Day-ahead solar forecasting improved for energy sector by up to 13%

In brief

Researchers at North Carolina State University have developed machine-learning techniques that improve day-ahead solar power forecasts by up to 13% compared with the most consistently performing individual model. Accurate forecasts are increasingly important as solar generation expands, because variations in sunlight complicate the task of balancing electricity supply and demand. The researchers evaluated seven statistical and artificial-neural-network models using weather and power-generation data collected from 2019 to 2022 by two California utilities: the Imperial Irrigation District and the Los Angeles Department of Water and Power.

No single forecasting model performed best under all conditions, so the team tested two ensemble approaches that combined information from multiple models or locations. Weighted averaging improved forecast accuracy by up to 11% for the Imperial Irrigation District, while a multi-input method achieved gains of up to 13% for the Los Angeles utility. The results show that ensemble forecasting can make solar integration more reliable, but the most effective method depends on local weather patterns and grid characteristics, meaning models must be tested and adjusted for each region.

Source: Tech Xplore

‍

‍