performance modeling and testing of distributed … · 2015-09-17 · performance modeling and...

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Performance Modeling and Testing of Distributed Electronics in PV Systems

APEC 2015, Charlotte NC

Chris Deline

3/18/15

2

Roadmap

• Overview of Distributed MPPT (DMPPT) in PV

• Performance benefits of DMPPT

• Shade modeling in SAM

3

Roadmap




4

Examples of Distributed MPPT products

Frame - attached

Credit: Enphase Energy, Tigo Energy, Maxim Integrated

J-box embedded Laminate embedded

2.15 cm

5

Examples of Distributed MPPT products

Frame - attached

Credit: Enphase Energy, Tigo Energy, Maxim Integrated

J-box embedded Laminate embedded

2.15 cm

DMPPT = MLPE, microinverter, power optimizer…

6

US residential market (1H 2014)

DMPPT market share: 56%

Source: GTM Research U.S. PV Leaderboard, Q2 2014

7

US residential market – 3rd party ownership

3rd-party-owned in 2014 : 68%

Source: GTM Research U.S. Residential Solar Financing 2014-2018

8

DMPPT pro/con

Pro:

• Flexibility of design

• Meets NEC 690.12

• Data monitoring

• Shade performance

• Warranty

Con:

• Cost

• Efficiency (maybe)

• More potential points of failure

• O&M concerns

9

NEC 690.12 (2014)

Code requires rooftop PV conductors greater than 10’ from the array to be de-energized in event of emergency

Compliance options:

Remote relays Roof-mounted inverters

Credit: Midnite Solar, SMA, Tigo

Microinverters and Power Optimizers

10

NEC 690.12 (2014)

Code requires rooftop PV conductors greater than 10’ from the array to be de-energized in event of emergency

Compliance options:

Credit: SolarPro

11

Other emerging inverter trends:

• California Rule 21 (grid support capabilities for PV inverters)

• Storage capabilities for utilities without net metering, or with time-of-day pricing

• NEC 690.12 (2017): proposed module-level shutdown

12

Roadmap




13

Recoverable vs. Non-recoverable loss in PV

0

10

20

Volts

Cu

rre

nt

0 50 100 1500

500

1000

Po

we

r

Global max [A]Local max [B]

Partial shading leads to power loss from reduced irradiance (non-recoverable) and panel mismatch (recoverable)

The better the peak-power tracking (MPPT) granularity, the higher the recoverable power

Mismatched panels in a series string. Either bypass diodes are turned on in the shaded panels (point [A]) or all panels operate at a low current (point [B])

C. Deline, B. Marion, J. Granata, S. Gonzalez. A Performance and Economic Analysis of Distributed Power Electronics in Photovoltaic Systems. NREL Report No. TP-5200-50003. Golden, CO: National Renewable Energy Laboratory, December 2010.

14

Overview of Distributed MPPT

S.M. MacAlpine, M.J.Brandemuehl, R.W.Erickson, “Potential for power recovery: simulated use of distributed power converters at various levels in partially shaded photovoltaic arrays,” in Proc. PVSEC 2011.

+2% +8% +9% +12% Cell

Annual performance improvement under heavy shade

Finer peak-power tracking granularity leads to reduced mismatch

Submodule String Module

MPPT Granularity

15

Sub-module power conversion

1,200

1,300

1,400

1,500

1,600

1,700

0.00 0.20 0.40 0.60 0.80 1.00

PR

OD

UC

TIO

N (

KW

H/K

WP

)

GROUND COVERAGE RATIO

ANNUAL ENERGY - COLORADO RTC SITE

Conventional

Maxim

Sub-module DC-DC converters: • Enable 10-20% tighter row pitch, -or- • 1-3% more energy / panel at the same row pitch

C. Deline et al., “Evaluation of Maxim Module-Integrated Electronics at the DOE Regional Test Centers”, 40th IEEE Photovoltaic Specialists Conference, Denver, CO, 2014

*Mismatch occurs within the module

16

Shading Survey gives real-life conditions

Survey of >60 residential installations provides distribution of shade extent

Credit: Solmetric

Imaging tool shows shade obstructions for residential rooftops Deline, C., Meydbray, J., Donovan, M. (2012) NREL Report No.

