See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268365410

PERFORMANCE RATIO OF A PV POWER PLANT: DIFFERENT PANEL

TECHNOLOGIES COMPARISON

Article

CITATIONS READS

4 2,131

4 authors, including:

S. Leva Emanuele Ogliari

Politecnico di Milano Politecnico di Milano
173 PUBLICATIONS   2,261 CITATIONS    23 PUBLICATIONS   345 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Day-ahead PV power forecast View project

All content following this page was uploaded by Emanuele Ogliari on 22 May 2015.

The user has requested enhancement of the downloaded file.

 PERFORMANCE RATIO OF A PV POWER PLANT:
DIFFERENT PANEL TECHNOLOGIES COMPARISON
F. degli Uberti*, R. Faranda**, S. Leva**, E. Ogliari**
*
juwi Energie Rinnovabili Srl
Via Sommacampagna 59/D – Verona, E-mail: [email protected]
**
Politecnico di Milano – Dipartimento di Energia
Via La Masa, 34 - Milano, E-mail: [email protected]

Abstract – In this paper a direct comparison between two different PV panel technologies is
presented. The calculation of PR, after one year of analytic monitoring of a PV plant, allows some
comparative considerations of special interest.
By comparing the performance between CdTe modules and the polycrystalline silicon ones, put
in equal environmental and plant conditions, it has resulted a lower annual average PR of the latter
than the one of the former.

Keywords: Performance Ratio, PV System, Monitoring, Thin Film

INTRODUCTION

The production of electrical energy from renewable sources and particularly by means of
photovoltaic systems has increased a lot in the last years and it‟s increasing further nowadays. At
the same time, the efficiency of the energetic production from this kind of plants, is becoming a
greatly interesting indicator. With regard to the standard conditions of operation, it is becoming a
parameter of great interest in the analysis of photovoltaic systems especially when the plant size is
medium to large.
The Performance Ratio (PR) is one of the main indicators to evaluate the good quality or not of
a photovoltaic system into operation. This is a dimensionless quantity indicating the net amount of
daily/monthly/annual energy output from the entire PV system compared with the theoretical input
from the PV array. The calculation of this indicator is becoming increasingly important both to
evaluate the efficiency of photovoltaic systems in service and in pursuit of new plants, especially
for their private and national funding.
Furthermore, the analytic monitoring of a photovoltaic system, as well as the calculation of the
PR, also aim to improve the knowledge of photovoltaic technology and the deepening of its
problems in the operational phase.
This study presents some results of the Performance Ratio analysis with reference to the
installation conducted by juwi Energie Rinnovabili Srl, located in the province of Lecce and
analyzed by the monitoring system and the site of Meteocontrol.
In particular, the purpose of this study is:
- to compare the PR fields with different technology in the monitored months;
- to analyze in detail the performance of the polycrystalline fields in the winter months.

At the same time the article also presents an overview on the PR values measured by different
institutions/research centers and available in the bibliography in order to have references for
assessing the quality of a photovoltaic system.

Proceedings of the Solar Energy Tech 2010 ISBN 978-1-4467-3765-1

NORMALIZED INDICES

PV systems with different configurations and placed in different locations can easily be
compared by evaluating their normalized indices of system performance such as yields, losses and
efficiencies. The amount of energy yields are normalized to the nominal power of the modules.
Yields of the system are normalized to the area of the arrays and losses are calculated as the
differences between the energy yields [1],[2].
The performance indices of systems connected to the distribution network of autonomous
systems and hybrid systems may differ significantly because of their adaptation to loading and other
special operating characteristics.

Daily mean yields are the quotient of energy quantities over the installed array‟s rated output
power P0 (kW). The yields have units equal to kWh·d-1·kW-1 or h·d-1 and indicate the amount of
time during which the array of modules would be required to operate at P0 to provide a particular
amount of monitored energy. The energy yields indicate actual array operation relative to its rated
capacity. The average daily energy performance are:

Array Yield
The Array Yield (YA) is calculated as the net daily energy from the PV array (EA,d) divided by the
installed array‟s rated output power P0.

