GEOMATICS AND ENVIRONMENTAL ENGINEERING • Volume 16 • Number 4 • 2022

https://doi.org/10.7494/geom.2022.16.4.31

Lidiia Davybida1, Ireneusz Wyczałek2, Artur Plichta3

Using GIS and SDSS Tools

in the Design of a Photovoltaic System
for a Built-up Roof

Abstract: The design and installation of solar panels on the roofs of urban buildings often
require consideration of the specific spatial conditions that affect their efficien-
cy. The primary purpose of this work is to develop a procedure for designing
and optimizing photovoltaic installations using geomatics methods and specif-
ic tools of GIS and CAD systems. The roof of the historic building A2, which
is a part of the Poznań University of Technology campus, was selected as the
tested object. Solar radiation modelling and determination of suitability zones
were performed using SEBE (Solar Energy on Building Envelopes) in QGIS.
Possible options for the placement of photovoltaic modules on the roof were
simulated with CAD technique in the web-based HelioScope software. The re-
sults of the simulation show that the current roof area can generate electrical
power of 99.9 MWh/year. The proposed methodology is universal for photo-
voltaic installations on built-up roofs and can be applied to other buildings
and, consequently, the results obtained can be used to improve the content of
the solar data urban geoportal.

Keywords: solar irradiance, historic building, shadow effect, obstructions, geospatial as-
sessment, roof-top photovoltaic system

Received: 13 February 2022; accepted: 5 July 2022

© 2022 Authors. This is an open access publication, which can be used, distributed and repro-
duced in any medium according to the Creative Commons CC-BY 4.0 License.

1
Ivano-Frankivsk National Technical University of Oil and Gas, Department of Geotechnical Safety
and Geoinformatics, Ukraine, email: [email protected],
 https://orcid.org/0000-0002-9796-7124
2
Poznan University of Technology, Institute of Civil Engineering, Poland,
email: [email protected],  http://orcid.org/0000-0003-3963-8186
3
Poznan University of Technology, Institute of Civil Engineering, Poland,
email: [email protected],  https://orcid.org/0000-0001-6134-290X

31
32 L. Davybida, I. Wyczałek, A. Plichta

1. Introduction
1.1. Solar Photovoltaic (PV) Technology
By the start of the 21st century, 160 years after A.E. Becquerel first demonstrated
the phenomenon of photoelectricity and 60 years after the first commercial solar cells
were produced [1, 2], photovoltaics (PV) had found its rightful place as a source of
electricity, ranging from small electronic equipment and night lighting [3] to powering
satellites, space stations and spacecraft. Along with their technological development,
photovoltaic cells have become an increasingly important alternative source of pow-
er for homes and household installations [4], even generating electricity surpluses
which have now been taken over by institutions that manage energy distribution. In
the era of climate protection, solar energy has become an important component of the
energy mix for individual countries. Currently, there are three groups of distributed
electricity producers using PV technology: individual pro­ducers for their own needs
(island-type power systems), individual prosumers (produc­ers discharging surplus
energy to the open electrical grid) and photovoltaic farms (power plants – PVP) [4].

a) b) c)

Fig. 1. Photovoltaic panels in an anthropogenic landscape: a) on public utility building;

b) on the roof of residential house; c) on a frame on the ground
Source: author ownership (photos from 26.11.2021)

The conversion of solar radiation into electricity takes place in solar cells as
a result of the photovoltaic effect, i.e. the flow of electrons between the elements
of the cell type “n” and “p” due to the excitation of valence electrons with quan-
ta of solar radiation [5]. The current method of PV installation (Fig. 1) is mainly
based on monocrystalline (or, less frequently, polycrystalline) quartz cells combined
into modules with dimensions of approx. 1.0 × 1.6–1.7 m, which are assembled into
panels [2, 6]. Most often, one module consists of several dozen (e.g. 60, 72, ...) cells
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 33

collectively (thanks to the series connection) collecting solar energy and transmitting
the received electricity through an inverter to the electrical installation or the batter-
ies. Modules can be divided into 2–3 sectors operating independently [6], however
one-piece solutions predominate. Batteries of PV panels are installed in new public
buildings (Fig. 1a), on residential houses (Fig. 1b) or in their close vicinity (Fig. 1c).
The efficiency of typical PV installations based on silicon solar cells ranges from
about 14% to over 22% (at 25°C) [3, 7] and depends on static factors affecting the
panels (latitude, shape and orientation roof, topography and its coverage) and dy-
namic (duration and intensity of sunlight, air temperature, humidity, cloud cov-
er, etc.) [8–12].

1.2. Factors Taken into Account

for the Location of the Panels
The authors of various publications focused [8–10, 13–19] on the design of pho-
tovoltaic installations take into consideration different analyses and configurations
of the components of these analyses.
Romero Rodríguez et al.  [10] listed most of them in the form of a flowchart,
divided into two groups of factors:
1) solar suitability:
– orientation,
– roof slope,
– separation of photovoltaic panels;
2) architectural suitability:
– construction restrictions,
– protected buildings,
– shadows effects,
– service area.

The calculation of the amount of solar energy reaching the roof is the basis of
the design of a photovoltaic installation. This value depends on the angle of inci-
dence of sunlight (geographic coefficient) and the temporal weather conditions (at-
mospheric coefficient). Other factors that were omitted here, such as those resulting
from local law, environment protection plan or the possibility of mounting panels
on the ground, should also be taken into account in the process of designing photo-
voltaic installations [20].

