Contents

I LECTURES 3
INTRODUCTION: PV CELLS AND MODULES . . . . . . . . . . . . 5
SOLAR GEOMETRY AND RESOURCE . . . . . . . . . . . . . . . . 31
SOLAR GENERATOR . . . . . . . . . . . . . . . . . . . . . . . . . . 65

SE
R
U
O
C
ST
SY
PV

Bruno Wittmer and André Mermoud

PVSYST SA
Route du Bois-de-Bay 107,
1242 Satigny - Switzerland
Any reproduction or copy of the course support, even partial,
is forbidden without a written authorization of the author.

1
2
PV

PVsyst SA
SY
ST
C
O
U
R
SE
 Part I

SE
LECTURES

R
U
O
C
ST
SY
PV

3
PV
SY
ST
C
O
U
R
SE
 INTRODUCTION: PV CELLS AND MODULES

PVsyst Training for Engineers

SE
PV Cells and Modules

Bruno Wittmer

R
Bruno.Wittmer@pvsyst .com

U
PVSYST SA - Route du Bois-de-Bay 107 - 1242 Satigny - Suisse
www.pvsyst.com
O
Any reproduction or copy of the course support, even partial, is forbidden without a written authorization of the author.
C
Contents
ST

• Semiconductors
SY

– Introduction
– pn-Junction
• PV Cells
PV

– Working Principle
– Technologies
• PV Modules
– Functionality
– Characterization

Page
Page
2 2

PVsyst SA 5
 INTRODUCTION: PV CELLS AND MODULES

Semiconductors
Semiconductors:
- Introduction
- Doping
- pn-Junction

SE
R
U
Page
O
Page
3 3
C
Charge and Current
ST

Electrical Charge:
SY

• Can be positive or negative, electrons are negative,

protons (nuclei) have positive charge
• Measured in Coulomb (C)
PV

Electrical Current:
• Charges in motion, measured in Ampere (Coulomb
per second Cs-1)
• Mostly the current is carried by electrons (negative
charge). In semiconductors there are also holes
(positive charge).
• To get electrical current one needs electrical charges
that can move freely.

Page
Page
4 4

6 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Semiconductors
Conductor:
• Almost infinite resources of electrons (charge carriers).
• Conductivity is limited by the ‘friction’ that charges
experience.
Insulator:
• No free charges are available.
Semiconductor:

SE
• Very limited number of charge carriers
(electrons and holes).
• Conductivity is mainly limited by the amount of available
free charge carriers.

R
• The amount of available charge carriers can be controlled
by doping, small electrical currents, electrical fields, light,

U
temperature, etc.
=> very complex components can be designed. O
Page
Page
5 5
C
Semiconductor Doping
ST
SY
PV

Pure Silicon n-doping 5 el. p-doping 3 el.

(ex: phosphorus) (ex: boron, aluminum)

• Adding materials with an extra electron (P) or a missing electron (B) in the valence
band, the semiconductor is doped to give it a positive or negative intrinsic polarity.
• The bulk crystal is doped at the growing time. Additional doping is performed by
vacuum evaporation ( < 1 mm thickness) .
• Typical concentrations : 1013 – 1017 cm-3 (1 impurity per 109 – 105 Si atoms)

Page
Page
6 6

PVsyst SA 7
 INTRODUCTION: PV CELLS AND MODULES

pn-Junction
pn-Junction:
- Interface between p-doped and n-doped semiconductors.
- The pn-junction conducts only in one direction (from p to n).
Diffusion of charge carriers leads to an intrinsic electrical field at the
junction, this is the place where the charges that are generated by the
photoelectric effect get separated.
- The pn-junction is the origin of the photocurrent and shapes the IV-curve of
the PV cell.

SE
R
U
Page
O
Page
7 7
C

IV Curve of pn-Diode
ST

Shockley Equation:
• Describes current in diode
SY

I : diode current
Is : reverse bias saturation current
PV

VD : voltage across the diode

VT : thermal voltage
n : ideality factor

PV Cells are pn-diodes

operated in forward direction
Voltage drop in forward direction
depends on band gap Egap

Page
Page
8 8

8 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

PV Cells
PV Cells:
- Photovoltaic effect
- PV Cell Working Principle
- Cell characteristics
- PV Cell Technologies

SE
R
U
Page
O
Page
9 9
C

Light
ST

Electromagnetic Wave  Particle

Intensity  Nr. of photons
SY

Frequency  Energy of photons

PV

h: Planck constant photon energy [eV]

c: speed of light
n: frequency = w/2p
l : wavelength
1eV = 1.602 10-19J wavelength [nm]

Visible light is a narrow band in the electromagnetic spectrum

Incident light on earth atmosphere (solar constant): 1363 W/m2
On surface: Around 1000 W/m2 perpendicular to the sun.

Page
Page
10 10

PVsyst SA 9
 INTRODUCTION: PV CELLS AND MODULES

Photovoltaic Effect

Convert photons into electrical energy Photon

-
Photovoltaic Effect: Egap
Electrons get from valence to conduction band
Electrons in conduction band can move +
Holes in valence band can move
Both can sustain a current (photocurrent)

SE
Material Egap Do not confuse with
Si (crystalline) 1.12 eV Photoelectric effect :
Si (amorphous) 1.7 eV Photoelectric Effect:
GaAs 1.43 eV Electrons leave the material

R
CdTe 1.49 eV
CuInSe2 (CIS) 1 eV

U
Cu(In,Ga)Se2 (CIGS) 1 eV

Page
O
Page
11 11
C
Photovoltaic Cell
ST

Photovoltaic Cell:
- Light creates electron-hole pairs
SY

- Charges get separated by built-in field of pn-junction

- Electric field pointing from n to p
- Current flow from n to p
PV

1. Front metallization
2. AR coating
3. Textured front surface
3-4. Junction and electrical
field
4. p doped substrate
4-5. BSF, back surface field
5 p+ layer
6 Aluminum metallization
7. Solderable metallization

Page
Page
12 12

10 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Photovoltaic Cell
Photovoltaic Cell:
- Charge separation leads to photocurrent front contacts
- Losses due to reflection and transmission
n layer
- Losses due to recombination
p substrate
A
B
back contact
C
C A reflection and shading

SE
losses
B transmission losses

E
C absorption

R
D D charge separation
E recombination losses

U
Page
O
Page
13 13
C

Si-Crystalline spectral sensitivity

ST

Solar spectrum Use of the solar

spectrum:
SY

Energy dissipated Egap = 1.12 eV

as heat
l = 1.1 mm
PV

Photons < 1.12 eV:

not absorbed
Not absorbed
Energy
Usable
Energy Photons > 1.12 eV
Only 1.12 eV used

PVsyst SA

35% of the spectrum useable – A fundamental limit to the efficiency …

Page
Page
14 14

PVsyst SA 11
 INTRODUCTION: PV CELLS AND MODULES

Effect of Egap on Efficiency

The band gap influences the efficiency of a PV Cell:
- Higher band gap leads to larger voltage drop across junction
=> more electrical power per electron-hole pair
- Higher band gap looses low-energy part of the spectrum

increasing Egap

SE
R
U
Page
O
Page
15 15
C

IV curve of PV Cell
ST

-IL
- Light acts as a constant current source
SY

- Photocurrent splits into current in diode

and current in load

Iph = IL + ID
PV

convention is to
show the current
in the load

Page
Page
16 16

12 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Power curves of PV Cell

Power = I x V => Maximum power point MPP

SE
R
MPP depends on irradiance and temperature
Cells in series or parallel with different MPP lead to mismatch losses

U
Page
O
Page
17 17
C
PV Cell Characterization
ST

Parameters:
SY

- Open circuit Voltage Voc

Isc
- Short circuit current Isc MPP
Impp
- Maximum power point Impp , Vmpp
- Fill Factor FF: (Vmpp x Impp)/(Voc x Isc)
PV

should be as large as possible. FF

- Efficiency :
Electrical power output divided by
total Energy flux incident on cell.
Vmpp Voc

Page
Page
18 18

PVsyst SA 13
 INTRODUCTION: PV CELLS AND MODULES

Single Diode Model

- Single Diode Model:
- Equivalent circuit describing the behavior of a single PV cell or module.
- Considers also parasitic resistances:
Shunt Resistance Rsh and Series Resistance Rs
- Parametrization of Temperature dependences
- This is the very core of the PV simulation in PVsyst.

SE
PV cell V + I * Rs Use (load)

Rs
I
Photocurrent Diode

Iph Rsh V RL

R
U
Page
O
Page
19 19
C
PV Cell Technologies
ST

Crystalline Silicon:
- Mono-Crystalline
SY

- Poly-Crystalline Crystalline Silicon ( )

dominates global PV production
Thin Film:
- Amorphous Silicon (a-Si)
PV

- Cadmium Telluride (CdTe)

- Copper-Indium-Selenide (CIS/CIGS)
New Technologies:
- Silicon Heterostrucutures (HIT)
- Organic/Polymer Cells
- Dye-Sensitized PV Cells
- Perovskite Cells
- Microcrystalline silicon
- Quantum Dots

Page
Page
20 20

14 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Silicon Raw Material

Quartz or Silica (SiO2)

Silica mixed with charcoal, wood, coal

Electric arc furnace at 1900°C:
SiO2 + 2C -> Si + 2CO

Metallurgical Silicon (> 95% pure)

SE
Siemens process Fluidized Bed Reactor (FBR) Upgraded Metallurgical
CVD of Trichlorosilane (HSiCl3) Modified Siemens process Grade (UMG)
Si gets deposited on seed rods Deposition of Silane (SiH4) on seed less energy consumption
9N-11N purity (electronics grade) granules in a continuous process ≤ 5N purity
Most common process 5N – 6N purity

R
Solar grade Polysilicon 5N – 6N (99.99999% pure)

U
Page
O
Page
21 21
C

Crystalline Silicon Wafer Production

ST

Silicon: Mono- or Poly-crystalline

SY

• Ultra-pure silicon (10-7 to 10-9 impurities)

• Ingot "pulled" (mono) or cast (poly)
• Sawed as "wafers", thickness typ. 200 mm),
• Typical sizes: 5" = 12.5x12.5 or 6" = 15.6x15.5 cm²
PV

• EFG (poly): wafers extruded as ribbon – no sawing

Poly-crystalline Mono-crystalline
Mono-crystalline Silicon Wafers
Silicon Ingot (round) Silicon Ingot (square)

Page
Page
22 22

PVsyst SA 15
 INTRODUCTION: PV CELLS AND MODULES

Crystalline PV Cell: Basic Structure

Wafer substrate: p substrate
• Typically p-type bulk (Boron doping)
• Doping performed at Ingot production

n-type Implant:
• n-doping by surface diffusion (Phosphorous) n implant
• Creates pn-junction a few hundred nm deep

SE
Texturing:
• Structured front-side enhances light collection
(light trapping) textured surface

R
• AR coating with Silicon Nitride Si3N4 ( or Titanium
Dioxide TiO2) applied by Plasma Enhanced Chemical
Vapor Deposition (PECVD)

U
• Passivation Si3N4 also reduces recombination losses
O
Page
Page
23 23
C

Crystalline PV Cell Electrical Contact

ST

Front Contacts:
• Silver Electrodes are screen-printed on front side (bus bar and fingers)
SY

• Baking of silver paste creates electrical contact

• Methods to get more narrow contacts are constantly improving
=> less shading losses
Back Contact:
PV

• Back side covered with Al paste and heated to form ohmic contact
• Leads to Back Surface Field (BSF) that reduces recombination losses

bus bar

fingers

crystal boundaries

Mono-crystalline cell Poly-crystalline cell

(rounded corners) (rectangular)
Page
Page
24 24

16 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Special PV Cell Structures I

Back contact cell:

- PV cell where both contacts are on the back side.
- Increased efficiency (> 23% in production) due to
lower shading losses.
- Growing importance in PV cell production.

PERC Cell:

SE
- Passivated Emitter Rear Cell
- Passivation layer on back enhances electrical
gradient

R
- Reduced recombination => higher efficiency

U
Page
O
Page
25 25
C
Special PV Cell Structures II
ST

Metal Wrap-through (MWT):

SY

- Contact to the front side is made from the back by

metal vias passing through the silicon bulk
- Bus bars can be placed on the back side
PV

Emitter Wrap-through (EWT):

- Contact to the front side is made from the back by
n-implants passing through the silicon bulk
- Approx. 1 feed-through per mm2
- Charge separation is also possible in horizontal
direction

Page
Page
26 26

PVsyst SA 17
 INTRODUCTION: PV CELLS AND MODULES

Special PV Cell Structures III

Bi-facial Cell:
- PV cells or modules that can collect light on
both sides (front and back).
- Back side is less sensitive than front.
- Yield increase due to additional collection of
reflected or scattered light.
- Useful where vertical mounting is mandatory.

SE
pin Cell:
- p-type, intrinsic, n-type structure.
- Intrinsic layer enhances depletion zone

R
- Penetrating irradiation can be absorbed =>
higher quantum efficiency.