This yield indicates the daily energy output per kW of the installed PV array. YA indicates the
amount of time during which the array would be required to operate at P0 to provide the monitored
daily energy.

Reference Yield
The Reference Yield (YR) can be calculated by dividing the total daily in-plane irradiation HI,d
(the monthly or yearly average value) by the module‟s in-plane reference irradiance GSTC (kW·m-
2 1
):

This yield represents the number of hours per day during which the solar radiation would need to
be at reference irradiance levels in order to produce the same incident daily energy as it was
monitored.
Thus YR would be the number of peak sun-hours per day.

Final Yield
Final Yield (YF) is calculated dividing the total system output energy (EUSE) by the installed
array‟s rated output power P0.
The final PV system yield YF is the portion of the daily net energy output of the entire PV plant
1
STC (Standard Test Conditions – CEI EN 60904-3):
 Irradiance GSTC=1.000 W/m2
 Irradiance spectral distribution of AM1,5
 Working temperature of the solar cell: 25°C±2°C

14
which was supplied by the array per kW of the installed PV array:

This yield indicates the number of hours per day that the array would need to operate at its rated
output power P0 to equal its monitored contribution to the total system output energy EUSE.

Performance Ratio
The Performance Ratio (PR) is calculated by dividing the Final Yield by the Reference Yield.
This index indicates the overall effect of losses on the rated output power of the array:

The Performance Ratio is a dimensionless quantity that indicates the amount of the net output
energy of the PV system, compared with the theoretical one in input for a certain period. It doesn‟t
represent the produced energy - in fact systems with low PR in a place with high radiation can
produce more energy than a system with a high PR located in a place with little radiation -, but the
overall components efficiency by the actual operating conditions with the exception of the PV
array.
The main factors of PR reduction are:
- High temperature on the PV array;
- Incomplete utilization of the irradiation;
- Inefficiencies or failures of the system components.

TYPICAL MEASURED VALUES OF PR IN PV SYSTEMS

The first reference is provided by the publications of the data collected and processed by the
Photovoltaic Power Systems Programme of the International Energy Agency (IEA PVPS)2, which
is based on the data from numerous PV systems distributed around the world.
In particular, according to reference [3], which assesses both residential and photovoltaic power
plants distributed in 17 different countries and built between 1983 and 2006, it may be noted (from
the data provided in relation to monitoring campaigns of the last 15 years) that the PR reaches only
in a few cases values of 88%.
35% of the monitored plants have a PR between 72,5% and 77,5% (Fig. 1). The average value of
the PR increases from 0,68, for plants built in 1994, to 0,72 for those constructed in 2004 (Fig. 2).
Analyzing the data concerning the production of the 15 biggest photovoltaic plants, listed in the
PVPS IEA [4], in which the peak power varies between 1,14kW and 3.000kW, it can be seen a
significant variability of the PR: the annual average PR value is equal to 0,66 and the best has a
value equal to 0,81 (Fig. 3).

2
PVPS is a collaborative research and development created within the IEA, which covers projects in the applications for the production of
electricity through photovoltaic technology.

15
 Figure 1. PR distribution compared to the Figure 2. PR trend in the year of the plant construction.
number of the monitored plants.

Figure 3. PR of the 15 largest systems in the IEA PVPS Database.

From publication [5] it can be inferred that the PR measured in 2007 of a grid connected
photovoltaic system, located on the island of Crete with a peak power of 172kWp with
polycrystalline silicon panels that has been operating since 2002, is between 58% and 73% with an
annual value of approximately 67%.
Work [6] was based on a two years monitoring (from August 2003 to July 2005) of a 5kWp plant
with amorphous silicon panels. In this case, the PR varies between 0,59 and 0,83 with an annual
value of 74% in the first year and 71% in the second year.
Finally, some significant data:
- the present studies for calculating the lifecycle of PV modules consider a PR of 0,75 for roof
PV systems and 0,8 for ground PV systems (references [7] and [8]);
- the calls for public funding set a minimum value for the PR of 0,75;
- one of the most widespread software to estimate the energy production from a PV system,
uses an estimated PR value of 0,77;

16
 From the analysis of the above-mentioned data and considering the most recent installations, the
typical values of PR, for well monitored and maintained large systems, are within the range from
0,75 to 0,82. The PR with values greater than 0,8 are considered excellent.
In any case, the PR is growing in parallel with the improvement of the efficiency of all single
components of the system and with the perfecting of the operation & maintenance plans: the IEA
PVPS itself has the aim in the next years (for plants that will enter service in 2014) of reaching an
86% PR for solar power plants.

