PowerPointPrsentation

1
IP PERFORMANCE and possible links of its goals to
IEA PVPS Task 2
European research initiatives towards new quality
standards for PV modules and arrays performance
characterization Tadeusz Zdanowicz Wroclaw
University of Technology SolarLab
2
IP Performance subprojects
SP1 Traceable performance measurement of PV
devices (TUEV, DE) SP2 Energy delivery of PV
devices (ZSW, DE) SP3 Performance Assessment and
Evaluation of Photovoltaic Systems (UNN, UK) SP4
Modelling and analysis (LU, CREST, UK) SP5
Service life assessment of PV modules (FhG-ISE,
UK) SP6 PV as a building product (ECN, NL) SP7
Industry interaction and dissemination (EPIA) SP8
Standardisation processes (JRC, EU)
3
Subproject 2 Energy delivery of PV devices
WP2.1 Assessment of actual outdoor evaluation
procedures CEA WP2.2 Influence of performance
relevant parameters WrUT WP2.3 Minimum set of
characterising module parameters CREST WP2.4
Translation between indoor and outdoor
performance measures JRC WP2.5 Preparation and
implementation of harmonised procedures WP2.6
Performance evaluation at system level ZSW
4
Definition of Extended measurement campaign

• Locations
• ZSW, Widderstal (Stuttgart, Germany)
• CIEMAT (Madrid, Spain)
• WrUT (Wroclaw, Poland)
• Measurement methods have been defined and agreed
• Experimental setups at partners sites have been
  modernized and adjusted to new requirements
• Number and type of PV modules (commercially
  available) to be tested discussed and agreed
  (sc-Si, mc-Si, CI(G)S, CdTe, a-Si (1J, 2J and
  3J)) 2-3 pieces of each type, 20 modules in
  total

5
Definition of Extended measurement campaign
(contd)

• What and how is to be measured - agreed
• What should be calculated - agreed
• Database (data format and data availability) -
  agreed
• Definition and identification of possible
  pitfalls and error sources under discussion and
  ongoing work

6
What is to be measured?

• PV module power rating
• I-V curve of a PV module
• PV module temperature
• Irradiance horizontal (global, diffuse), global
  (POA) GPOA measurement both with pyranometer as
  well as Si sensor
• Set of weather parameters (especially of those
  affecting solar spectrum and module temperature)
  ambient temperature, relative humidity,
  atmospheric pressure, wind (speed, at least )
• Solar spectrum (minimum range 0.25 1.1 mm,
  preferable range 0.2 1.7 mm)

7
What is to be calculated?

• Thermal coefficients, NOCT (is it really useful
  ???)
• Energy rating and/or module PR (DATA VALIDATION
  UNCERTAINTY ESTIMATION!)
• Additional parameters (calculated)
• a)      AOI angle of incidence
• b)      Eph equivalent or average photon energy
  
• c)      Solar spectrum calculated (additionally
  to measured)
• d)      AM air mass factor
• e)      Water vapor thickness

8
What affects mainly data reliability and accuracy?

• Problems with estimation of true module
  temperature mainly due to high thermal inertia of
  a module especially significant in case of
  frequent and abrupt irradiance changes (e.g. due
  to quickly running clouds)
• Time inertia of irradiance sensor (pyranometer)
• Time inertia of temperature sensor
• Nonuniform temperature distribution over module
  area e.g. due to elements of mounting
  construction, cell mismatching, partial
  shadowing, shunts, hot spots etc.
• Spectral effects (especially in case of thin-film
  a-Si devices) and spectral mismatch error
• Extended in time memory and recovery effects
  in case of some thin-film devices (e.g. CIS),
  degradation (a-Si) etc

9
What affects mainly data reliability and
accuracy? (contd)

• Fluctuations of solar irradiance, especially
  significant in case of slow I-V scan
• Errors due to applying curve translation
  procedures (where needed)
• Incorrect numerical procedures applied when
  determining module parameters e.g. applying
  linear regression where it is not suitable
• Capacitance and transient effects (rarely
  usually I-V scan is slow enough to neglect the
  problem, except dye-sensitized)
• Possible errors in measuring circuits and/or
  procedures (usually minimized, can be neglected)