U
Page
Page
O
27 27
C

Thin film technologies

ST

Films deposited on various substrates (glass, metal, plastic)

High absorption coefficient materials => Typical thickness 1-2 mm
SY

Market Share of Thin-Film Technologies

Percentage of Total Global PV Production
PV

Main Technologies
• CdTe
• CIS / CIGS: CuIn(Ga)Se
• a-Si: amorphous Si

Data: from 2000 to 2010: Navigant; from 2011: IHS. Graph: PSE AG 2015

©Fraunhofer ISE: Photovoltaics Report, updated: 26 August 2015

Page
Page
28 28

18 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Thin Film: CdTe

CdTe
• Simple and cheap Technology
• Record Cell efficiency > 20%, record module
efficiency >18%
• Typical efficiencies > 15%
• Band Gap : 1.49 eV
• Potential environmental issues with Cadmium

SE
• Limited Tellurium resources

Group IA IIA IIIB IVB VB VIB VIIB VIIIB IB IIB IIIA IVA VA VIA VIIA VIIIA

R
P1 H He

P2 Li Be B C N O F Ne

P3 Na Mg Al Si P S Cl Ar

P4 K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr

U
P5 Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe II – VI Semiconductor
P6 Cs Ba Lu Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn

P7 Fr Ra Lr Rf Db Sg Bh Hs Mt Ds Rg Cn Uut Fl Uup Lv Uus Uuo

Page
Page
29 29
O
C

Thin Film: CIS / CIGS

ST

CIS or CIGS
• CuInSe2 - Copper-Indium-Selenium + option. Gallium
SY

• Band Gap : 1.0 – 1.6 eV (tuned via Ga-concentration)

• Direct bandgap => high absorption coefficient
• Record Cell efficiency > 20%
PV

• Typical efficiencies 12% - 15%

Al/Ag contact grid

n-type ZnO/ITO window
n-type CdS buffer
hetero-junction
p-type CIGS layer

Mo back contact

Substrate

Page
Page
30 30

PVsyst SA 19
 INTRODUCTION: PV CELLS AND MODULES

Thin Film: a-Si

Amorphous silicon:
• a-Si:H, single cell, tandem, triple junction
• Egap = 1.7 eV,
• p-i-n junctions (intrinsic layer)
• Short diffusion length => resistivity 
• Typ. efficiencies 6% (simple) 8% (tandem) 10% (triple)

SE
c-Si a-Si a-Si:H
amourphous
crystalline amorphous
hydrogenated

R
U
Page
O
Page
31 31
C
New Technologies I
ST

Silicon Heterostructures (HIT):

- Heterojunction with Intrinsic Thin-layer
SY

- Mono-crystalline cell with front-doping

replaced by amorphous a:Si-H film
- High efficiency solar cells (>25%)
Source: T. Mishima et al.
- High costs ‘Development status of high-efficiency HIT solar cells’
PV

in Solar Energy Materials & Solar Cells

Organic and Polymer Solar Cell:

- Flexible cell made from polymers
- Low efficiency
- Degradation Problems
- Very few commercial manufacturers

Page
Page
32 32

20 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

New Technologies II
Dye-sensitized PV Cells:
- Principle close to the chlorophyll absorption
Namely developed by Pr. Graetzel at EPFL (Swiss
Institute of Technology)
- In principle simple technology at very low cost
- More than 10% efficiencies in laboratory
- Not yet in the industrial phase.

SE
Perovskite Solar Cells:
- Cell with perovskite absorber (lead or tin halide)

R
- Cheap and simple to manufacture
- Efficiencies in lab up to 20%

U
- Not yet on the market

Page
O
Page
33 33
C
New Technologies III
ST

Nano-crystalline Silicon:
- Amorphous silicon with very small embedded Si crystals
SY

- Band-gap as c-Si (1.1 eV), but useable in thin films

- Combined with a-Si in tandem junctions
PV

Quantum Dot Solar Cell:

- Semiconducting Nano-Crystals
- Band Gap can be tuned by size of dots
- Useful for multiple junctions to cover solar
spectrum

(Source: Vladimir Mitin/U. Buffalo)

Page
Page
34 34

PVsyst SA 21
 INTRODUCTION: PV CELLS AND MODULES

Efficiency of PV Cells

SE
R
U
This plot is courtesy of the National Renewable Energy Laboratory, Golden, CO.

Page
Page
O
35 35
C

Market Shares of PV Technologies

ST
SY
PV

Data: from 2000 to 2010: Navigant; from 2011: IHS (Mono-/Multi- proportion by Paula Mints). Graph: PSE AG 2015

©Fraunhofer ISE: Photovoltaics Report, updated: 26 August 2015

Page
Page
36 36

22 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

PV Modules
Module Structure:
- Serial and parallel connection of Cells
- Protection Diodes
- Encapsulation
- Glass cover

SE
R
U
Page
O
Page
37 37
C
PV Module structure
ST

PV Module:
- Connect PV cells in series to form string
SY

- Bypass diodes protect each string

- Serial connection of strings
(very rarely parallel connections)
- Encapsulate module for protection
PV

: PV cell : bypass diode

IV-Curves of Cells add up

60 cells in series

Page
Page
38 38

PVsyst SA 23
 INTRODUCTION: PV CELLS AND MODULES

PV Cell Strings
Manufacturing steps:
• Cells flash-test and sorting:
All cells inside a module should have a similar IV-curve
• Assembly of cell strings (tabs soldering)
• Solder cleaning, electrical and optical control

Connect front bus bars to backside Connect Cell strings inside a module

SE
R
U
Page
O
Page
39 39
C
Bypass Diode
ST

Hot-Spot:
- Situation when a PV cell dissipates power instead of generating it.
- Can happen with strong imbalance in series connection of PV cells (e.g. due
SY

to shading).
- Hot-spots can lead to overheating and destruction of a PV cell.
Bypass Diode:
- By-pass diodes protect the modules from hot-spots and
PV

reduce the inefficiencies caused by strongly unbalanced

series connections.

Page
Page
40 40

24 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Module Encapsulation and Frame

The cover – mechanical support – is most often in tempered glass
 High transparency
 High refractive index => favors the Anti-Reflective coating action
 Dilatation coefficient close to the one of the cells.

The backside protection is often in Tedlar or Polyester

 Sometimes second glass => connections at the edges, or holes in the glass
 Encapsulating material EVA (Ethyl-Vinyl-Acetate)

SE
Frame
 Most often in aluminum, rarely plastic
 Without frame ("laminate"): fragile during
mounting !

R
For integration in buildings or for use on
tracking structures

U
Mounting with very low tilts are possible
(avoids accumulations of dirt and mosses) O
Page
Page
41 41
C
Typical PV Modules
ST

Crystalline Modules:
- Power ranges from 120 to 350 Wp (36 to 96 cells)
- Connections with sealed junctions boxes and cables + connectors
SY

- With or without frame (laminates)

- Glass-glass modules for integration, with cells spacing
PV

Poly-crystalline Mono-crystalline Mono-crystalline Bifacial panels

72 Cell in Series 72 Cells 60 Cells Glass-glass
Al frame Al frame frameless cell spacing

Page
Page
42 42

PVsyst SA 25
 INTRODUCTION: PV CELLS AND MODULES

PV Modules

PV Module characterization:
- IV and PV curves

SE
- Module Efficiency
- Flash Testing and Data sheets
- Modules in PV Systems

R
U
Page
O
Page
43 43
C
IV Curve of PV Module
ST

IV Curve:
- Current of a device as function of the applied voltage.
SY

- The IV curve is specific for a module, and depends on the technology, the
irradiance and the temperature.
- The IV curve of a cell or module has to be known with good precision to be
able to perform the simulation.
PV

Voc drops with

temperature

IV curves for different irradiances IV curves for different temperatures

Page
Page
44 44

26 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

PV Curve of PV Module
PV curve:
- Power generated in the module as function of the applied voltage.
- PV curves are useful when studying the working point of a PV module.
- The ideal working point of a PV cell or module is the maximum of the PV-
curve.

SE
R
U
PV curves for different irradiances PV curves for different temperatures

Page
O
Page
45 45
C
PV Module Efficiency
ST

PV Module Efficiency:
- Electrical power output divided by total Energy flux incident on module.
SY

(Pnom / 1000 Wm-2 / module area)

- Depends on technology, irradiance and Temperature.
- Efficiency drops for low light irradiance and high temperature
PV

: STC
25 °C
1000 W/m2
Efficiency drops for Efficiency drops for
low irradiance high temperature

Module efficiency for different irradiances Module Efficiency for different temperatures

Page
Page
46 46

PVsyst SA 27
 INTRODUCTION: PV CELLS AND MODULES

PV Module Efficiency
Efficiency of PV modules:
- A few % lower than for bare PV Cells
- Losses due to cell interconnection and encapsulation
- Module Efficiency is increasing over time as technologies improve

SE
R
U
© Fraunhofer ISE: Photovoltaics Report (2014) Updated: 28 July 2014

Page
O
Page
47 47
C
Standard Test Conditions (STC)
ST

Standard Test Conditions (STC):

- Standardized conditions for testing PV modules (flash test).
SY

- One of the first existing standards for testing PV modules.

- Irradiance: 1000 W/m2
Ambient Temperature: 25 °C.
Spectrum: 1.5 AM
PV

This does not always represent typical working conditions:

At this irradiance, the cell temperature is usually higher than 25°C
At low irradiance the efficiency of a PV cell decreases
- International standard for PV module testing, allows comparison of module
yields between different manufacturers, models and series. Reference
measurement from which one can extrapolate (with some limitations) to
different operating conditions.

Page
Page
48 48

28 PVsyst SA
 INTRODUCTION: PV CELLS AND MODULES

Module Specifications
Modules are systematically measured
Example from a module datasheet (First Solar)
by the manufacturers using flash testers

Minimal requirement:
STC measurement
(Standard Test Conditions)

Temperature dependence

SE
Measurement closer to real conditions:
lower irradiance (800 W/m2)
higher temperature (NOCT 45°C)

R
Manufacturers provide flash test results

U
for each individual module
=> sorting can reduce mismatch losses

Page
O
Page
49 49
C
Test Standards and Nominal Power
ST

Test Standards:
- IEC 61215:
Crystalline silicon terrestrial photovoltaic (PV) modules - Design
SY

qualification and type approval.

- IEC 61853:
PV Module performance testing and energy rating
- IEC 61646:
PV

Thin-film terrestrial photovoltaic (PV) modules - Design qualification and

type approval.
Nominal Power (Pnom):
- Power produced by a module at STC conditions, as measured or declared by
the manufacturer
- Also called nameplate power. Some manufacturers treat Pnom as a lower
tolerance.
- Pnom is used to rate the power of a PV installation. Pnom is used as reference
for calculating the Performance Ratio.

Page
Page
50 50

PVsyst SA 29
 INTRODUCTION: PV CELLS AND MODULES

PAN File
PAN File:
- File in PVsyst, holding all information on a PV module type.
- PAN files gather the parameters defining a module in the database of
PVsyst.
- Data can be imported from the PHOTON database (partial data).
- PAN files can be visualized and edited in PVsyst.
- Module manufacturers sometimes provide PAN files for their products.
Third party testing laboratories can provide PAN files.

SE
- The internal PVsyst database contains more than 10 000 module types. It is
crucial to use a correct PAN file for obtaining good simulation results.

R
U
Page
O
Page
51 51
C
Next Step: PV System
ST

PV System:
- Connect several PV Modules in series to form a String.
SY

- Connect several Strings to an Inverter.

