Prediction and Optimization of Power Loads Using AI Forecasting Models for Cost and Supply-efficient Power Grids

Project Description

Load forecasting errors have significantly escalated over the years as energy system modeling techniques have proven to be too static for frequent changes of contributing factors such as PV and e-mobility due to rapid technological advancements. These inaccuracies necessitate costly last-minute adjustments, burdening not only energy suppliers but ultimately also end consumers. POWERCAST aims to introduce a new paradigm for load forecasting that considers the dynamic nature of consumption and production structures within the energy system. By leveraging adaptive artificial intelligence (AI) methods, the developed forecasting models can adjust to rapid changes by producers and consumers within the area under consideration and can be transferred to novel but similar scenarios. UseCase#1 focuses on improving short-term forecasts that (i) contribute to economic efficiency of power grids and (ii) facilitate better verification of compliance with safety standards by grid operators. An early warning system for negative loads developed in UseCase#2 paves the way for faster expansion of PV and thus contributes to the faster achievement of Austria’s renewable energy goals.

Project Consortium

  1. RISC Software GmbH
  2. Energieinstitut an der Johannes Kepler Universität Linz
  3. Austrian Power Grid AG
  4. HAKOM Time Series GmbH

Project duration

01.04.2024 - 31.03.2027

Funding provider


Funding program

AI for Green
Bmk Logo Srgb
Bundesministerium für Klimaschutz, Umwelt, Energie, Mobilität, Innovation und Technologie
Ffg Logo De Rgb 1500px
Scroll to Top