Performance assessment of photovoltaic power predictions using Univariate models

Integrations of photovoltaic systems to utility grids are widely implemented in many countries to meet increasing electricity demand for socio-economic development while reducing carbon foot print in their electricity generation sectors. However, the erratic nature of solar irradiance poses challenges to the utilities in operation and management. Thus, predicting the photovoltaic power generation is crucial to ensure the usage of the renewable energy output is optimized. Several prediction methods using Univariate models for predicting photovoltaic power were evaluated and compared in this article. The autoregressive and the Hourly-based Prediction models have demonstrated comparable ability in predicting PV power generation based on the climatological conditions selected for this study.