1.3. Using IT Resources in the Design of

Photovoltaic Installations
The tools enabling spatial analysis of the relevant factors are SDSS (Spatial De-
cision Support Systems) procedures implemented in GIS or CAD systems. They are
mainly used in strategic or conceptual analyses when determining the locations of
34 L. Davybida, I. Wyczałek, A. Plichta

the optimal PV installation on groups of buildings, housing estates or other territo-

rial units to determine the energy potential of these areas or groups of facilities [8, 9,
13, 14, 19–24]. Meanwhile, to locate panels on specific roofs, specialized CAD soft-
ware is usually used to perform analyzes on vector models of a given roof [7, 25–27].
In recent years, specialized programs have emerged and are still being developed
(GIS or CAD-type) to assess the insolation of areas and objects, as well as to design
solar panels of various scales. Specialized applications are implemented in commer-
cial environments (Area Solar Radiation in ArcGIS, PVCAD in AutoCAD) and freely
distributed (UMEP (Urban Multiscale Environmental Predictor) in QGIS, Module
Potential Incoming Solar Radiation in SAGA) but also separate programs (PVGIS,
PVSyst, Aurora, Homer, HelioScope etc.).
Nowadays particular attention should be paid to the development of web tech-
nologies and cloud geodatabases which enable the use of interactive tools imple-
mented in specialized geoportals to assess the insolation of the area, determine the
optimal panel characteristics (angle, azimuth), model and forecast energy produc-
tion and assess the payback period of installations.
An analysis of publications focused on the design and optimization of PV sys-
tems allows the identification of two main approaches that are determined by the
scale of research – ground-mounted installations or solar farms within a country
or region are considered on the macro-scale level, whereas solar panel installa-
tions on building roofs within a city, district or individual object are considered
on the micro-scale level. Appropriate PV installation site macro-scale analysis
is usually carried out by multicriterial GIS analysis associated with a variety of
relevant factors (climate, land surface slope, aspects, distances from building ar-
eas, roads etc.) [8, 9, 11, 19–22]. The micro-scale analyses are primarily based on
the spatial variation of incoming solar radiation on a building rooftop surface.
This type of analysis needs full 2.5–3D geometrical representations of buildings
(LOD 2 and LOD 3 building models) and CAD tools for optimal location determi-
nation [10, 16, 26–31].
The final step of the majority of such projects is to show results on the Web-
GIS platforms or specialized geoportals, where everyone can access the information
about solar energy potential and supplement data with interactive maps and mod-
els [20, 29, 31]. As examples, we can cite solar data portals and cadasters of differ-
ent cities [32–39], including the Portal of the Spatial Information System of Poznań,
where PV potential and suitability of roofs are visualized [33].
A specific issue for many urban areas is the integration of PV  installations
on historic buildings. As to [40, 41], it is noted, that the design and installation of
a PV system need to be carefully considered so that its efficiency can be maximized
whilst avoiding damage to the significance of the building, its fabric and its setting.
Supplementing the solar cadastre information block with data on restrictions for the
PV installations for historic buildings is appropriate for urban spatial planning and
energy management.
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 35

1.4. Problem Definition

A real problem related to the installation of photovoltaic panels on historic
buildings is planning their placement to ensure an appropriate (maximum) level of
power, simultaneously limiting the visibility of the panels for an observer located in
public spaces. Objects that require this kind of approach include public utility build-
ings that were built in the 1960s and 1970s – nowadays old enough to be classified as
protected buildings due to their historic character. It is generally accepted that pho-
tovoltaics should not be installed on such kind of facilities. However, many of them
are covered with flat roofs, so they are suitable for mounting various installations
invisible from the level of the road and sidewalk. Usually, different installations are
already mounted on the roofs (i.e. chimneys, fans, external air conditioning units...),
which means that the effective placement of a sufficiently large number of photovol-
taic panels requires additional spatial analysis.
The purpose of the research is to develop a procedure for designing and opti-
mizing photovoltaic installations using geomatics methods and specific tools of GIS
and CAD systems that allow the character of insolation for different parts of a roof to
be taken into account, as well as certain spatial constraints on the specific roof of the
historic building A2, which is a part of the Poznań University of Technology (PUT)
campus.

2. Methods and Materials

2.1. Tested Object

Within the practical experiment, a project was made of the positions of PV pan-
els on the roof of the abovementioned A2 building of PUT, erected in 1953–1955 and
now part of the Warta campus (Fig. 2). Poznań is located in the flat, central part of
western Poland, which is characterized by a relatively low level of solar potential
(1060–1090 kWh/m2/year, according to information obtained from the Solargis Glob-
al Solar Atlas 2.0 platform [42]) (Fig. 2а). This, in turn, affects the level of insolation
of the roofs of individual buildings.
According to the SIP Poznań municipal geoportal  [33], the insolation of the
buildings’ roofs in Poznań varies from 3 kWh/m2/year to 1050 kWh/m2/year (Fig. 2b).
These data also show that the tested object is one of the most promising for the
installation of solar panels (Fig. 2c) because the insolation of the considered roof
ranges from 800  kWh/m2/year to 1000  kWh/m2/year (except for the shaded parts)
with the limitation of the sunlight amount by cloudiness, fog and air pollution with
dust and aerosols.
The south wall of the study building is oriented at an angle (azimuth) of 211.6°,
which should be taken into account in the analysis of the impact of lighting on the
efficiency of PV installations.
36 L. Davybida, I. Wyczałek, A. Plichta

a) b)

c)

Fig. 2. Location of the research object and general conditions of solar radiation:
a) zoning of the territory of Poland according to the GHI insolation index; b) location of
the tested object in the city of Poznań and classification of roofs of city houses according
to the level of solar potential; c) the subject of research, i.e. building of the Faculty of Civil
Engineering and Transport (FCE building) against the background of the orthophotomap of
Poznań, taking into account the insolation of the roofs
Source: own study based on available data [33, 42, 43]

2.2. Proposed Methodological Framework

Since designing and optimizing photovoltaic installations is a complex prob-
lem and different criteria apply, a combination of GIS and SDSS methods was used
to perform the analysis. The steps are discussed below according to the proposed
framework as shown in Figure 3. The methodology is as follows:
1. Data collection and pre-processing, which include the spatial and temporal
data collection and data standardization processes.
2. Image processing and spatial analysis, which include geoinformation analysis
involving several methods and tools, e.g., solar irradiance analysis, distance
analysis, raster zonal statistics, geometric and overlay operations, visualization.
3. Installation design, which is conducted to create different options for the
photovoltaic installations and their performance analysis.
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 37