10
Acquiring and sharing data with the project
partners (SolarLabs SAS based DAS)

• Directly measured parameters
• a)   PV Modules   I-V curve, VOC, GPOA, Tm,
  date/time (90 s interval if GPOAgt10 W/m2)
  determined from I-V ? ISC, IRAT, IMPP, UMPP,
  PMPP, date/time (90 s interval), integrals
• b)      GHi, DHi, relative humidity, atm.
  pressure, Tamb, Wind speed and direction, (60 s
  interval)
•  
• Additional parameters (calculated)
• a)      AOI angle of incidence
• b)      Eph equivalent photon energy
• c)      Solar spectrum calculated
• d)      AM air mass factor
• e)      Water vapor thickness (calculated using
  Gueymards model)

11
Fitting I-V curves to Equivalent Diode Models
Diode models I-V curves are fitted, stored in
database for further analysis and treated as
additional parameters. Due to computational
complexity data are fitted on distributed
computers in local computer network
12
Fitting I-V curves (contd)
Arrhenius plots of the dark current components
and diode ideality factor A calculated for CIS
and CdTe thin-film modules using either of two
eqivalent diode models I-V curves used for
fitting were taken for full range of available
irradiance values.
Comparison of the band gap energy Eg determined
for CIGS thin-film module using Arrhenius plots
for both equivalent diode models.
13
Fitting I-V curves (contd)
a) b) Dependence of module shunt RSH (a)
and series RS (b) resistances on module
temperature Tm determined for CIS and CdTe
thin-film modules using I-V curve fitting to
either of two equivalent diode models I-V curves
used for fitting were taken for whole range of
available irradiance values.
14
Experimental vs. modelling energy gap
15
Experimental vs. modelling - Pm
16
Experimental vs. modelling Pm (contd)
Distributions of error for different levels of
irradiance
17
Looking (hunting) for STC
Frequency of occurrences of values close to STC
during of more than three years of outdoor
monitoring in SolarLab (irradiance 1000 ? 2
W/m2 and Tm 25 ? 50C).
18
Translation to STC - ST42 CIS PV Module
ISC
VOC
PM
Errors corresponding to basic values of
parameters determined after translation of I-V
curves measured in one four values of irradiance
(700, 900, 1100 or 1200 W/m2, respectively) to
Standard Test Conditions using each of three
basic numerical procedures (IEC 60891, Blaesser,
Anderson)
19
Translation to STC 3J a-Si UniSolar PV Module
(contd)
ISC
VOC
PM
20
KZ CM21 response time to abrupt change of
irradiance
21
Thermal inertia of PV module vs irradiance changes

• Nonfiltered data

• Filtered data

22
Data filtering
1. Using measured ISC value of PV modules as a
reference for irradiation stability reduces
effects of pyranometer time response to light
fluctuations
2. Introdution of a new parameter daily
irradiance stability index cGS helps to remove
outliers from the database
For sunny clear (stable) day CGS ? 0
23
Automatic data cleaning removing outliers
24
Example of effect of data filtering (1) (VOC vs
Tm at GPOAconst)
a) no filtering b) filtering using only ISC c)
filtering using ISC cGSlt0.07
25
Example of effect of data filtering (2) (Pm vs Tm
at GPOAconst)
a) no filtering b) filtering using only ISC c)
filtering using ISC cGSlt0.07
26
Energy Rating of PV modules- computing the time
integrals
trapezoid
rectangles
Examples of integrals computed taking various
time intervals and integration method
27
Effect of integration algorithms and sampling
interval on Energy Rating calculations

• Rectang.

