power production while the generator with a flame is
a storage system.
- Central inverter because of the decreased risk of EMI as required by LFV to follow
recommendation and not risk the PV system being deemed an EMI hazard.
- Storage system and power generation interconnected to increase the efficiency
between energy production and storage.
- The storage system will consist of a main large storage hub where the majority of
energy storage will occur but may have additional ESS at important loads. The
important loads consist of the chargers and to be able to supply the current LVP
with energy for at least two hours.
- The entire new part of the grid will be a separate microgrid where the power
production, storage and additional loads will be in this microgrid. The main
advantages might be an increased efficiency between the energy production and
storage as there are less conversions. Also having the system as a separate
microgrid makes it easier to expand the system in the future since it does not
require all operations to seize during the expansion.
6. Conclusion:
The main objective of this project was to investigate, evaluate and propose a possible
implementation of a future green airport. This project proposes using a pv system having a
total of 8,832 CSI Solar co. Ltd. CS3Y-470P solar modules arranged in a central inverter
architecture as this design when simulated can supply both the current facility and also able
to supply the addition of chargers into the system. Investigations of the current load profile
of Karlstad airport using the energy consumption profile, analysing specifications about
41
the current system and overlaying the additional consumption from the chargers into the
airports current load profile led to the conclusion that inserting the chargers directly into
the system with no modification would pose to great of a risk at overloading the current
facility. Additionally, this project investigated possible configurations of the future system
and lastly proposed a new future grid capable of producing and delivering the power
necessary to support the chargers and the current existing facility with the help of an ESS.
While this project has provided a suggested implementation the avenue of green airports
needs to be investigated to fully explore its potential as future green solution. To fully
explore the project like the one proposed a complete techno-economical study would be
required to explore the economic and technical implications for a system like the one
proposed, additionally because green airports could become significant energy producing
facilities studies about the power quality in a microgrid like the one suggested in this project
could be important because of their potential roles as power quality regulation centers and
large energy storage hubs. Lastly because of the large amount of energy produced and
exported in these types of facilities coupled with the suggestion of implementing a
microgrid into this future facility studies focusing on smart grid and smart energy market
management systems incorporated into similar facilities as these technologies could in the
future help in power quality management and more efficient power exporting algorithms.
42
7. Appendix:
43
44
Figure 24: Code Flowchart. This code is what generates Figure 6.
45
8. References
[5] A. D. Theocharis, J. Milias-Argitis and T. Zacharias, "A systematic method for the
development of a three-phase transformer non-linear model," International Journal
of Circuit Theory and Applications, vol. 38, no. 8, pp. 797-827, 2010.
46
[9] P. Aupke, A. Kassler, A. Theocharis, M. Nilsson and M. Uelschen, "Quantifying
Uncertainty for Predicting Renewable Energy Time Series Data Using Machine
Learning," Engineering Proceedings, vol. 5, no. 1, p. 50, 2021.
[14] O. Nyquist and P. Åberg, "SVT Nyheter," SVT, 13 August 2021. [Online].
Available: https://www.svt.se/datajournalistik/solligan-2021/. [Accessed 10 Mars
2022].
[17] B. Mow, "STAT FAQs Part 2: Lifetime of PV Panels," National Renewable Energy
Laboratory, 23 April 2018. [Online]. Available: https://www.nrel.gov/state-local-
tribal/blog/posts/stat-faqs-part2-lifetime-of-pv-panels.html. [Accessed 30
September 2022].
[18] N. Chen, K. Lu, W. Zhao, L. Wang and T. Dai, "Overload Analysis of Distribution
Transformers Based on Data Mining," Conference Series: Materials Science and
Engineering, vol. 439, no. 3, p. 1, 2018.
47
[19] K. Fors, S. Linder and T. Ranström, "Radiostörningar från solcellsanläggningar –
Kartläggning av störningsproblematik i Sverige och," Totalförsvarets
forskningsinstitut (FOI), Stockholm, 2020.
[23] An airports’ need of change to go 100% green using an energy storage system and
solar power, Karlstad: Karlstad University, 2021.
48