- Simulate the PV System.
PV

String definitions in PVsyst

String definitions in 3D scene
‘System’ dialog

Page
Page
52 52

30 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

USE OF PVsyst FOR GRID-CONNECTED SYSTEMS

SE
SOLAR GEOMETRY and RESOURCE

André Mermoud
andre.mermoud@pvsyst.com

R
U
PVSYST SA - Route du Bois-de-Bay 107 - 1242 Satigny - Suisse
www.pvsyst.com
O
Any reproduction or copy of the course support, even partial, is forbidden without a written authorization of the author.
C
ST

Contents
SY

• Solar geometry
• Energy from the sun
PV

• Horizontal Irradiance (meteo)

• Irradiance on a tilted plane
• Shed arrangement, tracking planes
• Shadings: far and narrow
• Practical evaluations

PVsyst SA 31
 SOLAR GEOMETRY AND RESOURCE

The Sun and the Earth

The Sun
• Gaseous star: ¾ Hydrogen and ¼ Helium
• Energy: fusion reactor 4 Hydrogen => Helium + energy
(i.e. nucleus: 4p  2p + 2n + 2 e+ + energy)
• Energy irradiated as photons and neutrinos
• Diameter 1.4 M km
The Earth
• Earth-sun distance 150 M km (149.6 +/- 1.7%)

SE
• Apparent diameter of the sun: 0.53°
• Rotation axis tilted by 23°27’ on the ecliptic plane

R
• Declination d = angle between the equator plane end the sun-earth axis
approximation Sin (d) = 0.4 sin t

U
where t = angular coordinate of the earth's position, origin = spring equinox :
t = 2 p · (NoDay - 80) / 365.25
O
Page 3
C
The Sun – Earth system
ST
SY
PV

July 6th

Page 4

32 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Sun's Trajectory

SE
R
From a terrestrial point of view:

U
Each day, the sun follows a circle around the rotation axis
At equinoxes, the observer is at the center of the circle
O
Page 5
C
" Solar angles " definition
ST

Zenith
SY

West
PV

South East

• Sun Height = angle with respect to the horizontal

• Sun Azimuth = angle with respect to South (+ toward west, i.e. afternoon)

Page 6

PVsyst SA 33
 SOLAR GEOMETRY AND RESOURCE

Solar Angles calculation

Hourly Angle: angle between the sun projection on the equator plane
and the south ("true" noon). Its varies by 15°/hour

sin HS = sin LAT * sin d + cos LAT * cos d * cos AH

sin AZ = cos d * sin AH / cos HS
Special values:

SE
Sunrise/sunset: condition. HS=0 => cos AH = - tg LAT * tg d
Height at noon: HSmax = 90° - LAT + d

R
U
O
Page 7
C
Sun paths at Geneva
ST
SY

summer
spring
winter
autumn
PV

Page 8

34 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Solar Time – Legal Time

Solar Time (AH):
• Day length defined by the sun passing in the meridian plane (real noon)
• Day length: 1 earth rotation + 2p/365 arc around the sun
• Not constant along the year for 2 reasons:
• Ellipticity => Area's law of Kepler, sinus. variation of +/- 7.8 min
• Obliquity => Sinus-like variation over 6 months, +/- 10 min
• => Time Equation, which varies slowly from year to year

SE
due to the equinoxes precession (26'000 years period)

Legal Time :
• Clock time: 24h / day 365.25 days/year

R
• Depends on the Longitude of the site (Time zones, with respect to GMT)

U
LT - ST = TZ + Long/15° + TimeEq
O
Page 9
C
The Time Equation
ST
SY
PV

Time Equation = 0.0072 cos J – 0.0528 cos 2J – 0.0012 cos 3J

- 0.1229 sin J – 0.1565 sin 2J – 0.0041 sin 3J
J = j * 2 p / 365.25 j = nr. of day in the year

Page 10

PVsyst SA 35
 SOLAR GEOMETRY AND RESOURCE

Energy from the Sun

Almost the whole renewable energy available on the earth :

• Extraterrestrial: Solar constant 1363 W/m² (ellipticity: +/- 3%)
• On earth: Average 137 W/m² horizontal at Geneva
x 8760 hours : (1200 kWh/m²/year)

Other resources:

SE
• Heat from the earth (geothermal, < 1 W/m²)
• Tides
• Fission (Uranium: limited resources)

R
• Fusion (unlimited, but not yet controlled)

U
O
Page 11
C

To be compared with our energy needs …

ST

World Annual Energy

SY

Consumption
16 TW·year
PV

16 = 16’000’000’000 kW·year
~ 475 exajoules ( 1018 J )

© R. Perez et al.

Source: Richard Perez et al.

Page 12

36 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Renewables
25-70 Wind
Par an

Waves 215 (13)

0.2-2 total

Natural Gaz

3-11
per year
240 (15)
Ocean's thermal total

Oil
2-6
per year arctic
Biomass
3-4

SE
16 TW · year per year 90-300 (5-18)
0.3-2 Hydroelectricity Total
0.3 per year
per year
Yearly
Geothermal
consumption Tides
Uranium

R
U
Limited reserves 900 (56 years) Coal
Total reserves

Source: Richard Perez et al.

Page 13
O
C
Renewables
© R. Perez and al. 25-70 Wind
ST

per year
Waves
0.2-2
215 (13)
total

Natural Gas
SY

Sun 3-11
per year

23’000 per year Oceans

240 (15)
total
Thermal
Oil
2-6 arctic
PV

per year

Biomass
3-4
16 per year 90-300 (5-18)
0.3-2 Hydroélectricity Total
0.3 per year
per year
Annual
Consumption Geothermal
Tides
Uranium

Limited reserves 900 (56 years) Coal

Total reserves

Source: Richard Perez et al.

Page 14

PVsyst SA 37
 SOLAR GEOMETRY AND RESOURCE

Air Mass (AM)

sun reference

Optical Air Mass: at clear sky:

atmosphere depth crossed by the sun rays /
reference atmosphere depth
(vertical at sea level)

=> varies with the altitude and the sun's height:

SE
At first approximation: AM = Exp (-Altitude / 7800m) / SinHS

For HS < 10-15°: take the earth curvature into account !

R
Highlight: Atmospheric pressure is also exponential with the altitude,

U
P = Po * Exp (-Altitude / 7800m)
O
Page 15
C
Beam, Diffuse and Albedo irradiances
ST

Extraterrestrial irradiance (average 1367 W/m²)

Atmosphere limit
SY

Beam Diffuse
PV

Global irrad.
about 1000 W/m²
at clear sky
Albedo

Global = Beam + Diffuse + Albedo

Page 16

38 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Irradiance components
• Beam: parallel rays without atmospheric diffusion
produces shadows, can be concentrated
• Diffuse: photons diffused by the atmosphere (air, clouds, aerosols)
At covered weather: admitted isotropic
At clear weather: "blue sky" , rather isotropic
+ Bright crown around the sun (circumsolar, 5° cone)
+ Reinforcement on the horizon ("horizon band")
• Albedo: Reflected part from the ground, depends on the site environment

SE
Contributes to the irradiance on tilted planes only

• Normal Beam DNI: direct measured perpendicular to the sun's rays

R
When measured on a non-perpendicular plane,
the value is lower as it irradiates a greater area ("cosine effect").
• Clearness Index Kt: Ground irradiance, normalized to the extraterrestrial

U
(measurement of the atmospheric attenuation)
O
Page 17
C
Irradiance Spectrum
ST
Spectral irradiance [W/m² / nm]

SY

Radiation of a Black body at 5'900°K)

PV

Atmospheric water vapour absorption

Diffuse component (clear sky)

Wavelength [mm]
Visible spectrum

Photon energy

Page 18

PVsyst SA 39
 SOLAR GEOMETRY AND RESOURCE

Radiation spectrum
Energy of each photon: E = h·n = h·c / l
h = Planck constant, c = speed of light,
n = frequency, l = wavelength

AM0 : Extraterrestrial radiation spectrum

≈ radiation of a Black Body at 5'900 °K
AM1.5: Normalised spectrum for a clear sky
after crossing of an air mass of 1.5 (SH = 48° at seal level)

SE
Distribution of the energy AM 1.5 AM 0
Ultra-violet UV : 0.2 < l < 0.38 mm 1.7 % 6.4 %

R
Visible 0.38 < l < 0.78 mm 53.7 % 48.0 %
Infra-red IR 0.78 < l < 10 mm 45.6 %

U
44.6 %
Total O 832 W/m² 1367 W/m²

Page 19
C

Meteorological data
ST

The meteorological stations record:

 The global irradiance on the horizontal plane
SY

(base of calculation for any solar system)

 Very rarely the Diffuse irradiance on the horizontal plane (or the DNI)
 The ambient temperature (sheltered, 2 m of the ground, on 100 m²gras)
 The wind speed, and possibly the direction (normally at 10 m)
PV

We define the following quantities:

Irradiance [ W/m²] energy flux per unit of time and surface (power)
Irradiation [kWh/m²] integration over a given period (energy)
Sun Hours for very old databases (beam component)

These data are usually available in Monthly, Daily or Hourly values

Page 20

40 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Irradiance Mesurements
Pyranometer: measurement of the temperature rise of a black surface
under a double cupola of glass or quartz (greenhouse effect)
Wide spectrum sensitivity (=> 3mm).
Accuracy of the order of 1-2 % (when very well calibrated)
Price: ≈ 1'500 to 3'000 euro.

SE
Pyranometer
very large irradiance spectra (3 mm)
large angular acceptance

Glass cupolas

R
Thermo-electric
voltage

U
5 - 15 mV

Page 21
O
C
Irradiance Mesurements
ST

Photovoltaic detector:
Spectral response: according to the crystalline PV cell's sensitivity.
SY

Bad angular response (reflections, IAM)

=> Best accuracy of the order of 5%
Reference cell: PV detector calibrated and sold by the official center JRC/Ispra
(Price around 300 euro)
PV

Reference PV cell
Spectral response specific to technology
Angular response limited by reflexions
Reference PV cell structure
Open circuit voltage The datalogger computes :
(around 500 mV)
Electrical
- the irradiance (current) and
separation - the temperature of the cell (voltage)
Short-circuit current:
30 mA / 1 ohm = 30 mV

Page 22

PVsyst SA 41
 SOLAR GEOMETRY AND RESOURCE

Irradiance Mesurements
Satellite measurements:
Analysis of the radiation reflected by the high atmosphere
At different wavelenghts
By geostationnary or polar satellites
Uses more and more sophisticated models
and other measurements (cloud coverage, turbidity, humidity, snow)
Information by image pixel  km resolution
The more reliable data for anywhere in the present time !

SE
Heliograph (of Campbell-Stokes): Sun hours measurements!

R
A glass ball focuses the sun's rays on a paper, which darkens
Measurement of the beam radiation, with a threshold of about
150-250W/m² (depending on the paper humidity and operator ! )

U
O
Page 23
C
Meteo databases: monthly data
ST

Directly available within PVsyst:

You choose your site on a Google map, anywhere on the earth, and import the corresponding
meteo data from

METEONORM: Software (fully integrated in PVsyst V 6)

SY

• Monthly data for about 1200 "stations" ( ≈15 in each big European countries)
• Typical year, based on 1961-1990 averages, MN6: sometimes 1983-1993
• Any location obtained by spatial interpolation
• Satellite data when terrestrial stations not available
PV

NASA-SSE : Monthly Satellite data, rather rough data

• cover the full earth by 1° x 1° (111 km x 80 km) in latitude/longitude

Available from Web :

PV GIS: PhotoVoltaic Geographical Information System (JRC/Ispra)
Monthly values of Global, diffuse and temperature
• "Classic" database: Monthly values for Europe (terrestrial measurements) and
Africa (satellite)
• “Climate SAF” new database: recent data, based on satellite data (1998 –
2013) – Significant increase of Irradiation by respect to "Classic"

Page 24

42 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Meteo databases: hourly data

For Europe: "Satel-light" project:
• Measured data in half-hourly values, five years 1996 .. 2000
• For any pixel of about 5 x 7 km² in Europe
• Free download from www.satellight.com
• Accuracy: competitive with terrestrestrial measurements as soon as you are at more
than 20 km of the closest station.
For USA and Canada (free):
• TMY3, Typical Meteorological Years provided by NREL for 1'020 US stations
• EPW, Typical Meteorological Years (CWEC) for 72 stations in Canada

SE
• Solar anywhere (SUNY model): satellite recent data for US
• Solar prospector: NREL satellite hourly data 1998 – 2005, resol. 10 km

For the whole world (for pay)

R
• SolarGIS, 3Tiers, Helioclim - Recent hourly satellite data
• cover the full earth

U
O
Page 25
C
Databases – Monthly meteo
ST

Database Region Values Source Period Variables Availability PVsyst import

Meteonorm Worldwide Monthly 1'200 Terrestrial 1960-1991 GlobH Included in Direct by

SY

stations averages Temp. PVsyst V 6 site choice

(+ complement satellite) V 6.1: Wind
1995-2005 Others Software Direct
Interpolations (optional) by file
Europe: 566 stations 1981-1990

PVGIS-ESRA Europe Monthly Interp. 1x1 km² averages GHI, DHI, TA Web Direct
PV

"Classic" Africa Africa: Meteosat 1985-2004 Linke turbidity free by copy/paste

S-W Asia (Helioclim-1 database)

PVGIS Europe + Afr. Monthly Meteosat and 1998 - 2010 GHI, DHI, TA Web Direct
"Climate SAF" 0° - 58° N EUMETSAT, 3x3 km² Linke turbidity free by copy/paste

NASA-SSE Worldwide Monthly Satellites 1983-1993 GlobH Web Included in

1°x1° cells (111 km) averages Temp free PVsyst

Helioclim - 1 Europe Monthly Meteosat 1985-2005 GlobH Web Direct

(SoDa) Africa 50x50 km² (each year) no temper. 1985-89 free by copy/paste
Hourly
WRDC Worldwide Daily 1195 stations 1964-1993 GlobH Web Direct
Monthly (each) no temper. free by copy/paste

