Forecasts for Operations Management Systems (AI-Based Load Forecasting)

Forecasts play a crucial role in the optimized operation of energy systems. Short-term forecasts allow the expected future behavior to be factored into real-time control, which enables more complex control strategies (particularly multi-use strategies) and conserves resources. A predictive energy management system is only as good as the reliability of the underlying forecasts. We therefore develop forecasting methods for electrical loads and use them in our operations management to optimally schedule the deployment of flexibility resources. We employ statistical methods, as well as techniques from the fields of machine learning and deep learning. In addition to load forecasting, we use our methodological toolkit to generate other relevant energy system variables, such as generation forecasts, forecasts of battery electric vehicle parking durations, or price forecasts for energy markets.

Our R&D Services in this Area Include:

• Development of short-term load forecasts based on classical statistical methods or through the use of artificial intelligence
• Development of short-term forecasts for decentralized energy generation plants based on historical weather and generation data as well as external weather forecasts
• Integration of forecasting algorithms into various internal or external operational management algorithms