• Trapezoid

28
Subproject 3 Performance Assessment and
Evaluation of Photovoltaic Systems

• Understanding and reduction of loss mechanisms
  experienced in the field
• Strategies for maintaining optimum system output
  throughout the system lifetime
• Harmonisation of PV system monitoring guidelines
• Strategies for implementation of Guaranteed
  Results approaches, technologically underpinning
  innovative financing and amortisation concepts
• Examination of the role of energy service
  companies (typically SMEs) and their impact on
  improved PV performance and user confidence
• Integration with SP2, SP4 and SP6 to achieve the
  tools to attain long-term system performance
  ratios of 75-80 for typical European
  installations (cf. less than 70 on many systems
  today)

29
SP3 Workpackages
WP3.1 Modernised European guidelines for PV
system monitoring JRC WP3.2 Data requirements and
procedures WrUT WP3.3 Tools and protocols for
analytical monitoring UNN WP3.4 Strategies to
improve system inverter performance ISE
30
Tasks in WP3.2
Data requirements and procedures
Definition of preliminary procedures for fault
definition and diagnosis (taking into account the
user needs definition and field experience review
from WP.3.1
Assessment of data precision requirements as a
function of user needs
31
Definition of preliminary procedures for fault
definition and diagnosis

• Definition of what we may call faulty operation
  of a PV system
• Categorization of faults regarding their impact
  on the PV system performance
• user needs and expectations must be taken into
  account
• field experience review to be taken from WP3.1

32
Definition of preliminary procedures for fault
definition and diagnosis according to application
and size of PV system
33
Fault definition and diagnosis according to
components of PV system proposed partners
involvement indicated
34
Methods of fault diagnosis
35
Urgency of intervention (idea of F.Baumgartner)
Cat A urgent intervention not necessary Cat B
intervention required within days or even
weeks and Cat C ALERT immediate intervention
required
36
Assessment of data precision requirements as a
function of user needs

• Three (at least) major factors deciding what kind
  of monitoring equipment is to be installed in the
  PV system. These may be defined as
• - required (expected) system reliability -
  monitoring equipment oriented mainly to early and
  reliable fault recognition
• - system costs limitations (ratio of cost of
  monitoring equipment to ovearall system cost may
  be here a measure) this may be main factor in
  case of domestic systems
• - special accuracy requirements
  (analytical/scientific monitoring)

37
Assessment of data precision requirements as a
function of user needs (contd)
Data precision requirements should be defined
taking into account the user needs (fit for need
but nothing beyond need) . The user needs depend
on type of the system, i.e. its application. Here
simple differentiation is proposed (partners with
leading role have been indicated) Domestic
systems if the produced energy is not the
object of trade but is used exclusively for user
needs then the required accuracy should be on the
level allowing for fast and reliable diagnosis of
the system malfunctioning low cost is often a
desisive factor Professional systems high
reliability and accuracy (the latter especially
important in case of analytical monitoring) are
of crucial importance, cost usually is not as
important Grid Connected PV systems margin of
accuracy determined by the grid energy
distributor obliged to purchase PV energy
generated by the system the accuracy should be
sufficient enough to diagnose system functioning
as well as quality of a.c. energy fed to grid
38
Assessment of data precision requirements as a
function of user needs (contd)

• Definition of the set of parameters to be
  monitored (including meteorological data)
• -    parameters on d.c. side allowing to estimate
  correctness of the PV array functioning
• -    parameters allowing to estimate correctness
  of the controller/inverter functioning (e.g. MPP
  tracking effectiveness
• -    parameters allowing to estimate status of
  the energy storage elements and effectivness of
  charging/discharging process
• -   output a.c. parameters as output energy,
  energy quality (frequency stability, harmonics,
  ripples, etc.)
• -   meteorological data .
• (Recognize equipment currently available on the
  market)

39
Subproject 4 Modelling and analysis

• Development of a coherent set of models of PV
  module and system performance these models will
  translate PV module data and PV component data
  (time varying where appropriate to take account
  of degradation etc) into system performance
• Calculation of life-time energy
• Tools for PV system condition appraisal/monitoring
  
• The aim is to halve the inaccuracy of energy
  yield prediction of any proposed system design.