RETScreen Worldwide Monthly Compil. 20 sources 1961-1990 GlobH, TA Software, Direct

incl. WRDC - NASA (averages) WindVel free by copy/paste

Page 26

PVsyst SA 43
 SOLAR GEOMETRY AND RESOURCE

Databases - Hourly meteo

Database Region Values Source Period Variables Availability PVsyst import

1960-1991
Meteonorm Worldwide Hourly Synthetic and GHI, DHI, TA Included in Direct by
generation 1995-2000 VindVel PVsyst V 6 site choice
Meteosat
Satellight Europe Hourly Any pixel of about 5 years GlobH Web Direct
5x7 km² 1996-2000 no temper. free by file

US TMY2/3 USA Hourly NREL, 1020 stations 1991-2005 GHI, DHI, TA Web Direct
Typical Meteo Years (samples) WindVel free by file

EPW Canada Hourly CWEC, 72 stations 1953-1995 GHI, DHI, TA Web Direct
Typical Meteo Years (samples) WindVel free by file

SE
ISM-EMPA Switzerland Hourly 22 stations 1981-1990 GHI, DHI, TA Included in Included in
Design Ref. Years (samples) WindVel PVsyst database
Solar GlobH, DiffH Web
Anywhere USA Hourly Satellites 1998 - today Temp. for pay 2010 - today Direct
(SUNY model) + Hawaii by 10x8 km² Wind for pay for pay by file
Web Direct
Helioclim - 3 Europe Hourly Meteosat from 02/2004 GlobH For pay by copy/paste

R
(SoDa) Africa => today no temper. 2005 free or PVsyst format

SolarGIS Worldwide Hourly Meteosat, ERA 1994 - today GHI, DHI, TA For pay Direct (file)
(Geomodel) PVsyst format

U
GHI, DHI
3Tiers Worldwide Hourly Satellites 1998 - today DNI available For pay Direct (file)
Spectroradiometer MODIS no temper. PVsyst format
O
Page 27
C
Comparison between databases
ST
SY
PV

See also: Global irradiation: average and typical year, and year to year annual variability
Pierre Ineichen, March 2011 (www.pvsyst.com)

Page 28

44 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Year-to-year variability
The Variability is the Variance (RMS) of the Gaussian distribution of yearly irradiations
Studied with data of 10-30 years in 30 locations
Depends on the irradiation (very sunny climates have lower ranges of variation)

Variability as function of Annual irradiation

8%

7%
Default values for
6%
P90 evaluation
5%
Variability

SE
4%

3%

2%

1%

0%

R
800 1000 1200 1400 1600 1800 2000 2200 2400
Annual irradiation [kWh/m²]

U
Ref: Global irradiation: average and typical year, and year to year annual variability
(Pierre Ineichen, ISE, University of Geneva, 2011), available on www.pvsyst.com O
Page 29
C
Climate change
ST

Since beginning of the years 2000, we observe an increase of the irradiation

of around 5% in whole Europe.
cf PVGis: evolution between classic and SAF databases:
http://re.jrc.ec.europa.eu/pvgis/PVGIS_new_features.html
SY

Example :
evolution of the irradiation in Geneva
ISM measurements
PV

Geneva-Cointrin - Yearly variation

15%
Geneva-Cointrin - Yearly irradiance
1400
1993-2002 average = 1197 kWh/m ² yr
10%
1350

1300 5%
kWh/year

1250
0%
1200

1150
-5%
2003-2012 average
1100
1317 kWh/m ² yr : + 10%
1050 -10%
DRY
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012

1000
DRY
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012

Page 30

PVsyst SA 45
 SOLAR GEOMETRY AND RESOURCE

Production previsibility – P50-P90

Probabilistic approach for the interpretation of the simulation results over several years:
Hyp: the year-to-year yield of the system will follow a Gaussian distribution
P50 (or P90) is the yield which will be attained in more than 50% (90%) of years.
This determination requires some hypothesis from the user !

P50 determination :
P50 is the mean value of the Gaussian distribution
Mainly determined according to the Meteo data used for simulation

SE
 If meteo data = average year (multi-year monthly averages or TMY):
=> The simulation result is the P50 value
=> for old databases: may be corrected for a climate change hypothesis

R
 If meteo data = one measured year
=> no possibility of defining a probabilty profile (no P50 nor P90 determination)

U
Except when we know the position of this year with respect to the average :
=> P50 value will be Yield (simul.) + deviation of this year
O
Page 31
C
Production previsibility – P50-P90
ST

Variability ( RMS or s ) determination

- Based on the observed Meteo data variability
- Increased by other uncertainties of the simulation process
SY

(PV module parameters, inverter, soiling, degradation): each uncertainty is quantified

by a RMS probability, which are added quadratically.
P90 yield evaluation = P50 - 1.28 * s
PV

Page 32

46 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Satellite measurements: accuracy

Accuracy of hourly values:
better as soon as the site is 20 km apart from a terrestrial station !

SE
R
U
Distance between 2 ground stations (km)
O
Page 33
C
Side by side comparison: Jungfraujoch station
ST

• ASRB station and CLIMAP station (MeteoSwiss)

• comparable Gh with 12% dispersion (42 Wh/m2h)
• comparable Bn with very high dispersion (extreme conditions)
SY

Hourly horizontal global irradiance Gh [Wh/m2h]

1200
Gh 365
Measurements (CLIMAP Jungfraujoch 2000) nb 3971
mbd 4 1%
1000 rmsd 42 12%
sd 42 12%
PV

R2 0.99

800

600

400

200

Measurements (ASRB Jungfraujoch 2000)

0
0 200 400 600 800 1000 1200

Page 34

PVsyst SA 47
 SOLAR GEOMETRY AND RESOURCE

Measurements reliability
Even when recorded by specialized institutions, irradiance measurements
may show surprising results !

Oct 2003 - Sep 2004:

ISM-Cointrin, calibration
error of 7% on the
solarimeters
(corrected here)

SE
R
ISM: Swiss Institute for Meteorology Measurement stations
BSRN : Baseline Surface Radiation Network 300 m apart !

U
O
Page 35
C
Models chaining
ST

Horizontal irradiance (Meteo)

Monthly data Hourly data

SY

Synthetic generation

Model for diffuse

PV

Hourly Global et Diffuse

Solar Geometry
Transposition
on tilted plane

Global, Diffuse, Albedo on plane

Horizon, near shadings

Optical effects (IAM)

PV Array
(useful energy)

Page 36

48 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Synthetic Model: Monthly  Hourly

Generation of synthetic hourly values
• The detailed simulation requires hourly values
• From monthly data (12 values for irradiation and temperature):
Creates a random sequence of hourly values (G and T)
with a distribution similar to a real data distribution
Sum and average values correspond to the specified monthly data

SE
• Irradiances: sequences built using Markov Matrices
Determined using data of more than 30 sites
Model established by Collares-Pereira (1988)

• Temperatures: irradiance dependence matrices 

R
Sinus-like daily behaviour with amplitude according to daily irradiance,
Phase shift of 2-3 hours depending on the climatic situation

U
O
Page 37
C
Diffuse model
ST

Model for the diffuse irradiance (from Global)

• Required for the transposition and the evaluation of shadings
SY

• In Hourly values :
 Liu-Jordan or Erbs correlations: Simple and steady models
estimation of the Diffuse/Global ratio as function of Kt (i.e. global)
PV

 Perez-Ineichen model: more sophisticated,

takes the kind of weather into account (hour before and after)
better only when original data for global are well recorded
(not adequate for synthetic hourly data)

Page 38

PVsyst SA 49
 SOLAR GEOMETRY AND RESOURCE

Measurement and Model for Diffuse

Ring for masking direct irradiance,
to be regularly adjusted
(about every 5 days).

Erbs model for the diffuse

Mesurements (Jonction 2004)
1.2
Erbs model

SE
1.0

Diffuse / Global
0.8

0.6

R
0.4

0.2

Applicability of the

U
0.0
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90
Erbs correlation Clearness Index Kt

Page 39
O
C
"Clear sky" model
ST
SY
PV

• Simple (PVsyst): Global according to air mass

• More sophisticated: Requires Humidity and aerosols ("Linke" trouble coefficient)
Advantage: evaluation of the maximal irradiance at any location
(requires only site coordinates)
Page 40

50 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Irradiance on tilted plane

Normal to plane

Collector plane
West

East
South

SE
Tilt b: angle with respect to horizontal
Azimuth g : orientation with respect to South ( > 0 towards West)

R
Incidence angle a (often noted i) :
angle between the normal to the plane with respect to sun rays

U
cos a = cos b · sin HS + sin b · cos HS · cos (AZ-g)
O
Page 41
C
Transposition model
ST

Circumsolar
(cone 5°)
Isotropic diffuse
SY
PV

Horizon band (about 5°)

Global P = BeamP + DiffuseP + AlbedoP

Page 42

PVsyst SA 51
 SOLAR GEOMETRY AND RESOURCE

Hay transposition model

The transposition applies differently to each component:

Beam: purely geometrical transformation :

BeamP = BeamH · cos a / sin HS
Isotropic diffuse: part of sky "seen" by the collector:
DiffP iso = DiffH iso · (1 + cos b) / 2
Circumsolar diffuse: fraction of the diffuse (determined by Hay model)

SE
behaves as the beam component
Horizon diffuse: Not taken into account in the Hay model
Albedo: Proportional to the Horizontal Global

R
(albedo coefficient r )
and to the ground fraction "seen" by the collector:

U
AlbP = r · GlobH · (1 – cos b) / 2
O
Page 43
C
Perez-Ineichen transposition model
ST

More complex model, established according to a lot of climates

• Beam : Geometrical, identical
SY

• Diffuse: The sky is patitioned into different zones,

Involves specific coefficients for each zone,
depending on the "quality" of the sky
PV

• Albedo: Similar treatment as for the Hay model

• PVsyst  V5.74: Hay model by default (Perez-Ineichen optional)

Since V6: Perez-Ineichen model is the default (settable in "preferences")
NB: the Perez model gives slightly higher annual values
(0 to 2 % ) depending on the climate and the plane orientation.
Ref: "Global irradiance on tilted and oriented planes: model validations", P. Ineichen, 2011,
Research report of the Institut of the Environnemental Sciences, University of Geneva,
available from www.pvsyst.com / publications

Page 44

52 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Orientation optimisation
Transposition Factor: Global on plane / Global horizontal ratio

Optimisation: Graphs of the Transposition Factor according to orientation,

depends on the PV system use !

Grid systems: maximize the annual production

Stand-alone systems: sizing on the critical periods
(winter, vacations, water needs for pumping, etc...)

SE
 PVsyst calculates transposition tables and graphs :

R
for a given site / meteo,
with choice of the period (months),
with possible horizon shading

U
O
Page 45
C
Example of annual TF graph
ST
SY
PV

Page 46

PVsyst SA 53
 SOLAR GEOMETRY AND RESOURCE

Example of TF graph for winter

SE
R
U
O
Page 47
C
Tracking planes
ST

• Mounting of the panels on tracking devices :

Tilted axis : N-S orientation, tilted according to the latitude
SY

Horizontal N-S axis: identical, daytime following

Horizontal E-W axis: follower acc. to sun height (season), not pertinent for PV
Vertical axis: fixed tilt plane, following the sun's azimuth
2-axes: heliostats, more complex mechanics
PV

• In any case: the course limitations should be taken into account

• Fields of trackers: carefully study the mutual shadings
(much more critical than for sheds)
• Tracking systems are suited for very sunny climates
(with strong beam component)
• Gains of the order of 30%, 35% in the best cases
• Winter gains enhanced

Page 48

54 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Sheds disposition

3 significant parameters:
• Pitch
• Sensitive (coll. band) width
(with inactive bands)
• Plane Tilt

 calculation of the
limit shadowing angle

SE
R
U
O
Page 49
C
Sheds mutual shadings
ST
SY
PV

Sheds towards South: shadings in the winter mornings and evenings

Usual sizing : limit angle = sun height on 21 december at noon.

Page 50

PVsyst SA 55
 SOLAR GEOMETRY AND RESOURCE

Sheds mutual shadings

SE
R
South-East or South-West: shading effects more important !

