Fuzzy logic approach for short term solar energy forecasting

Solar photovoltaic (SPV) technology is becoming one of the most promising among all renewable energy technologies. Due to intermittent nature of solar energy, output of solar photovoltaic systems highly affected by the metrological parameters like irradiance, temperature etc. In order to reduce uncertainty in power generation from SPV systems, forecasting of solar energy is utmost important. In the present work, a generalized fuzzy logic based model has been developed for short term solar energy forecasting using the measured solar irradiance data. The hourly data of solar irradiance (in W/m²) for the month of October 2014 has been measured and used as input and actual measured output. To avoid convergence problems the input and output data has been normalized in the range of 0.1 to 0.9. Obtained results are compared with the measured data and found accurate. The performance of the model is evaluated on the basis of mean absolute percentage error (MAPE), which is 1.052% and is in desired limits.