MONITORING SYSTEM AND THE MONITORED PLANT

The analytic monitoring system was provided by Meteocontrol which allows data acquisition
from a PV plant giving its own:

- Sensors
- Hardware
- Software

The analysis of the transmitted data and their communication with the customers is given by
safer‟s sun portal. This portal looks like a web interface, where it is possible to have a view of the
monitoring of one or more monitored quantities in “real time” and to calculate the Performance
Ratio of the monitored plant.
Meteocontrol supplies without any treatment the data as they are measured by the sensors to third
parties, such as for example Politecnico di Milano, in order to calculate the performance indices of
the monitored plants.
In this particular case, the monitoring system was used to analyze juwi Energie Rinnovabili‟s
plant located in the province of Lecce. It has a total rated installed power of 3,2 MWp and it has
completely started working since May 2009 and it is characterized to use different kind of PV
module: multi-crystalline and thin film cell technology.
Main characteristics of the plant are given farther:

Field 1-5
Array technology: thin film panel – CdTe
Number of inverters: maximum 2 inverters/field
Field 6
Array technology: Polycrystalline Silicon
Number of inverters: 7 inverters (for shadowing)
Overall characteristics
Type of installation: ground
Commissioning: 29/12/08
Azimuth: 0
Tilt: 25°

THE RESULTS

A better comparison between the different technologies built in the several monitored fields has
been done analyzing the normalized indexes changes in one year of observation. This comparison

17
has been made first by calculating the monthly performance ratio for all the fields with reference to
a calendar year starting from 01/04/2009 to 10/04/2010.
All these values are listed in the Table A below.

Table A. Monthly PR of the monitored fields. * calculated from 01/04/2010 to 10/04/2010.

Month Field 1 Field 2 Field 3 Field 4 Field 5 Field 6

Apr „09 91,75 93,83 90,95 87,87 91,82 92,24
May „09 89,79 93,37 91,43 91,33 91,20 89,27
Jun „09 93,66 92,00 91,34 90,48 90,28 86,90
Jul „09 91,94 92,38 91,81 83,70 90,39 86,70
Aug „09 89,33 92,02 91,27 89,21 89,60 85,91
Sept „09 91,46 91,66 90,95 87,25 93,99 88,17
Oct „09 90,30 89,87 79,94 88,39 87,61 88,26
Nov „09 88,42 91,25 89,14 87,63 87,48 85,20
Dec „09 83,47 86,31 84,46 83,86 83,09 78,34
Jan „10 87,44 89,85 86,25 87,29 85,95 81,54
Feb „10 87,80 89,25 87,27 84,03 80,06 92,09
Mar „10 91,10 91,93 87,28 88,70 89,23 93,65
Apr* „10 90,63 91,47 89,35 87,83 88,57 92,31

As it can be seen from Table A, the data obtained for the month of April 2010 are only partial and
in the analysis that follows they will not be considered.
In relation to this, Figure 4 shows the trends of the PR fields with thin film technology (Fields 1
to 5) and the evolution of PR in the polycrystalline one, Field 6.

Figure 4. Evolution of monthly PR, from April 2009 to March 2010, of the monitored solar fields.

18
Analyzing the data in Table A and in the graph in Figure 4 it can be observed that:
 the minimum monthly PR was recorded by Field 6 in December 2009;
 the maximum PR was recorded by Field 5 in September of the same year;

Table B. Minimum, maximum monthly PR and average annual PR of the analyzed Fields.