U
O
Page 51
C
Sheds Optimisation
ST
SY
PV

Electrical loss

Page 52

56 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Shading loss factor

Loss factor according to the profile

angle, one string in width
Total electrical loss as soon as the
bottom cell is fully shaded

SE
Loss factor for 3 strings

R
in the width of the shed

U
O
Page 53
C
Utilization ratio with respect to Ground
ST

The shading is calculated according

to the Profile angle, i.e. the angle
Shading limit between the plane passing through
SY

the shed's base and the sun, and the

horizontal.
PV

The utilization ratio = L/P depends

mainly on the collector tilt :
About. 45% at 30°
80% at 5°

Page 54

PVsyst SA 57
 SOLAR GEOMETRY AND RESOURCE

Sun shields in facade

SE
R
U
O
Page 55
C
Mutual shadings of sun-shields
ST
SY
PV

Page 56

58 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Backtracking
Standard Tracking:
• The distance between trackers is more
critical than for sheds
• because the useful irradiance is higher
when the sun height is low
• Electrical effect of strings is very
important

SE
Back-Tracking :
• Orientation in such a way that there are

R
no mutual shadings
• Avoids the electrical loss due to strings

U
• But incoming Irradiance is about the
same !!! O
Page 57
C
4 years of irradiance at Geneva (1978-82)
ST

Historical example of data recording 35 years ago… (P. Ineichen Thesis)

kWh/m²/jour

SY
PV kWh/m²/jour

One point = energy of one day [kWh/m²/day]

Page 58

PVsyst SA 59
 SOLAR GEOMETRY AND RESOURCE

Clear sky GlobHor versus latitude

Altitude 4000 m
Clear sky model, Linke = 3.5
Altitude 2000 m
3000
Altitude 1000 m
Altitude 0 m
Full year of the "Clear sky" model
2500 Diffuse clear sky
=>
Irradiance [kWh / m2 / year]

2000 Factor of 3 between Equator and Poles

1500

1000

500

0 Clear sky, altitude enhancement

-90 -60 -30 0 30 60 9040%

SE
Latitude
Altitude 4000 m
35%
Altitude 2000 m

Yearly Irradiance enhancement

30% Altitude 1000 m

25%
Effect of the altitude
20%

on the yearly clear sky

R
15%

10%

U
5%

0%
-90 -60 -30 0 30 60 90
O Latitude

Page 59
C
Horizontal irradiation versus latitude
ST

The highest irradiations are found in the tropical regions (lat. 20-30°, several deserts)
Varies by less than a factor of 2 between Geneva (1'250) and Tamanrasset (2'360 W/m²)
Optimale use for PV: Tracking in desertic regions
SY

Fixed tilted plane at middle latitudes

Meteo data and Clear sky model Altitude 2000 m

3000 CS, Altitude 0 m
PV

Equator
Diffuse clear sky
Meteo, Global
2500
Model, Diffuse
Irradiance [kWh / m2 / year]

2000

1500

1000

500

0
-90 -60 -30 0 30 60 90
Latitude

Page 60

60 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Resource versus latitude (clear sky)

Use on Horizontal plane:
Ex: Latitude 45° :
Summer 8.38 kWh/m²/day
Winter 1.66 kWh/m²/day (Factor of 5)

SE
Use on Tilted plane 30°
Ex: Latitude 45° :

R
Summer 8.09 kWh/m²/day
Winter 3.48 kWh/m²/day (Factor of 2.3)

U
O
Page 61
C
Shadings
ST

We distinguish between 2 kinds of shadings :

Far shadings (Horizon line)
SY

• Far enough to affect the total surface of the field at a given time (ON-OFF).
Criteria: about 10 x the PV system size
• Beam OFF when the sun is behind the horizon
• Diffuse not much affected (except when the obstacle is relatively close)
PV

Near shadings: partial shadows

• Require a 3D representation of the field and its environment
• Beam: "Linear" shading factor: shaded fraction of the field, irradiance deficit
"Acc. to modules strings": maximum limit of the electrical effect
"Real electrical effect": requires Module Layout – strings description
• Diffuse: Factor as an integral of the Linear shading factor over the part
of the sky portion as "seen" by the collectors
doesn't depend on the sun position  constant over the year !
• Albedo: Idem, integral on the part of the ground "visible" for the collectors

Page 62

PVsyst SA 61
 SOLAR GEOMETRY AND RESOURCE

Horizon measurements
Made on the site: Photographs
 Camera with a "fisheye" lens
=> a circular 360° picture can be superimposed on the sun paths polar graph
generated by PVsyst
Example: SunEye apparatus (www.solmetric.com, www.soleg.de)

 Several pictures of the horizon (panoramic)

Difficulty: how to define the horizontal plane and the vertical scale ?

SE
Example: Complete system for recording the horizon line:
• Panorama Master 2.0: Tripod with a rotative calibrated part for your camera
Allows to take a series of "normalized" pictures on 360°
• Horiz'on 2.0: The software merges the pictures as a continuous panorama

R
Allows to define a horizon profile (height/azimuth points) with mouse
Saved as a file to be imported in PVsyst

U
Ref: http://www.energieburo.ch/web/produkte/horizon (135 € + 199 €)

O
Page 63
C
Horizon measurements
ST

Record some reference points:

 Detailed map
Identify all interesting points (distance, azimuth, height): tedious
SY

 Geometer instrument: some reference points, may be reported on a picture

 If not available: Azimuth: Compass Hanging

PV

Height: Clisimeter (tilt meas.) Aimed point

"Emergency" instrument: wooden cross,

q
with analysis in EXCEL sheet B
H
M
[mm]

 GIS software Angle q = 90 - Acos ( ( B*B + H*H - M*M) / (2H * B) )

Ex: Carnaval (for Europe only)

The resolution should be very high
(< 100-200 m) otherwise the "higher" peaks may be "unseen"

Page 64

62 PVsyst SA
 SOLAR GEOMETRY AND RESOURCE

Inputs for near shadings

To build the 3D scene:
• Means described for the Horizon are not enough
• The view is different from each point of the field
=> the height/azimuth (horizon line) recording is not useable
• Requires the distance and sizes of the shadowing objects
• Requires the absolute height (or with respect to the array)
• The architect's plan (with surrounding houses)
can be used

SE
but should be completed by real height references

Electrical losses calculation:

• The calculation "according to modules" is a simplified tool,

R
gives an upper limit for the electrical losses.
• Detailed electrical loss calculation:

U
requires the exact definition of the position, the orientation, the structure
and the connections of the PV modules ("Module Layout" option).
O
Page 65
C
Conclusions
ST

• Knowledge of the irradiance is of crucial importance for the evaluation of

the PV system yield
SY

• Often rough data, given as monthly averages, without diffuse part

• Discrepancy between different data sources, availability according to regions
Historical averages , year-to-year variability,
Significant evolution since beginning of the years 2000
PV

Terrestrial measurements: not for the exact location, measurements quality

Satellite measurements: model's uncertainties
• Chaining of several models for getting the hourly irradiance on the fields :
Synthetic generation, Diffuse model, Transposition model
• Far shadings: rather easy to establish and treat
• Near shadings: require a 3D construction,
Geometrical data (achitect plans, on-site measurements)
Produce electrical mismatch losses according to module layout/connection

Page 66

PVsyst SA 63
 SOLAR GEOMETRY AND RESOURCE

SE
R
U
O
C
ST
SY
PV

64 PVsyst SA
 SOLAR GENERATOR

USE OF PVSYST FOR THE STUDY OF GRID-CONNECTED SYSTEMS

SE
SOLAR GENERATOR :
PV MODULES, INVERTERS, SYSTEMS

André Mermoud
andre.mermoud@pvsyst.com

R
U
PVSYST SA - Route du Bois-de-Bay 107 - 1242 Satigny - Switzerland
www.pvsyst.com
O
Any reproduction or copy of the course support, even partial, is forbidden without a written authorization of the author.
C

Contents
ST

• Modelling – Standard one-diode model (used in PVsyst)

SY

• Model validation
• Amorphous modules modelling
• Incidence correction
PV

• Module temperature
• "Hot-Spot" – Protection diodes
• Usual available modules
• Inverters
• Sizing PV array – Inverter
• Use of PVsyst: project, simulations and results

PVsyst SA 65
 SOLAR GENERATOR

Modelling the PV module

Objective: Reproduce the electrical behaviour of the module
Construct the I/V curve, and Vmpp, Impp, Vco, Isc values
for any irradiance and temperature conditions
1000 W/m2 25.3°C
I/V curves, 25°C
700 W/m2, 25.3°C
9 400 W/m2, 25.3°C
200 W/m2, 25.3°C
8
100 W/m2, 25°C
24.5°C
7
Model I/V curves, 1000 W/m2
35.1 °C
9 45.4 °C
6
Courasnt [A]

54.8 °C
8

SE
59.5 °C
5
Model
7
4
6

Courasnt [A]
3
5
2
4
1

R
3
0
0 5 10 15 20 25 30 35 2
PVsyst SA Tension [V]
1

U
0
0 5 10 15 20 25 30 35
PVsyst SA Tension [V]

Page 3
O
C

"Standard" 1-diode model

ST

The PV cell may be represented by the equivalent electrical circuit:

SY

PV cell V + I * Rs Use (load)

Rs
I
Photocurrent Diode
PV

Iph Rsh V RL

PVsyst SA

I = Iph - Io [ exp (q · (V+I·Rs) / ( Ncs·g·k·Tc) ) - 1 ] - (V + I·Rs) / Rsh

Photocurrent Current in the diode Current in Rsh

Page 4

66 PVsyst SA
 SOLAR GENERATOR

"Standard" 1-diode model

Module I/V Characteristics
0.9
Photocourant
0.8
R shunt
0.7

0.6
FF = (Vmp*Imp) / (Vco*Isc)
Current [A]

0.5
R serie
0.4

0.3 1000 W/m², 25°C

SE
Isc, (Vmp,Imp), Vco
0.2
R shunt
0.1
R serie

0.0

R
0 20 40 60 80 100 120
PVsyst SA Voltage [V]

U
I = Iph - Io [ exp (q · (V+I·Rs) / ( Ncs·g·k·Tc) ) - 1 ] - (V + I·Rs) / Rsh

Page 5
O
C

"Standard" 1-diode model

ST

Variables at the module terminals

I = Current generated by the module [A].
SY

V = Voltage [V].
5 Parameters to be determined
Iph = Photocurrent [A], proportional to the irradiance F,
Io = Reverse saturation diode current, depends on temperature
PV

Rs = Series resistance [W].

Rsh = Shunt resistance [W].
g = Diode ideality factor, normally between 1 and 2.
Constants
q = Charge of the electron = 1.602 · 10-19 Coulomb
k = Bolzmann constant = 1.381 · 10-23 J/K.
Ncs = Number of cells in series
Tc = Effective temperature of the cell [Kelvin]
q/kT = 26 mV at 300 K

Page 6

PVsyst SA 67
 SOLAR GENERATOR

Standard model parameters

To determine these 5 parameters (Iph, Io, Rsh, Rs, g) :
The measurement of one complete I/V curve at (Gref, TRef) is sufficient !

3.00
Shell ST40 Rsh is determined by the inverse
of the slope around Isc (V=0)
2.50

2.00
The I/V characteristics equation,
written at each 3 points given in any
Current [A]

1.50 specification at STC:

SE
Measured caracteristics)

1.00 Model, RS = 0
(0, ISC) (Vmp, Imp) (Vco,0)
Model, RS optimal

0.50 Model, RS = RSMax

gives 3 equations , the resolution of
Pmax, Isc and Vco
0.00
which leaves one free parameter, for

R
PVsyst SA
0 5 10
Voltage [V]
15 20 25
example Rs.

U
=> For a given Rs value, the remaining parameters Iph, Io et g are determined
giving the complete I/V curve (at STC) O
Page 7
C

Determination of Rserie
ST

Sigma Error (Model - Measurements) on Current

Using the measured I/V curve : 70.0

SY

60.0

=> Determination of Rs: 50.0

Value easily obtained by minimizing the

Sigma [mA]

40.0

(mes.- model) errors 30.0

20.0

NB: The min. RMS (Imes – Imodel)

PV

10.0

is usually of the order of 0.4% of Isc. 0.0

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80
PVsyst SA Serie resistance [ohm]

Using the manufacturer specifications: Isc, Vco, Imp, Vmp

at STC, i.e. Gref = 1000 W/m² Tcref = 25°C AM = 1.5 (AM = Air mass)
=> we have to do hypothesis on Rshunt and Rseries !

