Home > Published Issues > 2016 > Vol. 5, No. 3, July 2016 >

Validation of predictive models to estimate annual PV production: A case study of odisha

Author(s):Rakesh Kumar Tarai, Paresh Kale
Department of Electrical Engineering, National Institute of Technology Rourkela, Dist. Sundargarh, Odisha-769008, India
International Journal of Smart Grid and Clean Energy, vol. 5, no. 3, July 2016: pp. 160-167
ISSN: 2315-4462 (Print)
ISSN: 2373-3594 (Online)
Digital Object Identifier: 10.12720/sgce.5.3.160-167


Abstract:The decision to install a PV plant depends on three major factors: the climatic and environment conditions of the location, the viability of commercial operations, and the government policies. Economic feasibility of a PV system in the energy market is largely driven by the cost of technology, the cost of installation, and the yield of the plant. Considering uncertain nature of geographical parameters (solar radiation, temperature, dust accumulation, etc.), development of a reliable model to predict the energy output of a plant-to-be installed becomes essential. The model ensures the long-term performance criteria of the PV system. The proposed model considers a case study of Odisha by taking only two meteorological variables collected from 1195 locations: total annual incident global radiation on the PV module and annual average air temperature. The developed model is independent of longitude and latitude, elevation, and other environment conditions. Model is validated using the data collected from SN Mohanty solar power plant situated at Cuttack. The paper focuses on simplification at every stage of the development while validating the preciseness of the model.

Keywords:Solar PV production, predictive model, validation, PVGIS, Odisha

Full Paper.pdf