Fig. 3. Flowchart of the research methodology

38 L. Davybida, I. Wyczałek, A. Plichta

Data collection and pre-processing

Spatial and temporal data, the main characteristics of which are shown in Ta-
ble 1, were used to perform the steps of the framework and create a graphical visu-
alization of the research results.
Detailed (with 15 cm resolution) orthographic imagery, obtained from a drone,
and generated digital surface model (DSM) were used as the base dataset consisting
of ground and building heights and deciding the latitude and longitude used for the
calculation of the position of the Sun. Raster and vector datasets of topographic and ad-
ministration data were downloaded and transformed to the single projection into QGIS.
The variability of the dominative dynamic factors for the studied area of the
city of Poznań, obtained for the period  2013–2021 with the use of Copernicus
ERA5 ECMWF global monitoring data, USGS Landsat 8, is shown in Figure 4 pre-
pared with Google Earth Engine API tools [44] (Fig. 5).
The analysis results, compiled in the form of autocorrelation function graphs,
confirmed that the amount of solar radiation and air temperature are characterized
by distinct seasonality while cloud cover and precipitation are broadly less regu-
lar (Fig. 6).
Meteorological data at hourly temporal resolution were transformed into the
special format used in SEBE (Solar Energy on Building Envelopes), incorporated in
UMEP (Urban Multi-scale Environmental Predictor), a plugin for QGIS [45].
Image processing and spatial analysis
Solar radiation model SEBE allows estimates of solar irradiance on building
roofs and walls using an algorithm with a DSMs (and derivative rasters, such as wall
height and wall aspect) and the solar position to generate pixel-level information
of shadow or sunlit areas [45, 46]. The result added to map canvas is the horizontal
radiation, i.e. irradiance (kWh/m2) on the roof and ground around the tested object.
The suitability of the tested object’s rooftop has been already confirmed, and theo-
retical solar power potential has been calculated [33]. Therefore, the main task for
this step is to identify suitable and unsuitable zones for PV installations within the
rooftop using raster zonal statistic tools and taking into account the fact that rooftop
surfaces should receive at least 800 kWh/m2 in solar radiation if solar panels are to be
installed. Areas with low solar radiation have to be removed. Modelling the build-
ing rooftop with obstructions is also performed using QGIS.
A special unit of calculations, which allows taking into account the individual
geometric features of the tested object, provides an assessment of the location and mu-
tual shading of the panels, determining the visibility of panels from pedestrian areas.
The obtained results are used both in the GIS environment to create additional vector
layers of the unsuitable zones for solar panel installation, and in the CAD environment
to set the parameters of the installation. As a result, the layers, where it is not possible
or not appropriate to place the panels for various reasons, are combined into the final
layers of obstructions, which will automate the design based on the CAD technique.
 Table 1. Spatial and temporal datasets

Temporal Spatial
Dataset Source Format
aggregation resolution/Scale

Global Horizontal The World Bank, Source: Global Solar Atlas 2.0,

Irradiation in Poland annual average ~250.0 m Solar resource data: SolarGIS, 2019 raster
[kWh/m2] https://globalsolaratlas.info/download/poland [42]

Portal of the Spatial Information System of Poznan, 2020

Solar potential [kWh/m2] annual cumulative 1.0 m raster
http://wms2.geopoz.poznan.pl/geoserver/solarne/wms [33]

Sun exposure of roofs Portal of the Spatial Information System of Poznan, 2020
annual cumulative 0.5 m raster
[kWh/m2] http://wms2.geopoz.poznan.pl/geoserver/solarne/wms [33]

Geoportal Poland, 2021

0.5 m
Digital Surface Model [m] – https://mapy.geoportal.gov.pl/imap/ [43] raster
0.15 m
Author ownership

Geoportal Poland, 2021 vector

Buildings – 1:1000 https://integracja.gugik.gov.pl/cgi-bin/ (multipolygon),
KrajowaIntegracjaEwidencjiGruntow [43] .shp

Copernicus ERA5 ECMWF, 2021

Meteorological data: https://developers.google.com/earth-engine/datasets/catalog/
– solar radiation [kWh/m2] monthly cumulative ECMWF_ERA5_MONTHLY
– air temperature [oC] monthly average 30 m USGS Landsat 8 Collection, 2021 .csv
– precipitation [mm] monthly cumulative https://developers.google.com/earth-engine/datasets/catalog/
– cloudiness [%] monthly average LANDSAT_LO08_C01_T1_RT
Post-processed by Google Earth Engine [44]
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 39
40 L. Davybida, I. Wyczałek, A. Plichta

a)

b)

c)

d)

Fig. 4. Graphs of dynamic factors influencing the power of solar electricity in Poznań:
a) amount of solar radiation; b) air temperature; c) precipitation; d) cloud cover
Source: own study based on available data [44]
Fig. 5. Downloading data from global climate monitoring systems
using the Google Earth Engine API
Source: own study based on available data [44]
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 41
42 L. Davybida, I. Wyczałek, A. Plichta

Fig. 6. Graphs of the autocorrelation function and variability of factors over the year
(calculated for observational data in 2013–2021)
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 43

Installation design
The next step is import the file of obstructions into HelioScope Advanced Solar
Design Software, which is an online app for designing and engineering of rooftop
photovoltaic (PV) systems. The software provides access to a huge database of dif-
ferent solar panels and other components of PV systems and allows the panel angle,
orientation, raw spacing and azimuth to be changed automatically.
Due to the current wide range of solutions for photovoltaic installations, we
limit our overview to a few basic pieces of information about them influencing the
course of this study:
– photovoltaic roof installations consist of modules (i.e. sets of cells) connected
to panels [1, 6] properly arranged on the roof, then electrical installations
connect the panels in series, parallel or hybrid wiring, and transfer voltage
to inverters converting the generated DC current into AC used by electricity
network installations [4];
– modules consist of a regular grid of PV cells made of silicon – mono- or poly-
crystalline, or linked in pairs of minerals from blocks III and V of the table
of chemical elements, or of other combinations of minerals, including rare
minerals [20];
– the panels are combined with metal profiles and placed directly on the roof
or mounted on racks – fixed or rotating towards the Sun (heliotropes).

Monocrystalline quartz panels predominate in Poland, occupying about 60%

of the market – they are most effective concerning the price  [4]. For this reason,
they were taken as the basis for the spatial analyzes presented here – one-piece
monocrystalline SunPower Maxeon 3 (400 Wp) modules [47] with dimensions of
1690 mm × 1046 mm and efficiency of 22.6% were used (in the case of using the
60 kW 3 F 90 A Suntrio Plus inverter).
Simulations of PV  installations with different orientations and tilt angles are
performed for the parameters of the selected model of solar panels. Finally, the cal-
culation of the expected system energy losses and performance analysis of the pho-
tovoltaic installations have to be conducted.