Page 8

68 PVsyst SA
 SOLAR GENERATOR

Different (G, T) conditions

The model has been established for reference conditions:
• Gref = Irradiance when performing the measurement
• Tcref = Cell temperature during measurement

The photocurrent Iph is proportional to the irradiance

Iph = ( G / Gref ) · [ Iph ref + mISC · (TC - TC ref) ]
(small temperature dependence: mISC  0.05%·ISC / °C)

SE
The diode reverse saturation current Io varies with temperature:

R
Io = Io ref ( TC / TC ref )3 · exp [ ( q · eG / g · k) · ( 1/TC ref - 1/TC ) ]
(where eG = gap energy of the semiconductor material)

U
O
Page 10
C

Model Validations – crystalline modules

ST

The model should reproduce the real module behaviour

in any irradiance and temperature conditions
The model is established using one measured I/V characteristics (Vmp, Imp, Vco, Isc)
SY

+ 3 parameters to be adjusted Rshunt, Rsh(0), Rserie

• The model is compared to all measured I/V curves
• The model quality is estimated by the following indicators
PV

MBD: m = S (Val. mes – Val. model) / Nmes

RMSD: s = SQRT [ S (Val. mes – Val. model)2 / Nmes ]

Pmax Error, Meas - Model vs GlobP

4.0

3.0
Pmax Error (Meas-Mod) [W]

2.0
Ex: Si-mono 53 Wc module
1.0

0.0 standard "original" model

m = 2.0% s = 1.3%
-1.0

-2.0
(% of Pnom)
-3.0
0 < Tmod < 80°C Model
-4.0
0 200 400 600 800 1000
GlobP [W/m2]
PVsyst SA

Page 11

PVsyst SA 69
 SOLAR GENERATOR

Rshunt Correction
The "standard" model supposes a constant Rsh.
Then we measure an exponential-like distribution:
Siemens M55 - R shunt function of Irradiance
1400
Measurements
1200
Parametrization
R Shunt measured

1000

800

600

SE
400

200

0
0 200 400 600 800 1000
PVsyst SA Irradiance [W/m²]

R
Rsh = Rsh (GRef) + [ Rsh(0) – Rsh(Gref) ] * exp (- Rsh exp · (G / Gref))

U
( RshExp fixed at 5.5 for almost all module) O
Page 12
C

Model with Rshunt correction

ST

Pmax Error, Meas - Model vs GlobP

4.0

3.0
Ex: Si-mono 53 Wc module
standard model + Rsh correction:
Pmax Error (Meas-Mod) [W]

SY

2.0

1.0

0.0
Effect on Pmax (% of Pnom STC)
-1.0 Without corr: m = 2.0 % s = 1.3 %
-2.0
With corr: m = 0.4 % s = 0.7 %
-3.0
0 < Tmod < 80°C Model
PV

-4.0
0 200 400 600 800 1000
PVsyst SA GlobP [W/m2]

Pmax Model vs Pmax measured

45

40

35
Pmax model [W]

30

25

20

15
Effect on Vco:
Without corr: m = 3.1 % s = 6.1 %
10

corr: m =-0.2 % s = 1.0 %

0 < Tmod < 80°C Model
0 With
0 5 10 15 20 25 30 35 40 45
PVsyst SA Pmax measured [W]

Page 13

70 PVsyst SA
 SOLAR GENERATOR

With manufacturer's specifications

Don't confuse
Model accuracy and Parameters accuracy !!!
A model is of quality: if we can find parameters for reproducing
the behaviour in any situation
Manufacturer's Parameters : may be not representative of the module

Pmax Model vs Pmax measured

…. but these parameters are usually

SE
50

45

40 the only ones available !

35
Pmax model [W]

30
Ex: Si-mono 53 Wc
25

R
20
According to manufacturer:
15

10 m = -4.5 % s = 4.0 %

U
5
0 < Tmod < 80°C
0 at 1000 W/m² : -16%
0 5 10 15 20 25 30 35 40 45 50
PVsyst SA Pmax measured [W] O
Page 14
C

Amorphous modules modelling

ST

SHR-17 (old) - I/V Characteristics Measurements

2.5 Model For one measured characteristics, it is
Max. power
Ginc = 881 W/m²
always possible to find parameters to
SY

fit the curve (errors s(I) < 0.4%),

2.0

Ginc = 680 W/m²

Current [A]

1.5
Ginc = 501 W/m² but the parameters are not the same
1.0 from curve to curve !
Ginc = 329 W/m²
PV

0.5 Ginc = 209 W/m²

0.0
0.0 2.0 4.0 6.0 8.0 10.0 12.0
PVsyst SA Voltage [V]

Amorphous modules: require 3 corrections of the standard model:

• Rshunt correction – strong importance as Rshunt very low
• Recombination current in the intrinsic layer – i –
• Spectral sensitivity correction

Page 15

PVsyst SA 71
 SOLAR GENERATOR

Amorphous: Rshunt behaviour

SHR-17: R shunt function of Irradiance

1000

Measurements
800 Parametrization

R Shunt measured [ohm] 600

400

200

SE
0
0 200 400 600 800 1000

PVsyst SA Irradiance [W/m²]

R
Rshunt at STC is far lower than for crystalline modules
But with very high dynamics: Rsh(0) / Rsh(STC)  12

U
Page 16
O
C

Recombination correction
ST

Adds a new parameter in the I/V equation [Merten et al]

SY

I
Rs

Photo- Recom-
courant binaison Diode Utilisateur

Rsh V RL
PV

Iph Irec (Iph, V)

PVsyst SA

Voc Model vs Voc measured Voc Model vs Voc measured

14.0 14.0
Parameter
13.0
d²mt 13.0
Voc Model [V]
Voc Model [V]

12.0 12.0

11.0 11.0

10.0 Effect on the 10.0

9.0
Vco voltage 9.0
9 10 11 12 13 14 9 10 11 12 13 14
PVsyst SA Voc measured [V] PVsyst SA Voc measured [V]

Page 17

72 PVsyst SA
 SOLAR GENERATOR

Spectral correction
Correction proposed by CREST [University de Loughborough]
• Characterization of the photon energetic contents (APE – Average Photon Energy)
according to the Air Mass and Clearness Index Kt
• "UF = Utilization factor: convolution with the spectral sensitivity of each technology
(amorphous: eG = 1.7 eV)

Utilization Factor a-Si:H 0.670-0.700

0.640-0.670

SE
0.610-0.640
0.70
0.67
0.580-0.610
0.550-0.580 This parametrization is based on :
0.64 0.520-0.550

0.61
0.490-0.520 • The Loughborough climate
UF a-Si

0.460-0.490
0.58
0.55
• Computed for amorphous, single junction
0.52

R
0.49
0.46 It applies to the Photocurrent (Isc)
0 1
0.2 0.4 2
3

KTc
0.6
0.8
1 6
5
4
Air Mass
Improvement of the order of 10 % of the
s (meas – simul) value (low effect)

U
PVsyst SA

Page 18
O
C

Model for Amorphous and CIS

ST

Pmax M ode l v s Pmax me asure d

16.0

14.0
Ex: Unisolar SHR-17 shingle, Tripple junction
Standard model + all corrections
SY

12.0
Pmax model [W]

10.0

8.0 Pmax error m = -0.1 % s = 1.9 %

m = 0.1 % s = 0.7 %
6.0

4.0
Vco error
2.0 Isc error m = -0.8 % s = 2.1 %
PV

0 < Tmod < 80°C Model

0.0
0 2 4 6 8 10 12 14 16
PVsyst SA Pmax measured [W]

45
Pmax M ode l v s Pmax me asure d
Ex: CIS Shell ST40 module
40

35
Standard model + exponential Rsh corr.
Pmax model [W]

30

25

20
Pmax error m = 0.0 % s = 1.0 %
15

10
Vco error m = 0.0 % s = 0.5 %
5

0
0 < Tmod < 80°C Model Isc error m = 0.0 % s = 1.7 %
0 5 10 15 20 25 30 35 40 45
PVsyst SA Pmax measured [W] => CIS obeys perfectly to the standard model !

Page 19

PVsyst SA 73
 SOLAR GENERATOR

Long-term evolution
PVsyst SA Unisolar SHR-17 Wp Seasonal effect
10%
Unisolar SHR-17 (a-Si:H tripple junction):
8% over 6 years :
6%
Pmax error m = 0.7% s = 2.8%
Pmax (Meas - Model) difference

4%

2% Seasonal annealing not taken into account

0%
in the model
-2%

-4%

-6%

-8%

-10%
PVsyst SA Shell ST40 - CIS Seasonal effect
déc 04
mars 05

déc 05
sept 04

juin 05

mars 06
sept 05

juin 06

déc 06
mars 07
sept 06

juin 07

déc 07
mars 08
sept 07

juin 08

déc 08
mars 09
sept 08

juin 09

déc 09
mars 10
sept 09

juin 10
6 years
5.0%

SE
4.0%

3.0%

Pmpp (Meas - Model) error

2.0%

1.0%

0.0%

R
-1.0%

-2.0%

-3.0%

Shell ST40 (CIS) over 6 years :

U
-4.0%

Pmax error : m = 0.2% s = 1.0 %

-5.0%
déc 04

déc 05

juin 06

déc 06

déc 08
sept 04

mars 05
juin 05
sept 05

mars 06

sept 06

mars 07
juin 07

déc 07
sept 07

mars 08
juin 08

déc 09
sept 08

mars 09
juin 09
sept 09

mars 10
juin 10
Page 20
O
C

Summary of Meas. – Model comparisons

ST

PVsyst SA Error on Pmax Error on Voc Error on Isc

SY

-6% -4% -2% 0% 2% 4% 6% -6% -4% -2% 0% 2% 4% 6% -6% -4% -2% 0% 2% 4% 6%

Si-mono: Siemens M55, 1 year

Si-mono: Atlantis M55, 2.6 years

Si-poly: Kyocera, 5 years

CIS: Shell ST40, 6 years

PV

CdTe: First Solar FS267, 1.5 year

Si-a:H single: Flexcell, 1 year

Si-a:H tandem: EPV-40, 2.5 years

a-Si:H tripple: Unisolar SHR17, 1 year

idem, 6 years

a-Si:H tripple: Unisolar US32, 2.3 years

Microcryst: Sharp NAF121-G5, 7 months

Long-term measurements of modules of any technology ,

in any irradiance and temperature conditions

Page 21

74 PVsyst SA
 SOLAR GENERATOR

Uncertain parameters of the model

Crystalline and CIS modules:
Rshunt Effects on the low-light efficiency :
Rserie High (bad) Rserie  increases low-light efficiency
Rsh (0) Rshunt exponential  increases low-light efficiency

Amorphous and CdTe modules:

idem + Recombination parameter di²mt

SE
R
U
PVsyst SA PVsyst SA
O
Page 22
C

Effect of Rseries on low-light efficiency

ST

Efficiency f(Irrad) for different Rseries

17% Consider a module designed at low
irradiance.
SY

16%
R * I² loss Resistive Loss goes with
Rs * I² or Rs * Irrad²
Efficiency

15%

14%
Design Pmpp
A "good" module will have
13% Good Rserie = 0.3 ohm
higher STC performances
PV

Bad Rserie = 0.5 ohm

12%
0 200 400 600 800 1000 1200
PVsyst SA Irradiance [W/m²]
Efficiency f(Irrad) for different Rseries
17%

16%

But you buy the STC performances:

Efficiency

15%

For identical STC with bad Rs you

14%
should construct a much better Sold Pmpp at STC

module ! 13% Good Rserie = 0.3 ohm

Bad Rserie = 0.5 ohm

Therfore the good modules for Rs 12%
0 200 400 600 800 1000 1200
has a bad low-light performance ! PVsyst SA Irradiance [W/m²]

Page 23

PVsyst SA 75
 SOLAR GENERATOR

Effect of Rshunt on low-light efficiency

17%
Efficiency f(Irradiance) for diff. Rshunt Crystalline modules:
The shunt resistance has very small
16%
effects
15%
But the exponential Rshunt behaviour
Efficiency

Rsh=400, Rsh(0)=1600 ohm

14%
Rsh=300, Rsh(0)=1200 ohm enhances the low-light performance
Recovery due Rsh=400, Rsh(0)=400 ohm
to exponential Rsh=300, Rsh(0)=300 ohm
13% Rsh

12%
0 200 400 600 800 1000 1200
PVsyst SA Efficiency f(Irradiance) for diff. Rshunt - amorphous
Irradiance [W/m²]

SE
7%

6%

5%
Recovery due to
exponential Rsh(G)

Efficiency
Amorphous modules: 4%

Rsh=100, Rsh(0)=1200 ohm

the recovery of exponential is even 3%

R
Rsh=60, Rsh(0)= 720 ohm
more important ! 2%
Rsh=100, Rsh(0)=100 ohm
1%
Rsh=60, Rsh(0)=60 ohm

U
0%
0 200 400 600 800 1000 1200
PVsyst SA Irradiance [W/m²]
O
Page 24
C

Comparison with the Sandia Model

ST

Sandia model :
- established Outdoor (Albuquerque, NM)
(some few days, tracking, sunny climate)
SY

- defines evolution of 5 points only

- PVsyst extends I/V curve by the one-diode
model on each set of points
PV

PVsyst SA

PVsyst model adjustment:

Rserie = 0.531 W
g = 1.16

Sandia-One diode Model comparison

over more than 100 modules: PVsyst SA
 Average around g = 1.1 to 1.15

Page 25

76 PVsyst SA
 SOLAR GENERATOR

Low-light efficiency indoor measurements

Manuf. #1, 5 modules Poly 240Wp Manuf #2, 220, 230, 240 Wp ploy Manuf. #3, 1 mono and 1 poly
2.0% 2.0% 2.0%

1.0% 1.0% 1.0%

0.0% 0.0% 0.0%

-1.0% Mesure Mod. 1 -1.0% -1.0%

Mesure Mod. 2
Mesure Mod. 3 Meas. 220_0 Meas. Mono
-2.0% -2.0% Meas. 220_1 -2.0%
Mesure Mod. 4 Meas. Poly
Mesure Mod. 5 Meas. 230
-3.0% -3.0% Meas. 240 -3.0% Model Mono
Measured average
Model average Model average Model Poly
-4.0% -4.0% -4.0%
0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200
Irradiance Irradiance Irradiance

