For energy producers, power price prediction is an essential task and its accuracy is crucial for decisions in trading and risk management. Among the alternative methods for time series analysis, Machine Learning is becoming a promising alternative to predict power forecasting.
In this work, we will test alternative Machine Learning recipes (Neural Net, Random Forest etc.) and optimize both the model architecture and the feature space to use the model in production. This will create a robust tool to forecast power prices under the current unprecedented market conditions.
- Proven, strong background in Machine Learning methods
- Python programming skills
- Good oral and written communication skills
The project will be conducted as an external master thesis in collaboration with Energie Baden-Württemberg (EnBW) and the student will be employed at EnBW during the work.