We are seeking motivated and talented Master's students to participate in a cutting-edge research project focused on the transformation of electricity markets in the context of the transition towards high shares of renewable power generation. This project revolves around the development and application of the simulation tool-box called ASSUME (Agent-Based Electricity Markets Simulation Toolbox).

Background:

The transition towards renewable energy sources necessitates constant evolution in market mechanisms, sector coupling, and the emergence of new market platforms. However, introducing or altering market designs can have unforeseen consequences on other markets and their participants. Recent changes in the German reserve markets have highlighted the complex interplay between markets, necessitating the need for advanced tools and simulation models.

Project Description:

With ASSUME, our research team aims to create a highly modular and user-friendly energy market simulation toolbox that integrates state-of-the-art reinforcement learning methods. We have already conducted successful tests using various reinforcement learning algorithms in multi-agent simulations of electricity markets, yielding promising results. This toolbox will empower researchers and practitioners to dynamically analyze market designs and bidding strategies, adapting to the rapidly changing landscape of our energy system.

Objectives:

The Master's thesis project will focus on one or more of the following objectives:

• Exploration of Market Dynamics: Analyze the impact of different market designs, mechanisms, and policies on overall market performance and participant behavior. Evaluate the effectiveness of existing markets and propose improvements.
• Bidding Strategy Optimization: Develop and test innovative bidding strategies for different market scenarios, considering factors such as renewable energy integration, demand fluctuations, and policy changes.
• Investigate and model potential instances of market power abuse in response to various market conditions. Implement safeguards and strategies to mitigate such risks.

Requirements:

• Strong existing programming skills in Python.
• Prior knowledge in the energy industry (optional)

Formalities:

The thesis can be written in German or English. Please apply with a short letter of motivation (max. ½ page), your CV, and a current grade transcript. Work can commence immediately.

Relevant introductory literature:

Literature overview on Deep Reinforcement Learning (DRL) in the electricity market.

Zhu, Ziqing; Hu, Ze; Chan, Ka Wing; Bu, Siqi; Zhou, Bin; Xia, Shiwei (2023): Reinforcement learning in deregulated energy market: A comprehensive review. In: Applied Energy 329, S. 120212. DOI: 10.1016/j.apenergy.2022.120212.

Di Cao; Hu, Weihao; Xu, Xiao; Dragičević, Tomislav; Huang, Qi; Liu, Zhou et al. (2020): Bidding strategy for trading wind energy and purchasing reserve of wind power producer – A DRL based approach. In: International Journal of Electrical Power & Energy Systems 117, S. 105648. DOI: 10.1016/j.ijepes.2019.105648.

Publications with ASSUME.

Harder, Nick & Weidlich, Anke & Staudt, Philipp. (2023). Modeling Participation of Storage Units in Electricity Markets using Multi-Agent Deep Reinforcement Learning. 439-445. 10.1145/3575813.3597351.

Harder, Nick & Qussous, Ramiz & Weidlich, Anke. (2023). Fit for purpose: Modeling wholesale electricity markets realistically with multi-agent deep reinforcement learning. Energy and AI. 14. 100295. 10.1016/j.egyai.2023.100295.