Use and Validation of the Solar Advisor Model for Photovoltaic Applications

1
Use and Validation of theSolar Advisor Modelfor
Photovoltaic Applications

• Chris Cameron
• Solar Systems Department
• Sandia National Laboratories
• March 13, 2008

Sandia is a multiprogram laboratory operated by
Sandia Corporation, a Lockheed Martin
Company,for the United States Department of
Energys National Nuclear Security
Administration under contract DE-AC04-94AL85000.
2
Levelized Cost of Energy

• Why Levelized Cost of Energy?
• To measure cost-competiveness of solar energy
• Cost-competitiveness key to very wide-scale use
• Absent policy drivers
• Looking beyond incentives, net metering, and
  green markets
• Metric that captures all economic costs
• Unlike manufacturing cost (/Wp), module
  efficiency, etc.
• Note other metrics, e.g. first cost, ROI, etc.
  may be more important in the market place
• SAM developed to enable use of LCOE metric
• Also supports Systems-Driven Approach to
  technology development

Chris Cameron and Craig Cornelius, A
Systems-Driven Approach to Solar Energy RD,
http//ieeexplore.ieee.org/xpls/abs_all.jsp?arnumb
er4304320
3
Systems-Driven Approach

• Market analysis to establish targets
• Reference system designs for target markets
• Technology Improvement Opportunities (TIOs)
• Reduce cost, increase performance and reliability
  
• Perform Parametric Analysis to downselect TIOs
• Pursue multiple pathways to reduce risk
• Measure Progress
• Apply stage-gate management to evaluate progress

4
Market Analysis

• Targets defined relative to market prices
• Market prices reflect analysis of Energy
  Information Administration data
• Residential/Commercial market prices based on
  utility rates
• Utility market price based on Levelized Cost of
  Energy (LCOE) from new combined-cycle gas turbine
  in SW US
• Nondispatchable power has a market price of 4/kWh

Solar Program 2015 Targets (/kWh)
5

Example Analysis

• Impact of Inverter on First- and Life-Cycle Cost
• Residential System
• Note the impact of Operation and Maintenance Cost
  (OM)
• Suppose a 2 kW System at 5/W installed
• Annual output 4,000 kWh
• _at_ 10 /kWh, system output is worth 400/yr
• No room for 100 service calls
• PV modules may be reliable
• What about the inverter?

6
Inverter Contribution to LCOEAnalysis and
Assumptions
4 kW
6.9 /kWh2.6 /kWh
500/yr
4.7 kW
1.9 /kWh0.9 /kWh
2.4/kWh
0.3 /kWh0.4 /kWh
35/yr
5.9 kW
Component Costs _at_
1.8/kWh
5.51/Wp
0.8/kWh
0.90
5
Inv. Lifetime (yrs) Sys OM( of installed
cost)
0.69 100.3
0.30 200.2
50.5
Inv. Cost (/Wac)
Cost of Replacement/Repair (parts and labor) at
end-of-life assumed to be same as component cost
of original inverter in nominal dollars,e.g.
original inverter cost 2,000 in 2005, cost to
replace in 2015 is 2,000in 2015 s, including
parts and labor (cheaper in 2005 s)
2007-2011 MYPP
7
Inverter Efficiency

• Inverter efficiency (dc to ac) improvements have
  small effect on LCOE
• Not much room for improvement
• Array Utilization also Important
• Assumes (nearly) perfect
• Maximum power point tracking
• Wakeup and shutdown
• No power fold-back due to heat
• System design can interfere with max. power point
  tracking
• Shading
• Different string orientations

8
Primary system-level trade-off is first-cost vs.
reliability
Parametric Analysis
Contribution of 5 kW Inverter to LCOE
9
Parametric Analysis
Recall cost of repair/replacement including
labor is original cost.No other service cost is
in this analysis.
10
PV Performance Models in SAM
Performance Models

• Modules
• Single-point efficiency with single temperature
  coefficient module model
• Sandia PV Array Performance Model
• CEC/Wisc 5-parameter model
• Inverters
• Single-point efficiency inverter model
• Sandia Inverter Performance Model

11
Wont uncertainty in solar resource dominate
model errors?
Who Cares?

• Yes actual system output is dominated by
  weather
• But, improved accuracy enables technology
  choices A vs. B. 5 better performance is
  significant.
• Improved accuracy enables evaluation of impact of
  design changes, e.g. anti-reflective glass on
  annual performance
• Some module tests at Sandia are industry-funded
  to keep results of prototype testing proprietary
• Accurate models useful for performance monitoring
  to detecting problems

12
Module Model Comparison

• CEC 5-parameter model
• More readily available coefficients
• short circuit current
• open circuit voltage
• current and voltage at maximum power point
• temperature coefficient of the open circuit
  voltage
• temperature coefficient of the short circuit
  current

• Sandia Array Performance Model
• 40 coefficients measured in ½ day test on
  two-axis tracker
• Coefficients for similar modules can be estimated
• Model uses beam and diffuse data
• Coefficients include
• Isc, Voc, Imp, Vmp with 4 temp coeff.
• Polynomial representation of air mass
• Angle of incidence modifier
• Thermal model for cell and module operating
  temperatures vs wind speed and ambient
  temperature
• Demonstrated for moderate concentration
• Believed applicable to high concentration and
  multijunction cells
• Downside database is not current

13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
CEC Inverter Test Data
W
5-7 tests are done on a single inverter at each
power level and voltage. Coefficients are derived
from this or similar data
17
CEC Inverter Test Summary
18
Model vs. Weighted
19
System Derates

• System derate factors present major uncertainty
• Module model eliminates PV module nameplate
  rating
• Inverter model addresses inverter efficiency
• System degradation is entered as a rate
• Remainder of values are from Bill Marions
  PVWATTs site.

20
Near-term Efforts

• Validation of system performance against measured
  weather and system data, including for
  concentrators
• Navigant Consulting Inc developing
  SAM-interactive, spreadsheet-based, system cost
  model
• NCI also developing manufacturing cost model to
  support module cost inputs
• Support SAI partnership use of SAM