TP-5200-54876

M

17

Hanson A, Deline C, MacAlpine S, Stauth J, Sullivan C, “Partial-Shading Assessment of Photovoltaic Installations via Module-Level Monitoring”, IEEE Journal of Photovoltaics, 4 1618-1624, 2014

Shade Recovery Factor (SRF) measures the fraction of shade loss recovered by DMPPT

𝑆𝑅𝐹 = 𝑃𝐷𝑀𝑃𝑃𝑇 − 𝑃𝑁𝑜𝑟𝑚𝑎𝑙 𝑃𝑢𝑛𝑠ℎ𝑎𝑑𝑒𝑑 − 𝑃𝑁𝑜𝑟𝑚𝑎𝑙

=𝑅𝑒𝑐𝑜𝑣𝑒𝑟𝑒𝑑 𝑃𝑜𝑤𝑒𝑟

𝐿𝑜𝑠𝑡 𝑃𝑜𝑤𝑒𝑟

Assessment of typical shade conditions

18







Survey of > 300 systems with power optimizers

Estimated 36% shade recovery on average, 7% average shading loss

Shade %

SRF

19







Survey of > 300 systems with power optimizers

Estimated 36% shade recovery on average, 7% average shading loss Field validation

Shade %

SRF

20

Validation: before and after Power Optimizers

0.60

0.70

0.80

0.90

1.00

1 3 5 7 9 11

Tem

per

atu

re c

orr

ecte

d P

R

Month of year

Initial Results

With Tigo DC-DC


Monthly performance for 3 years

*5.8% more energy with module-level MPPT. 30% shade recovery

21

Roadmap



• Shade modeling in System Advisor Model

22

Inter-row shading in System Advisor Model

How much power is actually lost from inter-row shading? Can you improve kWh/m2 or $/kWh by closer row spacing?

C. Deline et al., “A simplified model of uniform shading in large photovoltaic arrays,” Solar Energy 96, pp 274–282, 2013. Image credit: Thakkar 2010, Applebaum 1979

https://sam.nrel.gov/

23

Experimental validation of SAM model

System 1 System 2

0 0.1 0.2 0.3 0.4 0.5 0.6

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Fraction of string submodules shaded S

Str

ing P

ow

er

Pstr/P

str

0

X = 0

X = 0.66

X = 1

System 1

System 2

C. Deline et al., “A simplified model of uniform shading in large photovoltaic arrays,” Solar Energy 96, pp 274–282, 2013.


24

New: 3D scene shading in SAM

• Relevant for residential rooftop simulations

• Now: linear shade model (low accuracy)

• In progress: new shade database (correct shade loss predictions)

• Future work: distributed (module-level) MPPT performance

SAM 2014 shading tool snapshot

S. MacAlpine, C. Deline, “Simplified Method for Modeling the Impact of Arbitrary Partial Shading Conditions on PV Array Performance”, 31st IEEE PVSC, 2015 (not yet published)


NATIONAL RENEWABLE ENERGY LABORATORY

Questions?

25

Acknowledgments

This work was supported by the U.S. Department of Energy under

Contract No. DOE-AC36-08GO28308 with the National Renewable

Energy Laboratory

Chris Deline

[email protected]

Ph: (303) 384-6359

mailto:[email protected]

Backup slides

27

Side by Side Shading study

• Side by side comparison, 2 identical arrays

o one with string inverter, one with microinverters (or DC converters)

• Synthesized shading using 60% opaque fabric

• Results are weighted based on real residential shading scenarios

Credit: PV Evolution Labs

Deline, C., Meydbray, J., Donovan, M., & Forrest, J. (2014). Photovoltaic shading testbed for module-level power electronics: 2014 Update. http://www.nrel.gov/docs/fy14osti/62471.pdf

(annual)

28

Side by Side Shading study - analysis

Deline, C., Meydbray, J., Donovan, M., & Forrest, J. (2014). Photovoltaic shading testbed for module-level power electronics: 2014 Update. http://www.nrel.gov/docs/fy14osti/62471.pdf

Raw performance data

Site histogram data

1

10

100

1000

10000

0%

10

%

20

%

30

%

40

%

50

%

60

%

70

%

80

%

90

%

Tota

l:

Ava

ilab

le Ir

rad

ian

ce, k

Wh

/m2

Percent of System Shaded

1

1.05

1.1

1.15

1.2

0% 10% 20% 30%

Per

form

ance

sco

re

Annual site shading

0.2

0.4

0.5

0.75

0.9More diffuse

Diffuse %

Microinverter shade benefit

X

=

*Microinverters produced 10-12% more under heavily shaded conditions

performance modeling and testing of distributed … · 2015-09-17 · performance modeling and...

Documents