High-frequency intraday trading for battery storages
David Schaurecker, David Wozabal, Nils Löhndorf, Thorsten Staake · 2025
Read High-frequency intraday trading for battery storages by David Schaurecker, David Wozabal, Nils Löhndorf, Thorsten Staake (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-06932.txt.
Introduction
David Schaurecker and three co-authors wrote this 2025 arXiv paper on trading electricity with grid batteries at millisecond speed. Intraday power markets update continuously, and a battery earns by seizing price gaps the moment they appear. The standard rolling intrinsic strategy re-optimizes too slowly to catch them. The team rebuilt it around the full order book, with a fast approximation replacing exact optimization.
Their dynamic programming shortcut ran a hundred to a thousand times faster than the exact solver without denting profits. Backtested on a year of German market data, reacting to every order update earned fifty-eight percent more than hourly re-optimization and fourteen percent more than minutely updates. A learned extension added another eight percent out of sample. The message to storage operators is blunt: in continuous markets, speed itself is a revenue line.
Catalog record
| CREATOR | David Schaurecker, David Wozabal, Nils Löhndorf, Thorsten Staake |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-06932
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
High-frequency intraday trading for battery storages David Schaurecker, David Wozabal, Nils Löhndorf, Thorsten Staake Abstract Maximizing revenue for grid-scale battery energy storage systems in continuous intraday electricity markets requires strategies that are able to seize trading opportunities as soon as new information arrives. This paper introduces and evaluates a computationally efficient, high-frequency implementation of the rolling intrinsic trading strategy for battery energy storage systems on the intraday market for power. By combining the established rolling intrinsic logic with a full limit order book representation and a fast dynamic programming approximation, our method explicitly considers the continuously updated list of buy and sell offers, market rules, a linear approximation of degradation, and other technical parameters at a millisecond resolution. The standard rolling intrinsic strategy is adapted for continuous intraday electricity markets and solved using a dynamic programming approximation that is two to three orders of magnitude faster than an exact mixed-integer linear programming solution. A detailed backtest over a full year of German order book data demonstrates that the proposed dynamic programming formulation does not reduce trading profits and enables the policy to react to every relevant order book update, enabling realistic rapid backtesting. Our results show the significant revenue potential of high-frequency trading: our policy earns 58 58 % more than when re-optimizing only once every hour and 14 14 % more than when re-optimizing once per minute, highlighting that profits critically depend on trading speed. Furthermore, we leverage the speed of our algorithm to train a parametric extension of the rolling intrinsic, increasing yearly revenue by 8.4 8.4 % out of sample. Journal: Applied Energy 1 Introduction Grid-scale battery energy storage systems (BESS) have emerged as a promising solution to address short-term variability in renewable electricity production. Batteries store excess energy generated during periods of high production and release it when generation is low, providing much-needed flexibility and stabilizing the grid ( Martínez-Barbeito et al., 2023 ; Ullah et al., 2024 ) . In addition, they help manage peak demand, minimize reliance on fossil fuel peaker plants, and boost overall energy system efficiency. The rapid increase in grid-scale storage installations underscores this potential ( Saldarini et al., 2023 ; IRENA, 2017 ) , with the projected adoption of grid-scale BESS to increase significantly by 2030. In liberalized electricity markets, the evaluation of viable business models for BESS is crucial to this transformation. The economic feasibility and revenue streams of these systems depend on their ability to participate in various short-term energy markets, engaging in energy arbitrage, frequency regulation, and demand response. Understanding the potential of these business models helps investors and legislators make the right decisions for a sustainable energy future and increases the adoption of storage technologies. In particular, informed trading decisions for intraday (ID) trading become increasingly relevant, with ID trading volumes growing each year ( Koch and Hirth, 2019 ) and market participants shifting their trading towards short-term markets. Finding good trading strategies requires extensive backtesting over long time