3. Results

3.1. GIS-based Estimation of Solar Irradiance for the Roof

Figure 7 shows the results of the solar radiation analysis for different places of
the roof in the two extreme visible positions of the Sun. The daily insolation here
ranges from 0.01 kWh/m2 in winter (Fig. 7b) to about 5.7 kWh/m2 in summer (Fig. 7a).
The summary annual solar radiation is approximately 750–1045 kWh/m2 (Fig. 7c),
except for shaded roof fragments.
44 L. Davybida, I. Wyczałek, A. Plichta

a)

c)

b)

Fig. 7. The daily sum of solar radiation reaching the roof surface during the days of
summer (a) and winter solstice (b) and the annual solar radiation reaching the surface (c)

Parts of the roof where the panels can be placed have been identified, given the
level of insolation, existing obstructions, and shaded areas. These data not only al-
low the amount of solar radiation and shading of some roof surfaces for a particular
day or period to be determined but also to conduct an accurate analysis of existing
superstructures, ventilation systems, chimneys, antennas and others, which consti-
tute obstructions to installing solar panels and sources of additional shading. To
eliminate the existing obstructions and the most shaded areas, a raster classification
of the total annual insolation was carried out on the principle: <800 kW/m2/year – the
area is unsuitable for the installation of solar panels; ≥800 kW/m2/year – the area is
suitable for the installation of solar panels. Given the shape and position of the roof
of the lower part of the building (6.6 m in height) relative to the upper part (15.0 m
in height), this one was excluded from further analysis as unpromising.
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 45

3.2. Arrangement of Panels

As a rule, designs for solar roof installations assume their arrangement on racks
located directly on the roof slopes, rarely on a frame at a certain angle to the roof
covering. The first solution relates mainly to slope roofs, and the second one – to flat
roofs (according to the “wind standard” PN-77/B-02011/Az1:2009, it was assumed
that these roofs have a slope of up to 5°), thus this value was omitted in further anal-
ysis. In the second solution, two approaches are presented in the literature:
1) Panels placed on frames with an angle of inclination optimal due to the
recovery of solar energy – for medium latitudes it is usually between 39°
and 41°. The optimal value for the tilt angle was calculated by an application
from the geographic information portal PVGIS [48], which also provides ac-
cess to databases of meteorological data and solar radiation based on three
components of lighting – direct, diffused and reflected in clear skies under
the real global lighting conditions for flat as well as sloping surfaces.
2) Panels folded flat on the roof or at a slight angle, 10° to 15° inclination. This
solution is supported by limiting the risks to the roof structure due to wind
gusts, as well as the unequal load of the roof with the PV installation by
itself. Moreover, the smaller the angle of inclination of the panels, the great-
er the possibilities of their orientation (facing) – while the inclined surfaces
must face south (or turned only a little to the sides), the horizontal ones can
be oriented completely arbitrarily. Thus, the inclination (and the angle of
orientation) of the panels is the first criterion considered for the arrangement
of PV installations on flat roofs.
At the latitude corresponding to the location of the tested building (φ ≈ 53.4° N),
the Sun rises above the horizon from 13.2° at noon on the winter solstice (Decem-
ber 21) to 61° in summer (June 21) [49]. The panels have a mounting dimension of
about 1.0 m × 1.7 m, so when they are oriented with their long side at the base and
with the slope of 15°, the opposite longer side will be got up to 0.26 m. Such a setting
is favourable due to the influence of wind, but it can be problematic in the case of
significant snowfall. Snowfall and the residual snow cover after it solidifies may
require periodic maintenance. When the panels are placed at an angle of 40°, the
elevation of the back edge relative to the front edge is 0.64 m.

3.3. Shadows
Direct (or indirect, diffused light) illumination of the entire surface of the pan-
el is the basis for the effective operation of a PV installation. Incomplete lighting
results in the almost complete shutdown of the entire chain (series connection) of
panels [50]. Therefore, it is necessary to arrange the panels in such a way that will
avoid their mutual shading and influence on other installations sticking out from the
roof. Mutual shading analyzes generally apply to panels installed on a sloping frame
on flat roofs (panels lying flat on the roof surface do not cast a shadow). Usually, the
46 L. Davybida, I. Wyczałek, A. Plichta

shadows are analyzed at extreme positions of the Sun concerning the Earth’s sur-
face, i.e. on the summer solstice (June 21–22) and winter solstice (December 21–22),
and sometimes also on the equinox days. The apparent position of the Sun at noon
local time is also considered. The relative position of the Sun to the selected place
can be calculated based on quite simple geometric relationships or using one of the
internet applications available. Figure 8 shows a graph of the apparent movement of
the Sun in relation to the test object (coordinates 52.4N and 16.9E), as presented on
the website of the University of Oregon, USA [49].

Fig. 8. The apparent movement of the Sun in the sky for the location of the tested object
on selected days of the year. The highlighted (darkened) first 10° of the height of the Sun
above the horizon corresponds to the minimum efficiency of electricity production
by the PV installation
Source: own study based on [49]

Converting the results from Figure 8 to the length of shadows (the function
d = h/tan α, where d – length of the shadow, h – the height of the obstacle, α – the
angle of incidence of sunlight), we obtain an illustration of the shadow variability in
subsequent hours of local time in selected and characteristic days of the year (Fig. 9).
For an object with a height of h = 1 m, the longest shadow at noon on December 21
is almost 4.0 m long, and at 9:00 or 15:00 reach almost 12.0 m. Limiting the operation
time of the panels to the lower limit of the angle of incidence of light (i.e. 10°), we
obtained the shadow, which reaches 5.8 m. Taking into account the orientation of
the Sun to the panel line, the maximum shadow lengths of the object of 1 m height
are 4.0 m and 9.1 m respectively, and for an angle of 10° above the horizon is 5.1 m.
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 47

Fig. 9. Shadows projected by a vertical element (gnomon) on a horizontal roof

due to sunlight on selected days of the year. The distance sh is the length of
the shadow calculated as a multiple of the height of the illuminated element (h)
Source: own study based on [49]

In the variants considered in this analysis (40° and 15° inclination of panels), it

corresponds to the following parameters of the height of the obstacle resulting from
the inclination of the panel: a) shadow length on December 21 at 12:00, b) shadow
length on December 21 at 10:00 or 14:00:
– 0.64 m (40°): a) 2.6 m; b) 3.7 m;
– 0.26 m (15°): a) 1.0 m; b) 1.5 m.

