PV Systems Optimization Through Improvements in Laboratory and Field
Characterization, Modeling, and Analysis
Charles HanleyDOE/Solar Energy Technologies Program
Annual Review MeetingDenver, ColoradoApril 17-19, 2007
Relevance/Objective
• FY06 AOP: “develop, validate and apply new testing procedures based on energy production to optimize the performance of multiple components in fully integrated PV systems, providing the basis for system performance modeling and testing standards.
• MYPP milestone links:– Determine PV configurations for government procurements and hardware
certification programs (2008)– Extend Solar Advisor Model (SAM) capabilities as a predictive energy
model (2009)– Demonstrate factory-integrated residential and commercial PV systems
(2011)• Links to other program areas:
– Inverter and balance-of-system development– Module and array manufacturing and reliability studies– Systems analysis, including baselines and tracking of TPPs in SAI– SAI testing, evaluation, and technical support
Overview of PV Systems Optimization Laboratory (PVSOL)
Controlled side-by-side array and system characterization
• Energy production comparison versus module design
• Inverter in-situ performance testing• Soiling effects for different glasses and coatings
Comprehensive Data Acquisition Systems• All ac and dc data• Meteorological data• Continuous archive• Protocols for others
Part of the PVSOL Module Setup
Inverters, combiners, disconnects – all reconfigurable
Budget History
SNL Project Areas: FY06 ($k) FY07 ($k)Inverter/BOS Development, Manufacturing R&D, and Testing 380 1000
Module, Array, and System Manufacturing, Reliability, Test and Evaluation 300 1500System Performance Optimization, Modeling, and Benchmarking 835 850
This materials in this presentation are covered in the line highlighted in yellow above.
Summary of Activities/ Accomplishments
1. Draft system test protocol completed for use in SAI evaluations
2. Empirical inverter model completed3. Complete system performance model updated for
inclusion in SAM4. Further characterization of new products in partnership
with U.S. industry leaders.
Table 1. Some of SNL’s PV Systems Partners Advent Solar BP Solar CSG Solar Evergreen Solar PV Powered
Powerlight Sanyo Sharp SMUD Solectria
Sun Edison SunPower SunWize Tucson Electric Xantrex
1. System Test Protocol
• Developed in partnership with USDA/Rural Utility Service
• Documented and structured approach for components and systems
• Support SAI objectives and activities– Establishing technology baselines
– Tracking program progress
– Stage gate evaluations
• Current Status: undergoing review by RUS; under revision by inter-lab Test & Evaluation team.
2. Inverter Model Parameters
0
500
1000
1500
2000
2500
3000
0 500 1000 1500 2000 2500 3000dc Power (W) or Array Pmp (W)
ac P
ower
(W)
Nominal dc Voltage, VdcoHigh dc VoltageLow dc Voltage
Paco
P dco
Pso
CurvatureCo
ac Power Limit
Pnt
• Model determines AC power based on DC power and DC voltage inputs• Employs up to 8 parameters, depending on information available
• 5 parameters shown in figure above, others are empirical coefficients • Defaults come from manufacturers’ specification sheets• Current CEC inverter tests/database provide all parameters for model
Inverter model maps output across full range of inputs
• Ready for inclusion in Solar Advisor Model (SAM)
• Database contains parameters for 47 inverter/voltage combinations, based on SNL tests, CEC inverter database, and manufacturers’spec sheets.
, , p
0
500
1000
1500
2000
2500
3000
0 500 1000 1500 2000 2500 3000
dc Power (W)
ac P
ower
(W)
40
50
60
70
80
90
100
Inve
rter
Effi
cien
cy (%
)
Meas at 252 VdcMeas at 302 VdcMeas at 483 VdcModel at 252 VdcModel at 302 VdcModel at 483 Vdc
Comparison of Sandia inverter performance model (lines) and CEC measurements (dots).
Inverter model closely tracks real-world measurements
-6
-4
-2
0
2
4
6
0 500 1000 1500 2000 2500 3000
dc Power (W)
Erro
r in
Mod
eled
Effi
cien
cy (%
)
For Pdc > 250 W, Std. Error = 0.50%
Error in modeled vs. field measurements of a 2500W inverter over a 13-day period.