WP4.1 Interfacing and Data Assimilation
CREST WP4.2 Environmental Modelling WrUT WP4.3
I-V Characteristic Based Modelling CREST WP4.4
Annual Energy Production and Device Comparators
SUPSI WP4.5 Life Time Energy Rating ISE
40
Modern Web Technologies and Data Exchange System
for application in PV area
The main Web technologies

• XML (Extensible Markup Language)
• Style Sheet (XSL, XSLT, XSLFO)
• Web Services

41
XML (Extensible Markup Language)

• XML creates application-independent documents and
  data
• XML can be inspected by humans and processed by
  any application
• It has a standard syntax for meta data
• XML provides an effective approach to describe
  the structure and purpose of data
• It has a standard structure for both documents
  and data
• XML organizes data into a hierarchy
• allow applications to dynamically discover
  information about Web services

42
The Style Sheet
Style sheets allow to specify how an XML document
can be transformed into new documents, and how
that XML document could be presented in different
media formats
43
Examples of using combination XML Style Sheet
44
Web services
Web services are software applications that can
be discovered, described, and accessed based on
XML and standard Web protocols over intranets,
extranets, and the Internet.

• perform specific functions
• are based on XML (XML, standard supported and
  accepted by thousands of vendors worldwide)
• exchange information over intranets, extranets,
  and the Internet (and local net, too).

SOAP, developed as the Simple Object Access
Protocol, is the XML-based message protocol (or
API) for communicating with Web services
45
What kind of data format we need ?

• Data format should clearly and precisely to
  describe exchange of information between
• Partners of RD projects in purpose of
• - to exchange informations on details how
  measuremnts are being performed,
• - used sensors, their accuracy etc.
• - algorithms and procedures used to perform
  calculations
• User and Service specialized in maintanance of
  PV systems and monitoring facilities
• Reduce outage time and maintenance effort by
• automated yield monitoring of PV systems
• early identification of efficiency losses
• automated fault diagnostics
• notification of unsufficient energy production to
  the operator
• long term storage of operating data including
• permanent access
• Manufacturers and buisness and market community

46
SensorML? (next step forward)
SensorML is a key component for enabling
autonomous and intelligent control of web
connected sensors. SensorML provides the
information needed for discovery of sensors,
including the sensors capabilities, location,
and taskability. It also provides the means by
which realtime observations can be geolocated and
processed on-the-fly by SensorML-aware
software. SensorML describes the interface and
taskable parameters by which sensor tasking
services can be enabled, and allows information
about the sensor to accompany alerts that are
published by sensor systems. Finally, intelligent
sensors can utilize SensorML descriptions during
on-board processing to process and determine the
location of its observations1). 1) Open GIS
Sensor Model Language (SensorML) Implementation
Specification, OGC 05-086
47
Why SensorML ?
The Sensor Model Language defines an XML schema
for describing the geometric, dynamic, and
observational characteristics of sensor types and
instances. Sensors are devices for the
measurement of physical quantities. There are a
great variety of sensor types, from simple visual
thermometers to complex electron microscopes and
earth observing satellites. The Sensor Model
Language is a human-readable, XML-based language
that can be easily parsed by a wide variety of
existing tools. The current standard calls for
keywords in the English language, although
consideration for internationalization of
keywords should be considered if deemed
beneficial.
48
The purpose of the SensorML description

• provide general sensor information in support of
  data discovery
• support the processing and analysis of the
  sensor measurements
• support the geolocation of observed values
  (measured data)
• provide performance characteristics (e.g.
  accuracy, threshold, etc.)
• provide an explicit description of the process
  by which an observation was obtained (i.e. its
  lineage)
• provide an executable process chain for deriving
  new data products on demand (i.e. derivable
  observation)
• archive fundamental properties and assumptions
  regarding sensor

49
What does SensorML decribe?
50
Advantages of XML and Sensor ML IT technolgies ...

• Good idea
• Done a lot of work
• International range (337 companies)

51
... and disadvantages

• Join to OGC is not free of charge (
  http//www.opengeospatial.org/ogc/join )
• OGC standard is still under development ( loosed
  link to schemes)
• Lack of easy and useful tools for management to
  exchange of SensorML schemes

52
Proposal

• develop PV Information Framework
• use XML (Extensible Markup Language) for
  encoding PV information (PVML ?)
• use a JavaScript, HTML, XML for data presentation