On the longest day of the year (i.e. June 21), most of the shadows are turned
towards the south and do not interfere with the installation, while at noon they are
0.55 m long.
Ultimately, it was concluded that for the effective use of the installation, panels
should be inclined at an angle of 15° and their rows should be moved from each oth-
er at about 1.5–2.0 m. In the case of inclination at the angle of 40°, the rows of mod-
ules should be moved at about 3.5 m. The above-given parameters are close to the
results obtained by specialized software used for designing photovoltaic systems.
For further analysis, these two variants of panel inclination were adopted (with cor-
responding sizes of spacing between the rows), as well as two versions of the ori-
entation of the rows of panels: a) in the east-west direction (φ = 180°) or b) under
the azimuth of the southern elevation of the analyzed building (φ = 211.6°). Further
calculations also took into account – according to the above analysis – shadows cast
by lofty roof installations.
48 L. Davybida, I. Wyczałek, A. Plichta

3.4. Visibility of the Panels

For the analysis, the extent of the panels from the roof edge was calculated to
ensure their invisibility from the surrounding public places. In the beginning, it was
assumed that for safety reasons the panels should be at least 1 m away from the
edge of the roof. With this assumption, the rear edges of the outer rows of panels are
moved away from the edge of the roof – similarly to the calculations of shadows –
with the value b = h/tan α (where b is the panel’s projection onto the roof surface).
For panels inclination angle of 15° b = 0.97 m, their rear walls are almost 2.0 m away
from the roof edge, and for the panel inclination angle of 40° – b = 0.31 m, and their
distance from the edge roof respectively 1.31 m.

Fig. 10. Standardization of the obstruction layers

For a three-storey building (with an attic, as in the case of the considered object),
the edge of the roof rises above the ground to the height of about 15 m. Therefore –
according to the above given trigonometric rules – panels 0.26 m in height, remoted
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 49

from the edge for 2.0 m are not visible from a distance of 100 m (for the observer’s
eye at the level of 2.0 m above the base of the tested building), while for panels
0.64 m in height, separated from the edge for 1.3 m, equals a distance of 26.5 m. To
hide these panels from the eye of an observer who is 100 m away from the building,
they should be at least 4.9 m away from the edge of the roof (their higher edge), and
for the distance of 50 m (the maximum actual distance for an observer) it is 2.4 m,
which corresponds to the offset of the front line of these panels from the edge of the
roof for 2.1 m – in our analysis, we will assume a minimum distance of 4.0 m to the
edge of the roof along the front wall, and 2.0 m to the edge of the southern wall, less
visible for longer distances. The remaining edges of the roofs are not visible at great-
er distances than the predetermined 26.5 m.
The above-calculated values were used as parameters for the design of the ar-
rangement of the PV installation on the roof of the tested building.
Based on the calculations described in the previous chapters, the heights of the
rear edges of the panels and the minimal spacing distance between modules were
determined for tilt angles 15° and 40° for two options of the orientation of panels:
south (azimuth 180°) and parallel the edge of the roof (azimuth 211.6°). The obtained
layers, including the perimeter security buffer zone, the visibility area of the panels,
and the shaded areas, were combined into two vector layers, Obstructions_15.shp
and Obstructions_40.shp, correspondingly for tilt angles 15° and 40° (Fig. 10).

3.5. Detailed Design of the PV Installation

At the next step CAD models were built for shadow analysis, calculation of
shading losses and energy performance in the PV system and assessment of general
financial indicators in the HelioScope Advanced Solar Design Software platform [51].
Primarily, after converting obstruction layers to the *.kml exchange format, they
were imported into the HelioScope environment to fill the KeepOuts information
block and for further design.
For both variants of the tilt angle, the possibilities of placing the modules strictly
to the south and parallel to the edge of the roof were designed (Fig. 11). The summa-
ry results of the assessment are presented in Table 2. Calculations and models allow
the determination of the possibility of installing a solar power plant with a capacity
of about 40–99 kW on the roof of the A2 building of Poznan University of Technolo-
gy with an estimated average energy output of 41–100 MWh/year.
The projected distribution of expected system losses and the distribution of en-
ergy production are shown in Figures 12 and 13, respectively. Shading losses are
minimized, their values are less than the losses due to the technical parameters of
the PV system, and are approximately 3% for the 40° tilt and less than 1% for the
15° tilt (Fig. 12). The created prognostication graphs of energy distribution (Fig. 13)
show a significant variability in the amount of energy produced during the year, due
to the impact of the natural seasonality of the meteorological factors.
50 L. Davybida, I. Wyczałek, A. Plichta

a) b)

c) d)

Fig. 11. Options for the photovoltaic installations for: a) tilt angle 15° (azimuth 180°);
b) tilt angle 15° (azimuth 211.6°); c) tilt angle 40° (azimuth 180°);
d) tilt angle 40° (azimuth 211.6°) (created using the HelioScope Software)

a) b)

c) d)

Fig. 12. Distribution of the expected system energy losses for: a) tilt angle 15° (azimuth 180°);
b) tilt angle 15° (azimuth 211.6°); c) tilt angle 40° (azimuth 180°); d) tilt angle 40° (azimuth 211.6°)
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 51

a)

b)

c)

d)

Fig. 13. Potential distribution of energy produced for:

a) tilt angle 15° (azimuth 180°); b) tilt angle 15° (azimuth 211.6°);
c) tilt angle 40° (azimuth 180°); d) tilt angle 40° (azimuth 211.6°)
52 L. Davybida, I. Wyczałek, A. Plichta

Table 2. Possibilities and results of the placement of modules (photovoltaic panels –

SunPower Maxeon 3 (400 Wp), inverter – Suntrio Plus 60 kW)

Row Annual Annual

Azimuth Nameplate Performance
Tilt [°] spacing Modules shading energy
[°] [kW] ratio [%]
[m] [%] [MWh]

180.0 1.5 226 90.4 0.8 85.5 91.4

15
211.6 1.5 248 99.2 0.9 85.5 99.9

180.0 3.4 97* 38.8 2.9 84.4 41.4

40
211.6 3.4 120** 48.0 3.1 83.9 49.9

* An inverter with a capacity of 40 kW was used due to the less number of panels.
** An inverter with a capacity of 50 kW was used due to the less number of panels.