3. Integration of Overall System Performance Model
• Linkage of inverter model with previously-developed module/array model
• Covers a variety of climatic conditions, technology configurations, and installation conditions
• Predictive value: determine expected AC energy production for a particular system configuration
• Real-time value: monitor changes in output to determine system performance, identify faults.
IrradianceModel
SNLInverterModel
SNLArrayModel
AC outDC outPOAWeatherInstr.
TMY2or
System performance model closely tracks measured output
• 4-Day comparison of modeled and measured system performance at PVSOL
• Accurate match even in changing weather conditions
0
500
1000
1500
2000
2500
3000
3500
4000
1 721 1441 2161 2881Time (2 Min Samples)
AC
Pow
er (W
atts
)
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
Rat
io
Model DataMeasured DataModel/Measured
4. Characterizing New Products:e.g., Bifacial Modules
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0 10 20 30 40 50 60 70
Voltage (V)
Cur
rent
(A)
Pmp = 194.8 Watts Vmp = 49.47 Volts Imp = 3.938 Amps FF = 0.73 Eff = 16.09 % Voc = 61.64 Volts
Isc = 4.305 Amps
Pmp = 173.9 Watts Vmp = 49.52 Volts Imp = 3.512 Amps FF = 0.74 Eff = 14.32 % Voc = 61.37 Volts
Isc = 3.851 Amps
IV curves empirically obtained –dependent on installation conditions, such as ground reflectance.
Solar irradiance on front and back of bifacial modules for two clear and two cloudy days.
0
200
400
600
800
1000
1200
1400
0 500 1000 1500 2000 2500
Seq. No.Irr
adia
nce
(W/m
2 )
Front IrradianceBack Irradiance
Bifacial modules: modeling annual output
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec YearMonth
Dai
ly A
vg. d
c En
ergy
(kW
h/d)
No Bifacial EffectWith Bifacial Effect
Annual Energy Boost = 18%
Result: empirical model predicts 18% boost in annual energy!
• Partnered characterization effort with U.S. manufacturer
• Statistical analysis of time-series data shows marked differences (3-4%) in energy production
• System model parameters developed and utilized for overall comparisons
• Visual inspections showed different rates on soiling of panels
4. Characterizing New Products:e.g., Anti-Reflective Coatings
Anti-Reflective Glass Coatings on Equivalent Modules
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
7/12/06 8/11/06 9/10/06 10/10/06 11/9/06 12/9/06 1/8/0Date
Nor
mal
ized
Dai
ly A
rray
Eff.
(%)
Type AType BType C
Major Milestones
FY06 Program Milestones:• Demonstrate and document a new benchmark for accuracy in characterizing,
modeling, and analyzing energy-based performance of PV systems and components (inverters) in SNL systems optimization lab (PVSOL).
– Result: Models integrated to calculate overall system efficiency; letter report sent to DOE describing proposed system-level metric.
• Through comprehensive interactions with component manufacturers and system integrators, demonstrate and promote specific technology pathways in PV system integration/manufacturing/deployment that will achieve DOE and industry goals for levelized energy cost.
– Result: Reports to industry partners covering specifics of all technologies and pathways investigated, including those examples described herein.
FY07 Program Milestones:• Documentation of field-testing protocol for array, inverter, system performance
for SAI/TPP participants.• Document performance metrics for multiple array/inverter combinations in
PVSOL providing validation of testing and modeling protocols.• Initiate new method for accelerated array field-aging for SAI/TPP participants.
– Result: All milestones on track for FY07; several links to other milestones in test & evaluation, inverters &BOS, systems analysis, and SAI market transformation.
Future Directions
Specific follow-on activities include:• Refining models to address issues such as tracking and
shading, reliability, derates, different weather models, and other lifetime-related variables.
• Establish, validate, and implement accelerated testing protocols for PV components and systems (corrosion, packaging, interconnection, system reliability, etc.)
• Execution of test & evaluation activities in support of SAI, including lab and field measurements to refine analyses, establishing SAI baselines, and tracking program progress.
• Continuation of partnered investigations with industry on new materials, components, and systems integration methods.