PREDICT: Creating a predictive model for electricity transmission losses
In the era of rising electricity prices and the integration of renewable sources, an accurate forecast of losses on power lines is extremely important for the energy network operator. Incorrectly estimated losses cause increased network fees. Since the network fees can make up to 40% of the final electricity price for households, precise forecasts affect all electricity market participants.
The PREDICT project aims to:
- design an AI-based method for fast and precise line loss forecasting
- support the decision-making of network operators to increase network stability and save energy costs
Modern power systems are experiencing an increase in power demand and changes to complex interconnected power networks comprising conventional and renewable energy sources.
Maintaining a balance between generation and demand is important for the reliable operation of power networks. Besides forecasts of generation and demand, forecasts of transmission losses play an important role in the decision-making of system operators.
This project aimed to design and verify transmission loss prediction models using artificial intelligence methods. Predictions were based on historical data using state-of-the-art AI methods, such as support vector regression and a machine learning method based on gradient boosting and decision trees.
Enrichment of model attributes with a mix of weather measurements and data engineered attributes increased the overall prediction accuracy.
Application of the proposed methods for line loss prediction might help the system operators to reduce operating costs and ultimately save energy costs for all electricity market participants.
You can read a detailed report about the PREDICT here.
Meaningful use of artificial intelligence can lead to savings for each energy market participant.
Senior researcher, KInIT
Lead and Researcher
This project was supported by the Slovenská elektrizačná prenosová sústava Fund at the Pontis Foundation.