Comparative assessment of simplified indexes for the spectral characterisation of photovoltaic systems. (February 2019)

• Record Type: Journal Article
• Authors: Rodrigo, Pedro M.; Varona, Jorge; Soria-Moya, Alberto; Almonacid-Cruz, Bernardo; Fernández, Eduardo F.
• Is Part Of: Measurement. Volume 133(2019)
• Page Start: 1
• Page End: 8
• Publication Date: 2019-02
• Subjects: Meteorological databases -- Photovoltaics -- Spectral characterization -- Spectral index; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2018.09.077
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights : A comparative study of simplified spectral indexes for PV systems is presented. Two existing indexes and one novel introduced index are analysed. Five different PV materials are evaluated for the climate of Granada, Southern Spain. The accuracy of the indexes depends on the PV material. Maximum error relative to the spectral factor is 1.77% (annual) or 4.22% (monthly). Abstract: The electrical output of a photovoltaic system is often related to the plane-of-array global irradiance and the cell operating temperature. However, the spectral distribution of the irradiance also influences the system electrical behaviour, through its coupling with the spectral response of the photovoltaic material employed. Many authors have evaluated the spectral performance of different types of semiconductors under real operating conditions worldwide by using different methods and spectral indexes. However, there are not comparative studies on the accuracy of these methods in the literature. In this paper, two simplified existing indexes and a novel introduced index are compared to the spectral factor method in order to assess their relative accuracy for the climate of Granada, Southern Spain, and considering the most widespread used photovoltaic materials. Results show the election of a suitable index depends on the type of material, availability of spectral response information of the material, computing capabilities and required accuracy. The maximum annual relative error with respect to the spectral factor considering every index and material is 1.77%, while the maximum monthly relative error can reach 4.22%.