Solar Prediction Strategy for Managing Virtual Power Stations

Abstract

Indonesia is located on the equator, so it has the potential for solar power to be available and with good sunlight throughout the year. The Indonesian government finally has the consequence to develop a solar power plant and continue to protect the surrounding environment so that it is not polluted and global climate change occurs. Indonesia, which has direct sunlight exposure which is quite promising for predictions of solar power plants in the future. Solar energy generation in the last decade has continued to improve and develop in solar power predictions in a short time. Integration of solar power sources without accurate power predictions can hinder network operations and use of renewable generation sources. To solve this problem, virtual power plant modeling can solve as one of the successful solutions to minimize errors in predicting it. This research studies methods that can efficiently generate significant daily photovoltaic predictions at study sites using data available from the Meteorology, Climatology, and Geophysics Agency (MCGA). The second approach to the model based on RMSE and MAE, can be done virtual modeling of power plants to solve problems and as a management solution to minimize prediction errors. The performance of a verified prediction strategy on the PV module power output and a set based on geographical meteorological station data has been used to simulate a Virtual Power Plant (VPP). The power forecasting prediction refers to the LSTM (Long Short-Term Memory) network and gives an error close to other learning methods, based on the RMS characteristics of 4.19 W/m2 under lead time with different launch times. Applying the VPP model using RMSE can reduce global errors by about 12.37%, and shows great potential.