A0765
Title: Forecasting multiple attributes considering uncertainties in a coupled energy systems model
Authors: Ulrich Frey - German Aerospace Center (Germany) [presenting]
Felix Nitsch - German Aerospace Center (Germany)
Evelyn Sperber - DLR (Germany)
El Ghazi Achraf - German Aerospace Center (Germany)
Fabia Miorelli - German Aerospace Center (Germany)
Christoph Schimeczek - German Aerospace Center (Germany)
Anil Kaya - KIT (Germany)
Steffen Rebennack - KIT (Germany)
Abstract: Time-series prediction has improved enormously with state-of-the-art machine learning. However, it is hard to integrate ML forecasting methods into energy systems models (ESM) because the trained model has to conform to often strict requirements of the ESM, like class structure, computational limits, or restricted input and output. The open-source forecasting software FOCAPY trains and compares multiple algorithms from basic benchmarks to comprehensive machine learning models. Ways to integrate such production-ready ML models into ESM are also shown. The time series under consideration represents the optimized and aggregated grid interactions of three key actors within the open-source ESM AMIRIS: (a) rooftop photovoltaic systems with battery storage, (b) heat pumps, and (c) electric vehicles. The individual household decisions are obtained using an optimization model for each technology, representative weather regions across Germany, and household types. Results predicting the aggregate demand for a week ahead in an hourly resolution for one year in Germany are presented for each model.
