Photovoltaic power predictions using modified adaptive response rate exponential smoothing method

Abstract

Precipitous depletion of natural resources and exponential population growth are stimulating the rapid development of electric power industries. Increased awareness of the global community on the benefits of renewable energy is driving the conventional power grid systems to a new paradigm. Electric power grids are evolving rapidly where incorporation of higher proportion of photovoltaic power generation is observed. The modern grid structure requires contemporary energy management systems for better quality of services. Therefore, prediction is essential for effective energy management and solar energy trading. This paper proposed a Modified Adaptive Response-rate Exponential Smoothing method for predicting photovoltaic power generation for a location near the equator with tropical climate. Four other time series models were used to forecast the photovoltaic power generation for the same location and comparison between the proposed models with those time series models were performed. The results show that the proposed model demonstrates improved performance than the other time series models selected for this study.