4. Discussion

This publication presents the applied spatial analysis procedure for the selec-
tion and proper placement of photovoltaic panels on flat roofs, taking into account
the above-mentioned requirements (criteria) resulting from the need for safe usage
of the designed PV installations, as well as the elements of the roof structure and ob-
jects existing on them, and their shadows. Due to the historic nature of the building,
it was also necessary to take into account another criterion, namely hiding the panels
from observers at the street level. Assuming that the designer does not influence the
lighting conditions that get weaker with increasing latitude, an important task was
to find the best option to place the solar panels on the roof of the building to achieve
the maximum possible efficiency.
A review of previous research [6, 10, 12, 15–19, 24, 26–31] and existing practical
solutions [33–39] confirms the feasibility of combining GIS and CAD  tools (with
their decision-making modules), using high-precision spatial models to address is-
sues, concerning the design and optimization of photovoltaic systems, evaluation of
their effectiveness, planning and development of solar energetics.
The basis of the presented analysis is a spatial model of land cover developed
based on low-altitude aerial photographs (drone) correlated with the necessary data
available on internet portals [33, 43]. The analyses were preceded by a review of
the available PV panel systems and the design and implementation of studies for
rooftop PV installations gathered in the literature on the subject. Selected functions
of the programs mentioned in the text were used for the analysis. The final effect
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 53

was presented in the form of a 3D decision map showing the optimal distribution of
panels on the roof.
The best option allows the placement of the maximum number of solar panel
modules and at the same time, minimize energy losses due to mutual shading.
Considering the possibility of technology development is also essential. Cur-
rently, opportunities for development in the design of photovoltaics are mainly
considered in the technology of manufacturing cells. The technology of multi-tran-
sition cells stands out among the various solutions. In this technology, each of the
layers works independently and regulates productivity under the energy of solar
radiation, which reaches a single cell, depending on the current wavelength of light.
Moreover, silicon is being replaced by materials of greater efficiency and produc-
tivity. One of them is perovskite, a mineral consisting of calcium, titanium and ox-
ygen (CaTiO3). Investigations have shown that it absorbs much more solar energy
and in addition, is characterized by high conversion efficiency of solar radiation into
electricity, even 1/3 greater than silicon [52].
Another solution is to use polymers in which more electrons excited by light
move faster. This increases the efficiency of solar cells by as much as 15% [5]. Tech-
niques for increasing the amount of absorbed solar light using lenses printed on cells
with gallium arsenide are also being developed. Such a system will be much lighter
and relatively cheaper, it can be installed on old roofs that are sensitive to overload-
ing due to the reduced weight. In the field of miniaturization and load reduction, re-
search is being conducted on the production of amorphous [a-Si] or micromorphous
[a-Si/c-Si] silicon cells, which will be one micrometre thick. It is also possible to make
hybrid or even dye (DSSC) cells, enriched with a special dye, actively involved in
electricity production. When these solutions enter the production phase it will be
possible to cover the entire roof surface and facades [27, 30] of facilities such as the
A2 building of Poznan University of Technology with solar panels without fear of
their visibility from the sidewalk, mutual shading and consequently without previ-
ous comprehensive analysis for their effective placement.
Analysis of the content and interactive tools of existing solar data geoportals,
both for Poznan and other cities in Poland and Europe [32–39], shows the possi-
bility of developing and supplementing the geoportal with simulation results (ob-
structions, layouts of solar panel modules, graphs of potential energy produced)
that can be performed according to the proposed algorithm involving data with
greater spatial coverage. Input data can be obtained using drone photography or
LIDAR [28, 39, 40, 46].
An additional issue that needs to be addressed when the study area of the mod-
el is expanded concerns the development of special GIS plugins to automate routine
tasks, including the formation of unified files of obstructions, preparing them for
import, as well as integration of CAD tools in the GIS environment. The results of
modelling can be added to the solar data portal for the city and used in the tasks of
land and energy management.
54 L. Davybida, I. Wyczałek, A. Plichta

5. Conclusion

The purpose of the research was to use geomatics methods to place a photovol-
taic installation on a specific roof of a historic public building, which requires con-
sideration of certain spatial constraints. In addition to the standard placement crite-
ria, the state of visibility/invisibility of panels from the surrounding communication
spaces was taken into account. Methods for designing photovoltaic installations on
roofs using specialized SDSS software and data models were considered. Original
visibility and shading analysis were performed using GIS tools.
Based on the results of the analysis, the expected implementation effectiveness
of the designed solution was calculated. The obtained results confirmed the hy-
pothesis that the best solution was to install panels with a tilt angle of 15° parallel
to the south-eastern edge of the roof. This option is expected to generate 99.9 MWh
of electricity per year, while the least efficient solution is to install panels with an
angle of 40° and an azimuth of 180°, which will produce 41.4 MWh of electricity
per year. All parameters were evaluated for specific types of photovoltaic modules
and inverters.
Installing photovoltaic panels on the roof of a historic building may require
reinforcement, which should also be taken into account in such an analysis. It would
require a preliminary inventory of the load-bearing structure and the necessary
strength calculations.
The current study is limited by the spatial coverage of input data, which in-
cludes only one tested object. Expansion of the input database with the involve-
ment of aerial photography or LIDAR with high accuracy, as well as automation of
routine data processing tasks using specialized plugins, will create relevant models
with greater spatial coverage (district, city). The results of modelling, in their turn,
can be used to supplement and improve the functionality of the solar data portal for
the city.

Author Contribution
Lidiia Davybida: conceptualization, methodology, data curation, formal analy-
sis, visualization and writing – original draft.
Ireneusz Wyczałek: conceptualization, supervision, project administration, for-
mal analysis, visualization and writing – original draft.
Artur Plichta: methodology, validation, supervision, writing – reviewing and
editing. All authors approved the final version of the manuscript and agree to be
held accountable for the content therein.