Manuf #4, mono and poly Manuf #5, Mono / Poly Manuf #6, CIS
2.0% 2.0% 2.0%

SE
1.0% 1.0% 1.0%

0.0% 0.0% 0.0%

-1.0% -1.0% -1.0%

Meas. 200W Mono
Meas. Mono 230 Wp Meas. SF 135
-2.0% Meas. 250 W Mono -2.0% -2.0%
Meas. Poly 220 Wp Meas. SF 150
Meas. 255 W Mono
-3.0% Meas. 250 W Poly -3.0% Meas. Poly 230 Wp (1) -3.0% Meas. SF 165
Meas. 255 W Poly Meas. Poly 230 Wp (2) Model - Average

R
-4.0% -4.0%
-4.0%
0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200
0 200 400 600 800 1000 1200
Irradiance Irradiance Irradiance

Indoor measurements by independent institutes: 600-800 W/m²: values 0.5 to 1%

U
- no evidence of diff. between Mono and Poly 400 W/m²: values -1 to 0%
- weak indication as function of power O 200 W/m²: -3 to -4%

Page 26
C

Rserie effect on Low-light efficiency

ST

Evolution of the PVsyst Rs default: 2%

Efficiency f(Irrad) for different Rseries

1%
SY

PVsyst V6.26: New PVsyst default: 0%

-3% loss at 200 W/m2 -1%
Relative efficiency

PVsyst V6: default acc. to Sandia -2%

-3%
average: Gamma = 1.1 -4%
Sandia: outdoor measured values -5%
PV

(with spectral correction) -6%

-7%
Rs=0.70 ohm, Gamma=1.00, Indoor
Rs=0.60 ohm, Gamma=1.10, V6
PVsyst V5: Old default: Gamma = 1.3 -8%
Rs=0.53 ohm, Gamma=1.16, Sandia
strongly underestimated -9%
Rs=0.39 ohm, Gamma=1.30, V5
-10%
0 200 400 600 800 1000 1200
Serie resistance f(Gamma) Irradiance [W/m²]
PVsyst SA
0.75
0.7 NB: relative efficiency at 600 or 800 W/m2 lies usually
0.65
between 0 and 1% for modern modules.
Rserie [ohm]

0.6
0.55 Higher values are suspicious (boosted Rserie).
0.5
0.45
0.4
0.35 Linear dependency between
0.95 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35
Gam m a Rserie and Gamma

Page 27

PVsyst SA 77
 SOLAR GENERATOR

Effect on the simulation results

Yield f (RSeries) Effect on yield:
1.5%

1.0% + 0.5% at 600 W/m² - comparable in magnitude to

0.5% the low-light variations around 600 W/m²
0.0%
Sandia Model Gamma = 1.10 - doesn't depend much on the climate
-0.5%
Spectral gain
-1.0%
Berlin
- PVsyst doesn't take spectral effects into
-1.5%

-2.0%
Geneva account for crystalline modules
Sevilla
-2.5% Gamma = 1.30 Dakar
- Sandia model: spectral correction
-3.0%
0.3 0.4 0.5 0.6 0.7 0.8
acc. to Air Mass  apply to beam only ?
PVsyst SA
Rseries

SE
 When comparing module performance in PVsyst:
Make sure that the methodology for establishing the parameters is identical !!!

R
The PVsyst database contains modules:
- with undefined Rserie and Rshunt values (default depends on the PVsyst version)
- with Manufacturer specified values: we use to check the manufacturer's data using

U
low-light meaurements reports from independent institutes.
NB: The performance also depends on the temperature coefficient mPmpp and IAM
O
Page 28
C

Incidence Angle Correction (IAM)

ST

For all technologies.

The incident irradiance is partially reflected by the cover ,
IAM is the variation according to the incidence angle (Fresnel's laws).
SY

Default model in PVsyst: "ASHRAE"

parametrization with bo = 0.05:
PV

FIAM = 1 - bo · (1/cos i - 1)
where i = incidence angle

The default model is higher than the

Fresnel's laws between i = 65 and 85°.
The Sandia model (outdoor measurements)
PVsyst SA
is very close to Fresnel.

NB: IAM measurements are extremely difficult (methodology in IEC 61853-2).

Results got from specialized laboratories are often contradictory.
Page 29

78 PVsyst SA
 SOLAR GENERATOR

Module (cell) temperature

PVsyst simulation: Irradiance
Reflected

Produced
energy
according to a thermal balance :
Absorbed energy =
Evacuated energy (Pout,
convection, IR radiation)
PVsyst SA

SE
Gincid · a · (1 – h out) = U · (Tmod – Tamb)
a = Light absorption factor = (1 – Reflexion) (about 0.9)
U = Uc + Uv · WindVel = Thermal loss factor [W/m²·K]

R
Usually WindVel not reliable: assumed constant:

U
U = Uc = 29 W/m²·K Typical for "nude" collector (sheds)
= 15 W/m²·K Back insulated
= 18 – 25 W/m²·K intermediary cases (air duct)
Page 30
O
C

Module Temperature - NOCT

ST

Parameter specified by the manufacturer

NOCT: Normalised Operating Cell Temperature [°C]
SY

i.e. under the conditions:

Irrad = 800 W/m², Tamb=20°C, Wind velocity = 1 m/s,
Valid for a "nude" and not operating module !
PV

The specified NOCT is not always significant …

Remark:
Applying the thermal balance:
the NOCT value corresponds to our Uc:
NOCT = 45 °C  Uc = 29 W/m²·K
NOCT = 48 °C  Uc = 25 W/m²·K
… but this has not much meaning !

Don't use the NOCT approach !

Page 31

PVsyst SA 79
 SOLAR GENERATOR

One cell shading – "Hot Spot"

 The current in one cell string is limited by the weakest cell
 Imposing a high current: the shaded cell becomes "receiver":
all other cells will dissipate their energy in this cell
 May lead to heat up and destruction

SE
R
High voltage,
For an imposed
High current,
current:
=> high power
addition of voltages

U
on shaded cell !
PVsyst SA
O
Page 32
C

Protection diode
ST

Mounting a "by-pass" diode (anti-parallel)

The diode should be able to bear the module ISC current
SY
PV

PVsyst SA

PVsyst SA

36-cells module: one diode is not enough !

Page 33

80 PVsyst SA
 SOLAR GENERATOR

2 protection diodes
The module is divided into 2 half-modules, each one protected by a by-pass diode
Moreover, the diodes allow recovering a part of the shading mismatch loss

SE
PVsyst SA

R
PVsyst SA

U
For ensuring the module security, one diode per 22-24 cells is necessary
Requires a connection to the middle point of the cell string O
Page 34
C

Effect of one shadowed cell in an array

ST

PVsyst SA
One cell 80% shaded in an array of 3 strings of 6
modules each (i.e. 648 cells)
SY

Without diode:
Loss -17.1%
Shaded cell destroyed
PV

PVsyst SA

With 1 diode/mod 36 cells

Loss -10.7%
Danger for the shaded cell
Risk of secondary MPP

With 2 diodes:
Loss -4.7%
Cell protected

Page 35

PVsyst SA 81
 SOLAR GENERATOR

Typical modules
Traditionally
• 36 cells / module for 12V battery charging (possibly 40 for hot climates)
• Powers 40-75 Wc - Voltages Vmpp 14.3 to 15 V under 60°C (-80 mV/°C)
• Junction boxes with protection diodes
Modern for big systems and integration
• Typical modules 120 to 430 Wc (36 to 128 cells)
• Connections with sealed junctions boxes and cables + connectors

SE
• With or without frame (laminates)
• Glass-glass modules for integration, with cells spacing
Thin film modules
• 40 to 130 cells, high voltages and small currents

R
• Semi-transparent modules or coloured
• Flat-roof covering membranes of wide size (1500 Wp, 33 m²)

U
O
Page 36
C

Connections
ST

Traditional junction box

Contains by-pass protection diodes
SY

Watertight cable outputs

For big systems:

Pre-mounted cables, special watertight
PV

safety connectors (Multi-Contact, Tyco, etc)

PVsyst SA

On the roof: use of special cables, UV-resistant,

Be careful with the disposition, "dangling drop" avoiding penetration of water
Avoid big area conductor loops (electro-magnetic fields of lightnings).

Page 37

82 PVsyst SA
 SOLAR GENERATOR

Construction
The cover – mechanical support – is most often in tempered glass
 High transparency
 High refraction index => favours the Anti-Reflective coating action
 Dilatation coefficient near to the one of the cells.

The backside protection is often in Tedlar, Polyester, or other polymer

 Sometimes second glass => connections at the edges, or holes in the glass
 Encapsulant material EVA (EthylVinylAcetate)

SE
Frame
 Most often in aluminium, rarely plastic
 May be equipped with "tile" mechanic recovering systems

R
 Without frame ("laminate"): fragile during mounting !
For the integration

U
Mounting with very low tilts (avoids accumulations of dirt and mosses)

Page 38
O
C

Energetic pay-back time

ST

Depends on :
• technology (here: mono, wafer 300mm, effic. 14%, pessimistic)
• evolves quickly (Silicium SOG, wafer now 200mm, efficiency, etc)
SY

• geographic location (med-Eur, 1200 kWh/m², south-Eur 1700 kWh/m²)

• mounting and use (orientation, tracking)
PVsyst SA
Temps de retour
Energetic énergétique
pay-back time
PV

4.0
Système
3.5 Cadre alu
Module PV
3.0

2.5
Années
Year

2.0

1.5

1.0

0.5

0.0
SI- SI- SI- SI- Ruban, Ruban, CdTe, CdTe, SI- SI-
Mono, Mono, Poly, Poly, M-Eur. S-Eur. M-Eur S-Eur Poly, Poly,
M-Eur. S-Eur. M-Eur. S-Eur. 2010 2010

ECN - Energy Research Center Netherlands – Alsema, Fthenakis, 2004

Page 39

PVsyst SA 83
 SOLAR GENERATOR

Heavy metals emissions

Cd atmospheric emissions,
(Fthenakis, Brookhaven National Laboratory
Hyung Chul Kim, Columbia University)

SE
Atmospheric emission of heavy

R
metals, PV industry

U
O
Page 40
C

Inverter: Functionnalities
ST

Function: Convert the DC energy of the PV array

SY

into AC energy compatible with the grid

Input: Draw maximum possible energy from the PV array

(MPPT - Maximum Power Point Tracking )
PV

Automatic connect / disconnect acc. to availability

Safety: maximum power, overvoltages, lightnings

Output: Energy transfer to the grid

Synchronization
Sinus shape without distortions (harmonics)
No significant phase shift (or specified cosf)
Limit the electromagnetic HF perturbations
Safety: switch-off when grid absent, under- and overvoltages

Page 41

84 PVsyst SA
 SOLAR GENERATOR

Inverters parameters: AC output

AC output:
Pnom: Nominal power : output AC power
Vnom: Output voltage (nominal), grid voltage (range)

May also be specified:

Pmax: Maximum power
TPmax: Max. temperature for allowing Pmax
(Usually ambient temperature is considered)

SE
PLim1, TLim1
PLim2, TLim2 Power derates as f (Temperature)

Inom: Nominal output current

Imax: Maximum output current

R
(not used in Pvsyst)

U
O
Page 42
C

Inverters parameters: DC input

ST

DC input voltages:
VmppMin,
SY

VmppMax: Voltages range for MPP tracking.

