The increasing integration of renewable energies into modern electricity grids is creating new challenges for grid operators and energy traders. Wind and solar energy are highly dependent on the weather and are therefore particularly characterized by spatial and temporal fluctuations. Therefore, precise and reliable electricity forecasts for hours and days into the future are essential for grid stability, market optimization, and cost reduction. Moreover, the increasing availability and quality of power measurements enables a variety of methods to adapt weather and power forecasts to these measurements, with combining these diverse approaches often leading to improved accuracy. However, as different forecasting methods vary in accuracy over time, adaptive combination offers the potential for further enhanced forecasting accuracy and reliability.

Therefore, we present an adaptive combination method that produces a forecast based on power forecasts of different weather models and spatio-temporal information using different methods.

We compare online machine learning methods such as Online Sequential Extreme Learning Machines (OsELMs) that adaptively combine predictions with other benchmark approaches and individual forecasting models.

One particular focus is on the systematic analysis of defined update intervals (daily, weekly, monthly). During each interval, the adaptive learning process accumulates new measurement , forecasts data and adapts its weights automatically in order to optimize the prediction quality.

Both the combination and the adaptive training of the ML models show that the updates lead to imporoved forecasting accuracy. The approaches presented thus enable transmission system operators and wind energy traders to improve the reliability and efficiency of their processes.