Analyzing the Performance of Direct and Indirect Net Load Forecasting Strategies Under Varying Penetration Levels of PV and Wind Power

The increasing share of wind and photovoltaic (PV) power in modern power systems requires accurate net load (NL) forecasts for keeping a stable and economic power system operation. Researchers have been focusing on which net load forecasting (NLF) method can give the best possible accuracy and on whether it can be achieved through the direct or the indirect net load forecasting strategy (NLFS). The studies depended on datasets taken from particular power systems having a specific renewables penetration level (PL). In other words, the obtained accuracy levels were not validated on non-identical power systems and dissimilar renewables’ PLs. This research takes a different route by examining the dependency of the net load forecasting accuracy (NLFA) on wind and PV power PLs, the power system nature and the NLFS. Thus, we employed datasets from two different power systems: New England (NE) and Spain and forecasted the NL over four seasons of the year using both the direct and the indirect NLFSs. Besides the actual data, we assumed additional scenarios representing potential increase in either wind power, PV power or both and forecasted the NL in all of the cases and then compared the results.