VMaxAbs : Absolute maximum voltage, should never be exceeded
Sometimes specified:
VNom: Operating nominal voltage (rarely specified)
PV

VMinPmax: Minimum voltage for obtaining Pnom

(Corresponds to an input current limitation: ImaxDC = PNomDC / VMinPMax )

Operating: efficiency, as function of the power and the voltage

In PVsyst, may be specified as:
 three efficiency profiles as function of Power, for 3 input voltages
 one only efficiency profile, no voltage dependency
 two values hmax and hEuro (or eventually hCEC for USA)
 PVsyst will establish an "automatic" efficiency profile f (Power)

Page 43

PVsyst SA 85
 SOLAR GENERATOR

Inverter: maximum power tracking

Principle: matching the input impedance of the inverter
i.e. draw exactly the current corresponding to the MPP of the PV array

Voltage adjustment: the capacity is I array I Pulsed Switch

charged until the required voltage, then

the transistor draws the current from PV array
Inverter
circuits
the array + capacity Control: pulses

SE
I array

P=U*I
MPP tracking:

R
The operating point (voltage) is
permanently adjusted by successive steps,
for getting the maximum possible power

U
U array

O
Page 44
C

Inverter: sinus generation

ST

PWM technology
(Pulse Width Modulation):
SY

Create pulses of different

widths for getting just the
required power at a given time
in the sinus signal.
PV

Commutating devices:
Very old
IGBT or MOSFET transistors
technology
with thyristors (formerly thyristors)

Low frequency transformer

Pulse frequency or
20 – 80 kHz High Frequency transformer
or
Without transformer

Page 45

86 PVsyst SA
 SOLAR GENERATOR

Efficiency
6000
Pout vs Pin Transfer function:
POut lim = 5 kW AC
3 loss components:
5000
Without ohmic loss • Intrinsic efficiency (basic slope, orange)
With ohmic loss
4000
• Self-consumption (operating threshold)
PAC out [W]

3000 • Ohmic losses (quadratic: R * I²)

Pthresh = 100 W
2000

1000 Efficiency vs Pin

100%

SE
0
0 1000 2000 3000 4000 5000 600080%

Efficiency = PAC / PDC

PVsyst SA PDC Input [W] Without ohmic loss
With ohmic loss
60%

40%

R
=> simple variable change :
20%

PAC => Effic. h = PAC / PDC

U
0%
0 1000 2000 3000 4000 5000 6000
PVsyst SA O PDC input [W]

Page 46
C

"European" efficiency
ST

Standardized parameter (almost always given by manufacturers)

which characterizes the average efficiency during one year of operation
SY

Normalized efficiency (JRC/Ispra) = weighted average according to the usual power

distribution in middle Europe climates :
h (Euro) =
PV

0.03 · h(5%) + 0.06 · h(10%) +

0.13 · h(20%) + 0.1 · h(30%) +
0.48 · h(50%) + 0.2 · h(100%)

NB: in PVsyst, hmax and heuro

allow the construction of the complete
efficiency profile h (P)
PVsyst SA

Page 47

PVsyst SA 87
 SOLAR GENERATOR

Complete PV System
DC Part : Classe II insulation
Eventual lightning-rod

Junction box

Fuses > 1.5 * Isc

Fuses

DC switch

SPD
SPD

No connexion to ground (France)

Equipotential Section >= 10 mm² Cu

SE
Overcurrent Inverters
protections DDR 30 mA
AGCP DC
Public Grid Domain

Main switch AC

Compteur

DC switch
DC

R
AC
Compteur
S S Recommanded S S S S
P P when > 30 m P P P P
D D D D D D

U
Grid
PVsyst SA O
Page 48
C

Galvanic separation
ST

The needs for a galvanic separation may be discussed.

The advantages are essentially related to safety:
SY

 One polarity of the array may be connected to the ground

 Allows to choose the array voltage independently of the grid voltage
 In case of inverter failure, no risk of grid AC voltages on the array

PV

No risk of introducing DC voltages onto the grid circuit

 Low DC continuous currents may produce galvanic corrosions on contacts
But:
 The tranformer is a cumbersome and expensive element
 It penalizes the inverter efficiency (by about 1 to 1.5%)
The galvanic separation is mandatory:
• With amorphous modules (parasitic voltages damage the TiO2 layer)
• With old back-contact modules (Sunpower), which had to be polarized (+ to the ground)

Page 49

88 PVsyst SA
 SOLAR GENERATOR

Array – Inverter connection

 All strings connected on an inverter (same MPPT input)
should be identical (number and model of modules, orientation)

 Never connect several MPPT inputs in parallel

=> MPP tracking conflicts
Exception: inverter in Master/Slave mode
One inverter at low power, 2 inverters above half of the array nominal power

SE
The Master tracks the array MPP, and controls the slave operation
The arrays are connected in parallel on both inverters.

 Mandatory: bipolar switch at the input of the inverter (external)

R
U
O
Page 50
C

Inverter sizing
ST

Take into account:

 Specified Pnom = output AC power
SY

=> power at inverter input = Pnom / (1 – h max) (+ ≈5%)

 Real max. power of the array in operation depends on the orientation
 If the overpower behaviour of the inverter is "acc. to limitation"
 one can accept some hours over the nominal power without significant losses
PV

Page 51

PVsyst SA 89
 SOLAR GENERATOR

Array real operating power

30° south:
the nominal power is only reached
during a few hours per year

SE
PVsyst SA

In south facade:

R
the observed maximum
is ≈70% of Pnom

U
PVsyst SA
O
Page 52
C

PV array sizing : voltage

ST

Choose the type of module, acc. to :

• Market: provider, price, availability
SY

• Mechanical mounting
• Architectural aspects
Choose the number of modules in series
• Acc. to the inverter input voltage range
PV

V In
inverter • Vmpp (60°C) > VmppMin inverter
• Vmpp (20°C) < VmppMax inverter
• Voc (-10°C) < Vmax abs inverter
• Voc (-10°C) < Vmax isol. module
Optimisation:
• High voltage limits the wiring ohmic loss
• With some inverters:
PVsyst SA Vmin required for obtaining Pmax
• Inverter efficiency according to voltage
(new data for some inverters)

Page 53

90 PVsyst SA
 SOLAR GENERATOR

PV array sizing : current

Number of strings in parallel
• According to the desired power
• According to the available area
• If "string" inverter: acc. to available inputs
• Contractual conditions (physical meaning ?)
– Check Isc (at 1000W/m²) < Imax inverter if specified
– Check Pnom (STC) < PMaxDC (STC) if specified

SE
• Involves the PNom(Array) / PNom(Inv) ratio
• NB: The PVsyst simulation offers a good means for the adjustment of the PNom(Array) /
PNom(Inv) ratio : diminish the inverter power until obtaining an acceptable over-power
loss.

R
Final optimisations by trial-and-error
• Number of installable modules, according to module layout

U
• Desired power
O
Page 54
C

Sizing criteria
ST

PVsyst: Dynamic visualisation

of all sizing conditions
SY

NB: the following criteria may

be contractual (i.e. act on the
warranty) but don't have a
physical justification:
PV

Inv PMaxDC: If the array MPP

power exceeds the admissible
inverter PMaxDC, the inverter
displaces the operating point
on the I/V curve. Without
danger.
Inv IMaxDC: The max.
attainable current is not Isc,
it depends on the array
orientation.

Page 55

PVsyst SA 91
 SOLAR GENERATOR

Definition of the PV system

Minimal steps:
1. Choose Pnom required
or available area
2. Choose a PV module
3. Choose an inverter

SE
 PVsyst calculates your system
… and you can execute your first
simulation for this project.

R
U
O
Page 56
C

Sub-arrays, Multi-MPPT inverters

ST

- A sub-array is an homogeneous set of strings

(same module type, same number of modules in series)
SY

- A sub-array may concern several inverters or MPPT (identical) inputs

(ideally: same number of strings on each input)
- Multi-MPPT inverter: by default each input is considered as half an inverter
( identical Pnom  possible unbalance of overpower losses )
- But you can now manually allocate a nominal power sharing between inputs
PV

- Unbalanced Multi-MPPT : very asymetric inputs (ex. SMA Tripower series)

(  define one "main" input for several strings, and one "secondary" input for the rest of modules,
in one string - Allows a very flexible total number of modules )

Some possible sizing problems ( orange: warning, acceptable, red: impossible simulation):
- "Inverter strongly undersized " (overload conditions) :
in the Project's parameter increase the parameter "Limit Overload Loss for Sizing" (default 3%)
- "Inverter strongly oversized ":
in the Hidden parameters, topic "System design parameters":
diminish the parameter "The inverter power is strongly oversized" (default Parray/Pnom = 0.6)

Page 57

92 PVsyst SA
 SOLAR GENERATOR

Loss Factors definitions

Parameters defined in the "System" part, button "Detailed losses"

 Collector Temperature:
heat loss factor Uc and possibly Uv (according to wind velocity), see Help by F1
to be defined according to the installation type (integrated or not)
 Ohmic wiring loss: Basic parameter : resistance Rwiring
DC side: all cable resistances in parallel, on all inverter inputs of the sub-array
AC side: define the distance inverter-injection point, calculation takes mono/tri into account
Initial Default: Wiring loss may be specified as % loss at STC,
but the resulting wiring loss during operation will result in lower value (R * I² for each hour)

SE
Possibility of defining External transfo losses (iron and ohmic)
 Module quality loss
Represents the confidence you put yourself in the real performance of the modules
with respect to their specification (tolerance, degradation, yield warranty provision, etc).

R
Initial Default value: according to the specified tolerance range (1/4 of the range)
May be negative (= gain) for positive sorted modules: ex: Tol. 0 .. 3% will give MQ loss = -0.75%
 LID loss: Light Induced Degradation

U
Crystalline modules: degradation during the first hours of exposition to the sun
This degradation is with respect to the initial factory flash-test eventually given by manufacturer
O
Page 58
C

Loss Factors definitions (continued)

ST

 Mismatch loss – Differences between real modules, especially in a string

Fixed parameter in the simulation, there is a tool for its evaluation and understanding
 Long term degradation – Not explicitely defined yet in PVsyst
SY

The PVsyst simulation is supposed for the first year

You can use the "Module Quality Loss" for this

 Soiling loss - Very dependent on the site, Specified in yearly or monthly values
PV

 Incidence angle loss (IAM)

May be customized for special textured glasses
May be specified in the PV module data (PAN file)
 Auxiliaries and night consumption
Fans, monitoring and other auxiliary consumptions,
May be Specified as proportional to the produced power, and with a power threshold
If defined in the inverter's parameters, these are used as default value
 Unavailability (system failures or maintenance)
May be defined as several periods of system "off"

Page 59

PVsyst SA 93
 SOLAR GENERATOR

Power Factor
The energy calculated by the simulation at the inverter output is always the Active energy.

A power factor (phase shift, cos f) may be required by the grid manager:
It should be specified as operating parameter within the inverter.
The Apparent energy is the product U * I expressed in [kVAh]
If the voltage is sinusoidal, the active energy is U * I * cos f [kWh]
 the apparent energy will always be greater than the usual E_Grid value !

NB: A power limitation (at the inverter output or grid injection)

SE
corresponds usually to a current limitation:
 for many inverters, the PNom is expressed as Apparent power (kVA).
In this case, the power limitation will act for PNom (active) lower than Pnom (app).

R
Grid Power limitation
 May be required by the grid manager, either as Active or as Apparent power
 The limitation may be applied either at the injection point, or at the inverter output

U
O
Page 60
C

Simulation with PVsyst – first evaluation

ST

First evaluation, define the main parameters:

 Project:
SY

• Define the geographical site

• Choose the meteo file (if not existing: automatic synthetic generation)
• Optionnaly define special Albedo conditions (snow), sizing temperatures
• Give a name to the project, and save it
PV

 First calculation variant:

• Define the orientation of the field
• Define the system :
Required nominal power or available area
Choose a PV module,
Choose an inverter model,
=> PVsyst establishes the configuration (number of modules and inverters)
• If no red "Warning": perform the simulation
• Check the results ("Report")
• Save this variant (specify a comment)
Page 61

94 PVsyst SA
 SOLAR GENERATOR

Simulation – Further studies

During the first evaluation:
All the loss parameters are set to reasonable default values.
Further: analyze the effects of each loss or feature (incrementally)
 Far shadings (horizon)
 Near shadings + electrical effects (partition in module strings, module layout)
 Adjust the array losses (system definition, button "detailed losses"):
 Adjust operating conditions (power factor, grid power limitation, etc )

SE
 If needed, define custom special graphs, hourly values, monthly tables, etc
 Possibility of creating a CSV file with hourly results  msEXCEL
 Possibility of creating an economic evaluation and Carbon balance

R
 Batch mode for parametric studies

U
O
Page 62
C

Simulation - Results
ST

Simulation Report
• Lists all parameters used for this simulation
SY

• Shading diagrams, horizon

• Main results (Irradiance, produced energy, PR, efficiencies)
• Detailed arrow-diagram describing yields and all losses in the system
• P50 – P90 probabilistic yield evaluations
PV

• Economic evaluation, Carbon balance

The simulation report generates a printed (or PDF) document
which may be used as detailed offer to the end customer.
Further results
Selected results may be exported by copy/paste to EXCEL (runs comparisons)
All variables may be visualized as tables of monthly values
Graphs in hourly values
Special graphs defined prior to the simulation
Output CSV File in hourly or daily values for export to EXCEL

Page 63

PVsyst SA 95
 SOLAR GENERATOR

Economic evaluation
Definition of the costs (investment)
• Specify prices as global, per component, per Wp, per m2
• Prices may be defined for components in the database
• Possibility of defining your own labels
• In any currency (parity definition tool)

Computes the investment, taxes, subsidies, etc

• Supposes a loan for the whole investment, calculates constant annuities

SE
• + maintenance cost => yearly cost
 Determines the real cost of the produced kWh (LCOE)

R
U
O
Page 64
C

Operating results
ST

The production verification of a PV plant may be performed, in monthly values:

• From the reading of the selling energy counter
• With renormalization of the real irradiation
SY

(to be obtained from a near meteo station or satellite data)

Some inverters are equipped with Dataloggers
• Array energy, produced energy, max. power, etc
• If possible, connect an irradiance sensor (check calibration!)
PV

For a detailed system analysis (JRC/ISPRA recommendations):

• Monitoring of hourly values
• Measurement of the irradiance in the collector plane (POA)
… or better: in the horizontal plane !!!
• Ambient temperature (sensor under shelter)
• Voltage and current of the array at inverter input
(+ power U x I should be integrated at each time step !)
• Inverter output power / energy
• Operating duration, unavailabilty recording

Page 65

96 PVsyst SA