Field 1 Field 2 Field 3 Field 4 Field 5 Field 6

min PR (monthly) 83,47 86,31 79,94 83,7 80,06 78,34
Average (year) PR 89,78 91,17 88,57 87,51 88,41 87,74
max PR (monthly) 93,66 93,83 91,81 91,33 93,99 93,65
max deviation +% 4% 3% 4% 4% 6% 7%
min deviation -% -7% -5% -10% -4% -9% -11%
min PR (monthly) 83,47 86,31 79,94 83,7 80,06 78,34

Table B shows the minimum, the maximum monthly values of PR and the average annual values
for each field, except for the month of April 2010. The annual PR were calculated as the yearly
averages of the monthly values.

Figure 5. minimum, maximum and annual average PR of the monitored period.

An analysis of these data has led to the graph in Figure 5. It can be seen how Field 2 is the one
with the highest average of annual PR, the lowest deviation of the minimum monthly PR and the
lowest deviation of the maximum PR from this value rather than other fields. The major differences
are recorded for the polycrystalline Field 6.

PR Comparison with reference to summer and winter results

The analysis has continued by comparing the PR with reference to summer and winter months.
Several factors influence the trend of the recorded monthly PR, some are related to common factors

19
(such as radiation measured in real terms) and other phenomena due to different technology used in
the fields (mainly: the greater impact of shading, different energy production with light
diffused/direct and different behaviour of operating temperature).
Figure 6 a) shows the PR of June, July and August 2009 (summer) and Figure 6 b) the same data
for the months of November, December 2009 and January 2010 (winter). An examination of the
pictures can show that the PR of Field 6 is always lower than those of the other fields except for
Field 4 that records the minor PR in the summer series.
This is due to a low PR recorded from 3 to 7 July, 2009, which brought the monthly average from
an expected value of 90,2 to 83,7. At that time only one of the two inverters of Field 4 worked.

Figure 6. PR monthly comparison the fields considered in a) summer and b) winter.

The low PR recorded in summer by Field 6 (Figure 6a) is due to the fact that polycrystalline
technology undergoes a greater influence of high temperature than that of CdTe thin film. The first
technology loses, in average, about 4-5% of PR against the second one. Indeed, the temperature
coefficient for polycrystalline panels technology amounted to values of -0,45%°C-1 while in the

20
case of CdTe modules it is equal to -0,25%°C-1.
By a similar analysis for the winter months, especially for December, it is noticed that all the
fields have recorded a lower PR. During this month Field 6 has seen its lowest PR that was also
lower than that of the others (Figure 6b). In this case the poor performance of Field 6 PR is due
primarily to mutual shading between rows of photovoltaic panels that are present in all fields, but in
the case of polycrystalline technology they are actually more burdensome.
Given the fact that CdTe technology itself has diodes for each cell, while the polycrystalline
silicon panels used in Field 6 provide a total of three blocking diodes, in case of shading, the second
type of panel is put "in block" for a greater proportion than the first.
Analysis of a significant day
Figure 7 shows the graph of output power measured from all inverters in Field 6 during the 13th
December, 2009.

Figure 7. AC power output from the inverters of Field 6 on 13/12/2009.

With WR E1, WR E2 etc... the values of AC power output recorded by all inverters in Field 6
were given. It is easy to see how early in the morning and at dusk, the conditions of more shading,
the inverters WR E5 and E7 are the first to operate, while WR E6 is the last to set going nearly two
hours later. The repetition of this phenomenon in most other winter days affects the PR of Field 6,
rather than the other fields in CdTe, as such technology is affected by the partial shading. The study
of the winter shading of Field 6, was also undertaken through the simulation software for
photovoltaic PVSyst. The polycrystalline silicon panels were designed in CAD environment as if
they were leaning against the ground, not considering the supports because they raise the whole
field and then, in terms of shading, there is no change.

21
 a)

b)

c)

Figure 8. Shading of Field 6 on December 13, 2009 (a) at 7:45, (b) 9:45 and (c) at 16:15.