horizons. However, backtesting the continuous ID market is computationally expensive. Its millisecond frequency generates millions of decision points, making long-term simulations challenging. This paper introduces a highly realistic, high-frequency implementation of the rolling intrinsic trading strategy for battery energy storage systems. By combining the rolling intrinsic logic with a full limit order book representation and a fast dynamic programming approximation, this approach integrates the full granular evolution of the market with exchange regulations and physical storage constraints. As we argue below, literature on the subject is limited, as most authors adopt a rather stylized view of the intraday market, avoiding most of its complexity and thereby yielding suboptimal strategies and a biased picture of revenue potentials. Most relevant related literature on coordinated bidding in electricity spot markets focuses on the day-ahead market, such as Fleten and Kristoffersen (2008) , or treats the intraday market (IDM) as a single trading decision stage ( Faria and Fleten, 2011 ; Löhndorf et al., 2013 ; Kraft et al., 2023 ) . Applications range from pure trading strategies to asset-focused approaches, such as for hydro plant or battery system operation. Some publications consider a small number of repeated trading decisions on intraday markets for storages ( Löhndorf and Wozabal, 2023 ) and other applications ( Aid et al., 2016 ; Wozabal and Rameseder, 2020 ; Kuppelwieser and Wozabal, 2023 ) . Interest in short-term trading strategies for grid-scale BESS has surged recently, a topic explored by, e.g., Sage et al. (2024) ; Cornejo et al. (2025) ; Zhang et al. (2025) . Literature that exclusively focuses on (automated) trading strategies for the intraday market, while capturing its full complexity, is scarce. Some recent work presents approaches for single-, or multi-market trading strategies for storage. Jiang and Powell (2015) study short-term trading for a battery storage on real-time markets as a Markov Decision Process. Braeuer et al. (2019) , and more recently, Seifert et al. (2024) , present a European multi-market trading strategy for battery systems across three markets, trading at discrete time intervals and connecting the sequential reserve, day-ahead and intraday market decision stages. With a 4-hour time resolution, their ability to react to short-term changes in prices is very limited. Furthermore, the resulting smoothing of price spikes significantly reduces the profits a battery could extract from intraday markets. While co-optimizing storage assets across multiple markets (e.g., day-ahead, reserve, and intraday) is highly relevant in practice, integrating the continuous intraday market into multi-market frameworks currently requires heavy simplifications, such as artificially discretizing time into fixed decision stages. These simplifications severely distort the true profit potential of high-frequency trading. We focus exclusively on capturing the full complexity of the continuous intraday market because its rapidly growing volumes and high-frequency dynamics demand specialized, highly reactive algorithms. By isolating this market, we provide a highly accurate, fast-executing benchmark standalone model. We view understanding and accurately modeling this specific market as a critical foundational step before such algorithms can be expanded into broader multi-market environments without sacrificing intraday realism. Furthermore, sophisticated operation within the continuous intraday market is becoming increasingly critical for battery storage systems, as Europe’s ancillary markets face saturation from the surging volume of installed battery capacity. The myopic rolling intrinsic (RI) trading strategy is nearly optimal for gas storage operation ( Löhndorf and Wozabal, 2021 ) . The strategy naturally translates to other storage types, like batteries. Consequently, the RI is frequently used as a benchmark for more sophisticated strategies ( Bertrand and Papavasiliou, 2020 ; Boukas et al., 2021 , e.g.,) . Recent work by Semmelmann et al. (2025) is most closely related to our work by presenting an evaluation of RI trading for a battery on the continuous intraday market. However, their study relies on aggregated transaction data rather than the full limit order book, missing critical aspects of the market’s microstructure, and does not account for battery degradation costs. While we adopt a simplified linear degradation model in this paper to maintain computational tractability at high execution frequencies, explicitly incorporating these costs into the optimization maps trading decisions more realistically and prevents the over-utilization of the asset. Generally speaking, there are only a handful of papers that model the intraday market in its full complexity and implement trading strategies