Estimating photovoltaic energy potential from a minimal set of randomly sampled data

• Published in: Renewable energy Vol. 97; pp. 457 - 467
• Main Authors: Bocca, Alberto, Bottaccioli, Lorenzo, Chiavazzo, Eliodoro, Fasano, Matteo, Macii, Alberto, Asinari, Pietro
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2016
• Subjects: Electricity; Errors; Estimating; Fast energy potential assessment; Geographic information systems; Methodology; Photovoltaic cells; Photovoltaic solar energy; Renewable energy; Solar cells; Solar energy; Stochastic data models; MED, Italy; Renewable energy; Photovoltaic solar energy; Fast energy potential assessment; Stochastic data models
• ISSN: 0960-1481; 1879-0682
• DOI: 10.1016/j.renene.2016.06.001

The remarkable rise of photovoltaics in the world over the past years testifies of the great improvement in the use of solar energy. Opportunities for further new PV installations are being sought, especially power plants in areas with as yet little exploited solar energy potential. In this paper, we describe a methodology for generating estimation models of PV electricity for installations in large regions where only a few scattered data or measurement stations are available. For validation only, application of this methodology was performed considering Italy, where estimations can be benchmarked using the Photovoltaic Geographical Information System (PVGIS) by the Joint Research Centre of the European Commission. The results show that the mean absolute errors were usually lower than 4%, compared to the PVGIS data, for about 90% of the estimates of PV electricity, and about 6% for the greatest mean error. •Methodology for estimating PV potential where a few sampled data are available.•Methodology is validated by considering 21 test cases in Italy.•Modeling errors are typically lower than 4% compared to PVGIS database.•The methodology is indicated for developing countries with high solar irradiation.