Acknowledgements
This research was prepared with the support of the Centre for East European
Studies (University of Warsaw) in the frame of the Ivan Vyhovsky Award (2020/2021)
under the patronage of the President of Poland.
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 55

References

[1] Green M.A.: Silicon Photovoltaic Modules: A Brief History of the First 50 Years.
Progress in Photovoltaics: Research and Applications, vol. 13, 2005,
pp. 447–455. https://doi.org/10.1002/pip.612.
[2] Jastrzębska G.: Ogniwa słoneczne: budowa, technologia i zastosowanie. Wydaw-
nictwa Komunikacji i Łączności, Warszawa 2013.
[3] Soumya C., Deepanraj B., Ranjitha J.: A review on solar photovoltaic systems and
its application in electricity generation. [in:] AIP Conference Proceedings, 2396,
020011, 2021. https://doi.org/ 10.1063/5.0066291.
[4] Mirowski T., Sornek K.: Potencjał energetyki prosumenckiej w Polsce na przykła-
dzie mikroinstalacji fotowoltaicznych w budownictwie indywidualnym [Potential
of prosumer power engineering in Poland by example of micro PV installation in
private construction]. Polityka Energetyczna, t. 18, z. 2, 2015, pp. 73–84.
[5] Mughal Sh., Sood Yog R., Jarial R.K.: A Review on Solar Photovoltaic Technology
and Future Trends. National Conference on Recent Advances in Computer
Science and  IT  (NCRACIT). International Journal of Scientific Research in
Computer Science, Engineering and Information Technology, vol. 4, no. 1,
2018, pp. 227–235.
[6] Dondariya C., Porwal D., Awasthi A., Shukla A.K., Bhimte A.: Performance
simulation of grid-connected rooftop solar PV system for small households: A case
study of Ujjain, India. Energy Reports, vol. 4, 2018, pp. 546–553. https://doi.org/​
10.1016/j.egyr.2018.08.002.
[7] Bonomolo M., Di Lisi S., Leone G.: Building Information Modelling and Ener-
gy Simulation for Architecture Design. Applied Sciences, vol. 11, 2021, 2252.
https://doi.org/10.3390/app11052252.
[8] Yankiv-Vitkovska L., Peresunko B., Wyczałek I., Papis J.: Site selection for solar
power plant in Zaporizhia city (Ukraine). Geodesy and Cartography, vol. 69,
no. 1, 2020, pp. 97–116. https://doi.org/10.24425/gac.2020.131076.
[9] Yushchenko A., de Bono A., Chatenoux B., Kumar Patel M., Ray N.: GIS-based
assessment of photovoltaic (PV) and concentrated solar power (CSP) generation po-
tential in West Africa. Renewable & Sustainable Energy Reviews, vol. 81, 2018,
pp. 2088–2103. https://doi.org/10.1016/j.rser.2017.06.021.
[10] Romero Rodríguez L., Duminil E., Sánchez Ramos J., Eicker U.: Assessment of
the photovoltaic potential at urban level based on 3D City models: A case study and
new methodological approach. Solar Energy, vol. 146, 2017, pp. 264–275. https://
doi.org/10.1016/j.solener.2017.02.043.
[11] Khan J., Arsalan M.H.: Estimation of rooftop solar photovoltaic potential using
geo-spatial techniques: A perspective from planned neighborhood of Karachi – Paki-
stan. Renewable Energy, vol. 90, 2016, pp. 188–203. https://doi.org/10.1016/​
j.renene.2015.12.058.
56 L. Davybida, I. Wyczałek, A. Plichta

[12] Fuentes J.E., Francisco D.M., Oscar D.M.: Method for Estimating Solar Energy
Potential Based on Photogrammetry from Unmanned Aerial Vehicles. Electronics,
vol. 9, no. 12, 2020, 2144. https://doi.org/10.3390/electronics9122144.
[13] Zambrano-Asanza S., Quiros-Tortos J., Franco J.: Optimal site selection for
photovoltaic power plants using a GIS-based multi-criteria decision making and
spatial overlay with electric load. Renewable and Sustainable Energy Reviews,
vol. 143, 2021, 110853. https://doi.org/10.1016/j.rser.2021.110853.
[14] Bujarkiewicz A., Sztubecka M., Sztubecki J.: The Study of Using GIS Tools in
Sustainable Management of Solar Energy. Civil and Environmental Engineer-
ing Reports, vol. 28, 2018, pp. 26–39. https://doi.org/10.2478/ceer-2018-0003.
[15] Chow A., Alan S.F., Songnian L.: GIS Modeling of Solar Neighborhood Potential
at a Fine Spatiotemporal Resolution. Buildings, vol. 4, no. 2, 2014, pp. 195–206.
https://doi.org/10.3390/buildings4020195.
[16] Iñaki P., Izkara J.L., Usobiaga E.: The Application of LiDAR Data for the Solar
Potential Analysis Based on Urban 3D Model. Remote Sensing, vol. 11, no. 20,
2019, 2348. https://doi.org/10.3390/rs11202348.
[17] Redweik P., Catita C., Brito M.: Solar energy potential on roofs and facades in
an urban landscape. Solar Energy, vol. 97, 2013, pp. 332–341. https://doi.org/​
10.1016/j.solener.2013.08.036.
[18] Thebault M., Clivillé V., Berrah L., Desthieux G.: Multicriteria roof sorting for
the integration of photovoltaic systems in urban environments. Sustainable Cities
and Society, vol. 60, 2020, 102259. http://doi.org/10.1016/j.scs.2020.102259.
[19] Stachura T., Krzyś M.: GIS-based assessment of the feasibility of solar energy ap-
plications, in the case of Łazy village. Geomatics, Landmanagement and Land-
scape, no. 1, 2017, pp. 65–79. http://doi.org/10.15576/GLL/2017.1.65.
[20] Beseničar J., Trstenjak B., Setnika D.: Application of geomatics in photovolta-
ics. The International Archives of the Photogrammetry, Remote Sensing and
Spatial Information Sciences, vol. 37(B4), 2008, pp. 53–56.
[21] Guaita-Pradas I., Marques-Perez I., Gallego A., Segura B.: Analyzing territory
for the sustainable development of solar photovoltaic power using GIS databases. En-
vironmental Monitoring and Assessment, vol. 191, 2019, 764. https://doi.org/​
10.1007/s10661-019-7871-8.
[22] Ostapenko O., Koval V., Olczak P., Нren L., Matuszewska D., Postupna О.:
Application of Geoinformation Systems for Assessment of Effective Integration of
Renewable Energy Technologies in the Energy Sector of Ukraine. Applied Sci­
ences, vol. 12, 2022, 592. https://doi.org/10.3390/app12020592.
[23] Bukowski M., Majewski J., Sobolewska A.: Macroeconomic Electric Energy Pro-
duction Efficiency of Photovoltaic Panels in Single-Family Homes in Poland. Ener-
gies, vol. 14(1), no. 126, 2021. https://doi.org/10.3390/en14010126.
Using GIS and SDSS Tools in the Design of a Photovoltaic System for a Built-up Roof 57

