Impacts of Solar Penetration on Short-Term Net Load Forecasting at the Distribution Level

Abstract

Motivated by the environmental concerns and a desire to reduce electricity bills, the number of residential and commercial customers with solar panels has increased significantly. Since the installation of solar panels is Behind-The-Meter, utilities can only see the net load (load minus solar generation) of each customer. Distribution-level net load thus represents the net demand of a substation, a feeder, a transformer or a customer at the distribution side. Reliable forecasts of the distribution-level net load provide valuable decision support information for utilities, however, forecasting distribution-level net load is difficult since the net load is determined by complex interactions between customer energy usage patterns and weather conditions. This paper analyzes the impacts of different levels of solar penetration on short-term net load, identifies the key solar relevant factors, and provides a framework for distribution-level net load forecasting. Numerical testing demonstrates the high forecasting accuracy based on the synthesized load data and scaled solar generation. Our method has potential applications in forecasting peak load and time, substation/feeder with batteries, and distribution-level net load with wind or other dispatchable renewable resources.