Affiliations: Faculty of Electrical and Electronics Engineering, KSR College of Engineering, Tiruchengodu, India | Faculty of Electrical and Electronics Engineering, Info Institute of Engineering, Annur, Tamilnadu, India
Note: [] Corresponding author. T.R. Sumithira, Faculty of Electrical and Electronics Engineering, KSR College of Engineering, Tiruchengodu, India. E-mail: sumithra.trs@gmail.com
Abstract: In South India, Tamil Nadu has enormous solar potential, because of its location on the Deccan plateau, with an annual average solar radiation ranging from 4–7 kWhm-2. This study, presents an adaptive neuro-fuzzy inference system (ANFIS) based modelling approach to predict the monthly global solar radiation (MGSR) in Tamil Nadu (India). The statistical performance parameters on predicted and measured values validate the model. In addition, a case study was discussed on the application of predicted solar radiation for sizing of the standalone PV system design including battery and their results were observed to be satisfactory when compared with the other earth regions found in the literature. A detailed investigation on the complete standalone PV system emphasize that the performance is satisfactory from predicted solar radiation and sizing co-efficients for any geographical location and particularly in isolated sites where the global solar radiation data is not always available.
Keywords: Adaptive neuro-fuzzy inference system, monthly global solar radiation, root mean square error, sizing co-efficient, standalone solar photovoltaic system
