2194. BESS Optimization and the Value of Intraday Electricity Price Forecasts
Invited abstract in session WC-7: Power Trading , stream Risk Management in Commodities and Financial Markets .
Wednesday, 12:30-14:00Room: Clarendon GR.01
Authors (first author is the speaker)
| 1. | Simon Hirsch
|
| House of Energy Markets and Finance, University of Duisburg-Essen | |
| 2. | David Schaurecker
|
| Bits to Energy Lab, ETH Zurich | |
| 3. | David Wozabal
|
| Vrije Universiteit Amsterdam |
Abstract
Grid-scale battery energy storage is increasingly optimized for commercial success in wholesale electricity markets. The rolling intrinsic (RI) strategy for the continuous intraday market, a widely used and hard-to-beat benchmark, maximizes storage value in a myopic manner. This greedy optimization of the RI is also its major drawback: By failing to anticipate future price changes, it may sacrifice more profitable later trading opportunities in favor of short-term gains. To address this, we extend a recent high-frequency-trading dynamic programming formulation of the RI by integrating price forecast information. This enables the algorithm to trade not only based on immediate market conditions but also by submitting virtual positions against the forecast, allowing it to anticipate future price movements. We demonstrate that blindly relying on inaccurate forecasts leads to poorer financial outcomes, compared to the myopic RI. However, carefully constructed informed price forecasts, based on public information, can significantly enhance trading yields. Moreover, we argue that improved forecast accuracy, as measured by standard error metrics, does not automatically translate to higher trading profits, and introduce a novel pipeline for measuring the monetary value of intraday price forecast accuracy in grid-scale storage applications.
Keywords
- Electricity Markets
- Machine Learning
- OR in Energy
Status: accepted
Back to the list of papers