Predictive source management for low power domestic direct current grids

Abstract

The decrease in the price of solar photo voltaic (PV) panels has led to the widespread adoption of the solar power as a renewable energy source, not only at the grid level but also on the roof tops of the residential buildings. The solar PV panels produce direct current (DC) and can be readily used to drive DC powered loads or charge batteries. Direct powering of the loads from the solar panels is hindered by the highly variant nature of solar power generation which depends on a number of external as well as internal factors. The paper proposes a prediction based direct connection between PV panel and DC loads, only when the panel will be able to supply the required power to the load. The loads have been categorized based on how they need to be powered and their priority levels. The data specific to the rooftop was collected over two years using which the prediction of the panel output was carried out. Combining the load categorization and the power prediction a smart management system was designed which was able to decide on how the loads need to be powered and hence were connected to the appropriate source.