Electrical characterization intercomparison of high‐efficiency c‐Si modules within Asian and European laboratories

• Published in: Progress in photovoltaics 2019-07, Vol.27 (7), p.603-622
• Main Authors: Monokroussos, Christos, Salis, Elena, Etienne, Damien, Zhang, XiaoYu, Dittmann, Sebastian, Friesen, Gabi, Morita, Kengo, Stang, Johannes, Herbrecht, Thomas, Fakhfouri, Vahid, Rebeaud, Nicolas, Pavanello, Diego, Müllejans, Harald
• Format: Article
• Language: English
• Subjects: characterization; electrical performance; Electrical properties; Error analysis; Error correction; Error reduction; high‐efficiency c‐Si photovoltaic module; intercomparison; Irradiance; Laboratories; Modules; Photovoltaic cells; round‐robin; Sensitivity analysis; Simulators; Solar cells; Solar simulators; Spectra; spectral mismatch; Spectral sensitivity
• ISSN: 1062-7995; 1099-159X
• DOI: 10.1002/pip.3134

This work presents the results of a high‐efficiency (HE) photovoltaic (PV) module round‐robin intercomparison between five Asian and European ISO/IEC 17025 accredited laboratories and one industrial laboratory based in Europe. The scope of the round‐robin was to examine the measurements comparability for this PV technology with respect to ISO/IEC 17025 laboratory conformity assessment and also to examine the accuracy of step‐like methods towards transient errors against already validated methods. The devices under test were four types of HE c‐Si PV modules with efficiencies varying between 16.5% and 19.0%. The results indicate that a satisfactory agreement was achieved with maximum deviations of 1.59% in Pmax, 1.13% in Isc, and 0.64% in Voc for all devices under test. The weighted standard deviations in Pmax per device type, which can be seen as a conservative estimate of interlaboratory agreement for HE c‐Si PV, ranged within 0.82% to 2.23% (k = 2). The accuracy of step‐like methods towards transient errors was evaluated by comparing a second series of results at fixed Isc for each module under test, eliminating the influence of the effective irradiance measurement. This work suggests that the contribution of capacitive errors was in the range (0.47 ± 0.19) % (k = 2). A spectral mismatch sensitivity analysis showed that an accurate measurement of the spectral irradiance and of the involved spectral responsivities together with the punctual correction for the spectral mismatch can reduce the error in the measurement of PV modules performance of about 2% even in the case of c‐Si against c‐Si and class AAA solar simulators. The round‐robin results indicate that satisfactory agreement was achieved with maximum deviations of 1.59% in Pmax, 1.13% in Isc, and 0.64% in Voc for all devices under test. This work suggests that contribution of capacitive errors was in the range (0.47 ± 0.19) % (k = 2). A spectral mismatch sensitivity analysis showed that spectral mismatch correction can reduce the error in the measurement of c‐Si PV modules of about 2% even in the case of class AAA solar simulators.