From the simulation it can be seen that in the early morning hours of the 13th December, from
7:45 until 9:45, Figure 8a and Figure 8b, we have a partial shading of the rows which saves only the
first row consisting of two panels that are part of the inverter WR7, and the front panel of the
opposite inverter pair managed by the WR5 one.
Similarly, one can observe this phenomenon in the evening when, in the same simulation,
because of the sun that lowers on the horizon from 14:30 until 16:30, the front of the rear rows are
shaded. In particular, we can see in Figure 8c as the panels of the inverter WR7 are the only ones

22
not to be shaded at 16:15, while only one of the two panels operated by WR1, WR2, WR4 and WR6
are almost still fully irradiated while the WR3 and WR5 are almost totally in shadow. Even this
simulated behaviour has a direct feedback with the course of monitored quantities on the 13th
December 2009 and demonstrates clearly that the cause of low productivity in winter is solely due
to the mutual shading between the rows of Field 6. The design was made to maximize the
performance ratio in the winter months.

CONCLUSIONS

As already known, the Performance Ratio is a key indicator of the performance of a photovoltaic
system. The calculation of this indicator is becoming increasingly important both to evaluate the
efficiency of photovoltaic systems in service, and in sight of new plants installations and mainly of
their private and national funding.
The calculation of PR, with reference to the monitored plant in service for about one year
consisting of fields made with different technologies of photovoltaic panels, allows some
comparative considerations of special interest. By comparing the performance between CdTe
modules and the polycrystalline silicon ones, put in equal environmental and plant conditions, it has
resulted a lower annual average PR of the latter than the one of the former.
The main causes of this behaviour are due to:
 the different characteristics on conversion of sunlight into electrical energy of the two
compared technologies, and in particular that polycrystalline technology undergoes greater
influence of high temperature than that of CdTe thin film;
 shadings during winter months which, although common to both technologies, has been a
discriminating factor for the polycrystalline silicon. In fact, in this technology, even with
minimal shading, the entire module, or a large amount of it, is lost. As, given different sizes of
the cell, even if there is a blocking diode for cell (row) the percentage of form that doesn‟t
convert energy is higher in the case of polycrystalline panels rather than in the thin film ones.
It's important to emphasize that conclusions about panels of different technologies here exposed
are to be considered limited to the monitored plant and must not be inferred as general
considerations. It should be taken into account that a problem of shading as that exhibited in the
current article is only present in land installations and on flat roofs.
Finally, regarding the here analyzed plant, it has to be underlined that the annual PR always
measured above 85%, which means that the plant has been studied in terms of materials used and
designed in an excellent way.

REFERENCES

[1] CEI EN 61724, Photovoltaic system performance monitoring. Guidelines for measurement,
data exchange and analysis, (1999).
[2] IEA PVPS Task II, http://www.iea-pvps.org/.
[3] T. Nordmann, L. Clavadetscher, U. Jahn: PV system performance and cost analysis, Proc. 22nd
European Photovoltaic Solar Energy Conference and Exhibition, Milano - Italy, (3-7 September
2007).
[4] Performance Ratio for the 15 largest PV systems in the IEA PVPS Database (21-Aug, 2007).

23
 [5] E. Kymakis, S. Kalykakis, T. M. Papazoglou: Performance analysis of a grid connected
photovoltaic park on the island of Crete, Energy Conversion and Management, Vol. 50, (2009),
pp.433-438.
[6] Napat Watjanatepin: two-years performance of a 5 kWp amorphous silicon GPV system.
[7] W.M. Fthenakis, E. Anselma: Photovoltaics Energy Payback times, Greenhouse Gas Emissions
and External Costs: 2004-early 2005 Status, Progress in Photovoltaics: research and
applications, vol.14 (2006), pp.275-280.
[8] W.M. Fthenakis, H. C. Kim, E. Anselma: Emission from Photovoltaic life cycles,
Environmental Science & Technology, Vol. 42, n.6, (2008), pp.2168-2174.
[9] http://www.nrel.gov/about/, http://www.pvwatts.org/.

24

View publication stats