Identifying methodologies for medium-term ancillary market price forecasting

Thumbnail Image

Files

master_Bryan_Nathaniel_2025.pdf (6.84 MB)

URL

Journal Title

Journal ISSN

Volume Title

School of Engineering | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2025-10-24

Department

Major/Subject

Innovative Sustainable Energy Engineering

Mcode

Degree programme

Nordic Master Programme in Innovative and Sustainable Energy Engineering

Language

en

Pages

86

Series

Abstract

A qualitative and quantitative study of historical and future developments in Germany and Sweden’s Ancillary Service Markets. Ancillary services (AS) are increasingly important for balancing mechanisms, as more variable renewable energy sources (VRES) are integrated into the electricity systems. This creates attractiveness in the AS markets, especially for balancing providers to create new revenue opportunities. Having a thorough analysis about their historical and future price trends could unlock valuable insights related to price forecasting, especially for medium-term time range. This thesis first conducts a comprehensive historical analysis of the German and Swedish AS markets, combining both qualitative insights with quantitative methods. From the historical trends, Germany’s aFRR, mFRR, as well as Sweden’s mFRR products has shown an upward trend over time. Regarding the price movements, they are also influenced by external factors, such as the nuclear phase-out in Germany and hydropower generation in Sweden. The correlations analysis shows that Spearman correlation provides complementary insights to Pearson. Moreover, the link between renewable generation (especially solar in Germany) and AS prices has strengthened over time. The thesis then evaluates different forecasting approaches, from statistical methods to machine learning models. Random Forest (RF) and XGBoost (XGB) perform best in Germany, while Support Vector Regression (SVR) and XGB show stronger results in Sweden. Overall, Germany’s forecasts achieve better accuracy, mainly due to its market size, generation mix, and pricing mechanism. The study also discusses the outlook of external drivers from dynamic, periodic, or structural factors in both countries. Looking ahead, the future of ancillary service prices will depend on the technological, economic, and policy developments.

Description

Supervisor

Syri, Sanna

Thesis advisor

Nikzad, Dario

Keywords

ancillary service markets, medium-term price forecasting, electricity market dynamics, variable renewable energy, external drivers, statistical methods

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202512169329

Collections

[dipl] Insinööritieteiden korkeakoulu / ENG