ENERGY. FUTURE. ZAE. DEGRADATION AND RESILIENCE OF PRE-CRACKED PV-MODULES

Transcription

1 ENERGY. FUTURE. ZAE. DEGRADATION AND RESILIENCE OF PRE-CRACKED PV-MODULES AGCS Expert Days 2017, November 2nd, 2017, München Dr.-Ing. Claudia Buerhop

2 PV-inspection ZAE Bayern 2

3 Performance study of pre-cracked modules Visualization of cracks using imaging techniques Electroluminescence EL-imaging EL-images visualize e. g. cracks in solar cells normally no impact on electrical power output Electrically inactive cell area by differing shades of grey and black impact on electrical power output Nominal module power P 0 = W Actual module power P = W Module 165 ZAE Bayern 3

4 Study of crack evolution in PV-modules at real operating conditions 1. EL-imaging visualization of cracks in PV-modules, and now? 2. Experimental study field test, lab study and FEM-simulations 3. Case study I cleaning procedure 4. Case study II severe storm event Bayerisches Zentrum für Angewandte Energieforschung e. V. Alle Rechte vorbehalten, auch bezüglich jeder Verfügung, Verwertung, Reproduktion, Bearbeitung und Weitergabe sowie für den Fall von Schutzrechtanmeldungen.

5 Performance study of pre-cracked modules How does the module look like after one year of field exposure? At the beginning wind sn ow Temperature differences P0/P0 = 100% P/P0 = 99% Module 165 Option A: unchanged P/P0 = 98% Option B: Low impact P/P0 = 97% Option C: Strong impact ZAE Bayern 5

6 Performance study of pre-cracked modules Approach Field study 54 modules pre-cracked 1 year continuous monitoring of weather data, string power, module temp. 3 times: IR-, EL- imaging, module power Load testing in the lab 20 modules pre-cracked Static loading over- and underpressure simulating snow and wind loads EL-imaging, IV-curves, strain measurements FEM-simulations Abaqus Stress distribution ZAE Bayern 6

7 Oct 2015 Field exposure PV-plant and weather conditions (in Bavaria, Oct Dec. 2016) C. Buerhop, Wirsching, S., Bemm, A., et al., "Evolution of cell cracks in PV-modules under field and laboratory conditions," Progress in Photovoltaics, 2017, submitted. ZAE Bayern 7

8 loading pressure in Pa Lab testing Test facility Module mounting system Over- and underpressure EL-imaging (InGaAscamera, 100 Hz) Strain gauges Mobile flasher time interval ZAE Bayern 8

9 Lab testing EL-images with increasing loading and unloading At loading crack opening At unlaoding crack closure Module 165 ZAE Bayern 9

10 module power Ppk in W Lab testing Power outcome at loading and unloading intial, p =0 Pa at p = 400 Pa at p = 800 Pa at p =1600 Pa after p =1600 Pa at p =2800 Pa final, after p =2800 Pa underpressure in Pa unloaded loaded Buerhop-Lutz, C.,et al.: 'Performance analysis of pre-cracked PV-modules at realistic Module 484 loading conditions', Proc. 33rd EU-PVSEC, Amsterdam, The Netherlands, 2017 ZAE Bayern 10

11 pressure in Pa pressure in Pa Lab testing Wind and snow loads V max = 137 km/h V mean = 32 km/h wind velocity in km/h snow height [m] wind standard fresh settled old wet standard DIN EN , Eurocode 1 - Actions on structures - Part 1-3: General actions - Snow loads, 2010, DIN EN , Eurocode 1: Actions on structures - Part 1-4: General actions - wind loads, 2010, ZAE Bayern 11

12 Lab testing EL-images at loaded (400 Pa Pa) and unloaded state EL-images at ex-post loaded state simulating normal operating conditions initial p = 0 Pa at p = 400 Pa at p = 5000 Pa After p = 5000 Pa at p = 0 Pa Ex-post p = 300 Pa Ex-post p = 400 Pa P/P0 = 100% P/P0 = 98% P/P0 = 89% P/P0 = 98% P/P0 = 98% P/P0 = 99% Module 165 ZAE Bayern 12

13 FEM-simulations Stress distribution across the material layers Simulation assumptions: Static, planar loads No busbars, no gridfingers No cracks Linear, elastic material properties pressure ZAE Bayern 13

14 FEM-simulations Stress distribution on cell level Quarter model, p = 1000 Pa C. Buerhop-Lutz et al., "Lifetime and Degradation of Pre-damaged PV-Modules Field study and lab testing," in 44th IEEE, Washington, USA, ZAE Bayern 14

