TY - JOUR
T1 - Progress in Photovoltaic Module Calibration: Results of a Worldwide Intercomparison Between Four Reference Laboratories
AU - Dirnberger, D.
AU - Kräling, U.
AU - Müllejans, H.
AU - Salis, E.
AU - Emery, K.
AU - Hishikawa, Y.
AU - Kiefer, K.
N1 - Publisher Copyright:
© 2014 IOP Publishing Ltd.
PY - 2014/10/1
Y1 - 2014/10/1
N2 - Measurement results from a worldwide intercomparison of photovoltaic module calibrations are presented. Four photovoltaic reference laboratories in the USA, Japan and Europe with different traceability chains, measurement equipment and procedures, and uncertainty estimation concepts, participated. Seven photovoltaic modules of different technologies were measured (standard and high-efficiency crystalline silicon, cadmium telluride, single and double-junction amorphous and micromorph silicon). The measurement results from all laboratories and for all devices agreed well. Maximum power for the crystalline silicon samples was within±1.3% for all thin-film modules roughly within±3%, which is an improvement compared to past intercomparisons. The agreement between the results was evaluated using a weighted mean as a reference value, which considers results-specific uncertainty, instead of the widely used unweighted arithmetic mean. A further statistical analysis of all deviations between results and the corresponding reference mean showed that the uncertainties estimated by the participating laboratories were realistic, with a slight tendency towards being too conservative. The observed deviations of results from the reference mean concerned mainly short-circuit current and fill factor. Module stability was monitored through repeated measurements at Fraunhofer ISE before and after measurements at each of the other participating laboratories. Based on these re-measurements, stability problems that occurred for some thin-film modules and influenced the results were analyzed and explained in detail.
AB - Measurement results from a worldwide intercomparison of photovoltaic module calibrations are presented. Four photovoltaic reference laboratories in the USA, Japan and Europe with different traceability chains, measurement equipment and procedures, and uncertainty estimation concepts, participated. Seven photovoltaic modules of different technologies were measured (standard and high-efficiency crystalline silicon, cadmium telluride, single and double-junction amorphous and micromorph silicon). The measurement results from all laboratories and for all devices agreed well. Maximum power for the crystalline silicon samples was within±1.3% for all thin-film modules roughly within±3%, which is an improvement compared to past intercomparisons. The agreement between the results was evaluated using a weighted mean as a reference value, which considers results-specific uncertainty, instead of the widely used unweighted arithmetic mean. A further statistical analysis of all deviations between results and the corresponding reference mean showed that the uncertainties estimated by the participating laboratories were realistic, with a slight tendency towards being too conservative. The observed deviations of results from the reference mean concerned mainly short-circuit current and fill factor. Module stability was monitored through repeated measurements at Fraunhofer ISE before and after measurements at each of the other participating laboratories. Based on these re-measurements, stability problems that occurred for some thin-film modules and influenced the results were analyzed and explained in detail.
UR - http://www.scopus.com/inward/record.url?scp=84907152198&partnerID=8YFLogxK
U2 - 10.1088/0957-0233/25/10/105005
DO - 10.1088/0957-0233/25/10/105005
M3 - Article
AN - SCOPUS:84907152198
SN - 0957-0233
VL - 25
JO - Measurement Science and Technology
JF - Measurement Science and Technology
IS - 10
M1 - 105005
ER -