ML-Based Bidding Price Prediction for Pay-As-Bid Ancillary Services Markets: A Use Case in the German Control Reserve Market

TL;DR

This work tackles price forecasting for pay-as-bid ancillary service markets in the German control reserve context, aiming to maximize industrial revenues while maintaining grid stability. It proposes a day-ahead forecasting methodology evaluated across multiple machine-learning families, with a novel offset adjustment to mitigate asymmetry in the revenue function. Key findings show that model choice (especially SVR and XGBoost) and the offset adjustment drive most gains, yielding 27.43%–37.31% higher revenue than market baselines, while retraining frequency and hyperparameter tuning contribute marginal improvements. The approach offers a practical pathway for market participants to optimize bidding strategies in pay-as-bid markets and could extend to other markets with similar designs and constraints.

Abstract

The increasing integration of renewable energy sources has led to greater volatility and unpredictability in electricity generation, posing challenges to grid stability. Ancillary service markets, such as the German control reserve market, allow industrial consumers and producers to offer flexibility in their power consumption or generation, contributing to grid stability while earning additional income. However, many participants use simple bidding strategies that may not maximize their revenues. This paper presents a methodology for forecasting bidding prices in pay-as-bid ancillary service markets, focusing on the German control reserve market. We evaluate various machine learning models, including Support Vector Regression, Decision Trees, and k-Nearest Neighbors, and compare their performance against benchmark models. To address the asymmetry in the revenue function of pay-as-bid markets, we introduce an offset adjustment technique that enhances the practical applicability of the forecasting models. Our analysis demonstrates that the proposed approach improves potential revenues by 27.43 % to 37.31 % compared to baseline models. When analyzing the relationship between the model forecasting errors and the revenue, a negative correlation is measured for three markets; according to the results, a reduction of 1 EUR/MW model price forecasting error (MAE) statistically leads to a yearly revenue increase between 483 EUR/MW and 3,631 EUR/MW. The proposed methodology enables industrial participants to optimize their bidding strategies, leading to increased earnings and contributing to the efficiency and stability of the electrical grid.

Figures (20)

• Figure 1: Graphical overview of the methodology.
• Figure 2: Schematic representation of retraining.
• Figure 3: Comparison of forecast without (top) and with offset adjustment (bottom).
• Figure 4: German control reserve market and its submarkets. (based on nextkraftwerke_regelenergie)
• Figure 5: A sample of data for the different markets is shown.