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

2025-03-21

AI Summary

This paper addresses inaccurate price forecasting in Germany’s control reserve market, which undermines industrial participants’ profitability and grid regulation efficiency. We propose a shift-correction machine learning framework explicitly designed to account for the asymmetry of the revenue function. Integrating support vector regression, decision trees, and k-nearest neighbors—augmented by domain-informed feature engineering and error-revenue coupling modeling—we quantitatively establish, for the first time, a significant negative correlation between forecasting error (MAE) and annualized revenue: each 1 EUR/MW reduction in MAE yields an incremental revenue gain of 483–3,631 EUR/MW. Empirical evaluation across three control reserve market segments demonstrates a 27.43%–37.31% revenue improvement over baseline models. The framework delivers an interpretable, operationally deployable forecasting paradigm for ancillary service bidding optimization.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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.

Problem

• Predict bidding prices in ancillary service markets using ML
• Improve revenue for participants in German control reserve market
• Address volatility from renewable energy integration in electricity grids

Innovation

• Machine learning models predict bidding prices
• Offset adjustment enhances revenue forecasting
• Improved revenues by 27-37% over baselines