[24] Gassar A.A.A., Cha S.H.: Review of geographic information systems-based rooftop
solar photovoltaic potential estimation approaches at urban scales. Applied En-
ergy, vol. 291, 2021, 116817. https://doi.org/10.1016/j.apenergy.2021.116817.
[25] Pawłowska M., Pawłowski L. (eds.): Advances in Environmental Engineering
Research in Poland. 1st  ed. Routledge, London 2021. https://doi.org/​10.1201/​
9781003171669.
[26] Bremer M., Mayr A., Wichmann V., Schmidtner K., Rutzinger  M.: A new
multi-scale 3D-GIS-approach for the assessment and dissemination of solar income
of digital city models. Computers, Environment and Urban Systems, vol. 57,
2016, pp. 144–154. https://doi.org/10.1016/j.compenvurbsys.2016.02.007.
[27] Saretta E., Bonomo P., Frontini F.: A calculation method for the BIPV potential of
Swiss façades at LOD2.5 in urban areas: A case from Ticino region. Solar Energy,
vol. 195, 2020, pp. 150–165. http://doi.org/10.1016/j.solener.2019.11.062.
[28] Brito M., Redweik P., Catita C., Freitas S., Santos M.: 3D Solar Potential in the
Urban Environment: A Case Study in Lisbon, Energies, vol. 12, no. 18, 2019, 3457.
https://doi.org/10.3390/en12183457.
[29] Sánchez-Aparicio M., Jimenez Jose A., Del Pozo S., González E., Lagüe-
la S.: Ener3DMap-SolarWeb roofs: A geospatial web-based platform to compute
photovoltaic potential. Renewable and Sustainable Energy Reviews, vol. 135,
2021, 110203. https://doi.org/10.1016/j.rser.2020.110203.
[30] Desthieux G., Carneiro C., Camponovo R., Ineichen P., Morello E., Boulmi-
er A., Abdennadher N. et al.: Solar energy potential assessment on rooftops and
facades in large built environments based on LIDAR data, image processing, and
cloud computing. Methodological background, application, and validation in Geneva
(solar cadaster). Frontiers Built Environment, vol. 4, 2018, 14. https://doi.org/​
10.3389/fbuil.2018.00014.
[31] Stendardo N., Desthieux G., Abdennadher N., Gallinelli  P.: GPU-Enabled
Shadow Casting for Solar Potential Estimation in Large Urban Areas. Application
to the Solar Cadaster of Greater Geneva. Applied Sciences, vol. 10, 2020, 5361.
https://doi.org/10.3390/app10155361.
[32] Bieda A., Cienciała A.: Towards a Renewable Energy Source Cadastre – A Review
of Examples from around the World. Energies, vol. 14, no. 23, 2021, 8095. https://
doi.org/10.3390/en14238095.
[33] SIP Poznan. http://sip.poznan.pl/model3d/#/legend [access: 1.12.2021].
[34] Solar cadaster of Grand Genève. https://apps.sitg-lab.ch/solaire/ [access:
9.04.2022].
[35] Map of solar potential in Munich. https://geoportal.muenchen.de/portal/so-
larpotenzial/ [access: 9.04.2022].
[36] The London Solar Opportunities Map. https://maps.london.gov.uk/lsom/?_
gl=1%2a13sqp4w%2a_ga%2aMTY4OTIzNzMzOC4xNjQ5NTEzODc4 [ac-
cess: 9.04.2022].
58 L. Davybida, I. Wyczałek, A. Plichta

[37] Solar atlas of Amsterdam. https://www.zonatlas.nl/amsterdam/ontdek-de-zo-

natlas/ [access: 9.04.2022].
[38] Map of solar potential in Wrocław. https://gis.um.wroc.pl/en/maps/solarna/
[access: 9.04.2022].
[39] Mapping solar energy resources in San Francisco. http://app.dumpark.com/
sunlight/sf [access: 9.04.2022].
[40] Energy Efficiency and Historic Buildings: Solar Electric (Photovoltaics). Historic
England, Swindon 2018.
[41] Kandt A., Hotchkiss E., Walker A., Buddenborg J., Lindberg J.: Implementing
Solar PV Projects on Historic Buildings and in Historic Districts. National Re-
newable Energy Laboratory, Colorado 2011.
[42] Global Solar Atlas. https://globalsolaratlas.info/download/poland [access:
8.12.2021].
[43] Geoportal Poland. https://www.geoportal.gov.pl/ [access: 8.12.2021].
[44] Google Earth Engine. A planetary-scale platform for Earth science data and
analysis. https://earthengine.google.com/ [access: 8.12.2021].
[45] Solar Energy – Introduction to SEBE. https://umep-docs.readthedocs.io/​
projects/​tutorial/en/latest/Tutorials/SEBE.html [access: 9.04.2022].
[46] Garcia R., Polo J.: Evaluating the Solar Potential of Rooftops on Campus San Joa-
quín, Santiago – Chile using Opensource GIS. [in:] EuroSun2020 – ISES Con-
ference Proceedings, International Solar Energy Society, 2020. https://doi.
org/10.18086/eurosun.2020.01.12.
[47] SunPower. https://sunpower.maxeon.com/uk/solar-panel-products/max-
eon-solar-panels [access: 8.12.2021].
[48] PVGIS – Photovoltaic Geographical Information System. https://ec.europa.
eu/jrc/en/pvgis [access: 8.12.2021].
[49] University of Oregon. Solar Radiation Monitoring Laboratory. http://solar-
dat.uoregon.edu/SunChartProgram.html [access: 8.12.2021].
[50] Fialho L., Melicio R., Mendes V.M.F., Figueiredo J., Collares-Pereira M.: Ef-
fect of Shading on Series Solar Modules: Simulation and Experimental Results.
Procedia Technology, vol. 17, 2014, pp. 295–302. https://doi.org/10.1016/​
j.protcy.2014.10.240.
[51] HelioScope. Advanced Solar Design Software. https://www.helioscope.com/
[access: 8.12.2021].
[52] Green M., Ho-Baillie A., Snaith H.: The emergence of perovskite solar cells. Nature
Photon, vol. 8, 2014, pp. 506–514. https://doi.org/10.1038/nphoton.2014.134.