15 FEM-Simulation vs. EL-images Crack orientation loaded state FEM-simulation Module 744, 3600 Pa Difference image between EL-images of loaded and intial state Unloaded state ZAE Bayern 15

16 EL-image crack distribution Difference image between EL (4600 Pa) and EL (0 Pa) Module 555, 4600 Pa C. Buerhop-Lutz et al., "Lifetime and Degradation of Pre-damaged PV-Modules Field study and lab testing," in 44th IEEE, Washington, USA, ZAE Bayern 16

17 Frequency average wind speed gentle breeze strong breeze strong gale hurricane force Pressure we in Pa Okt 2015 Field exposure PV-plant and weather conditions (in Bavaria, Oct Dec. 2016) Temp. diff. 25% 20% 15% 10% 5% 0% Max. wind speed hurricane max daily wind speed in km/h wind pressure we in Pa C. Buerhop, Wirsching, S., Bemm, A., et al., "Evolution of cell cracks in PV-modules under field and laboratory conditions," Progress in Photovoltaics, 2017, submitted. ZAE Bayern 17

18 May 2016 / Oct 2016 Oct 2015 Field data EL-imaging No newly cracked cells changes in intensity distribution possible ZAE Bayern 18

19 module power in W Field data Module power measurement INV 1 INV 2 INV 3 INV 4 INV 5 INV 6 moderate + few cracks good + few cracks bad + cracks moderate + cracks mixed mixed module ID with description Oct 2015 May 2016 August 2016 C. Buerhop et al., "Evolution of cell cracks in PV-modules under field and laboratory conditions," Progress in Photovoltaics, 2017, submitted. ZAE Bayern 20

20 Performance study of pre-cracked modules Conclusion static planar loading = simulating snow or wind loads At the beginning Cracks open and close Open cracks power loss possible Moderate weather conditions including several severe wind scenarios no detectable / measurable changes Existing cracks do not necessarily impact the performance negatively at real operating conditions. Module 165 Option A: unchanged ZAE Bayern 21

21 Case study I - Cleaning of PV-systems ZAE Bayern 22

22 Case study I - Cleaning of PV-systems ZAE Bayern 23

23 Case study I - Cleaning of PV-systems p = 200 Pa Walking over modules P = 228W BEFORE loading procedure = cleaning P = 227W AFTER loading procedure = cleaning ZAE Bayern 24

24 Case study I - Cleaning of PV-systems Simulating subsequent normal operating conditions v wind = 55 km/h v wind = 96 km/h, storm v wind = 124 km/h h snow = 5 cm (wet snow) 20 cm (fresh snow) h snow = 15 cm (wet snow) 60 cm (fresh snow) p = 0 Pa p = 200 Pa p = 600 Pa p = 1000 Pa p = 0 Pa P/P0 = 100% P/P0 = 99% P/P0 = 91%W P/P0 = 89% P/P0 = 99% Module 124 ZAE Bayern 25

25 Case study II - Extreme weather events Hurricanes or hail storms Cadolzburg, Germany, 18th August 2017 Max. wind speed = 91 km/h source: ZAE Bayern 26

26 Case study II - hail storm Simulating subsequent normal operating conditions p = 0 Pa p = 200 Pa p = 400 Pa p = 1000 Pa p = 0 Pa v wind = 55 km/h, high wind v wind = 78 km/h, strong gale v wind = 124 km/h, h snow = 5 cm (wet snow) h snow = 10 cm (wet snow) hurricane force 20 cm (fresh snow) 40 cm (fresh snow) P0/P0 = 100% P/P0 = 100% P/P0 = 100% P/P0 = 99% P/P0 = 97% Module 336 ZAE Bayern 29

27 Degradation of pre-cracked PV-modules Cell cracks only half as bad? ZAE Bayern 30

28 Degradation of pre-cracked PV-modules loading tests in the lab At low loads existing cracks open At high loads new cracks are initiated -- power reduction due to open cracks field measurements At moderate weather no measurable/visible changes Long-term performance of cracked PV-modules? First answers: Resilience and mechanical stability of pre-cracked PV-modules ZAE Bayern 31

29 THANK YOU FOR YOUR ATTENTION! ACKNOWLEDGEMENT ZAE Bayern gratefully thanks the German Federal Ministry for Economic Affairs and Energy (BMWi) for financial funding of this project. We thank the Allianz Risk Consulting GmbH /Allianz Zentrum für Technik (AZT) gratefully for providing the batch of PV-modules and supporting this study. Bayerisches Zentrum für Angewandte Energieforschung e. V. Alle Rechte vorbehalten, auch bezüglich jeder Verfügung, Verwertung, Reproduktion, Bearbeitung und Weitergabe sowie für den Fall von Schutzrechtanmeldungen.