EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3779. Bidding Behavior Analysis in Electricity Markets Using Inverse Reinforcement Learning
Invited abstract in session MB-9: Electricity Market Design, stream Energy Markets.
Monday, 10:30-12:00Room: 10 (building: 116)
Authors (first author is the speaker)
| 1. | Ezgi Polat
|
| Industrial Engineering, Yıldız Technical University | |
| 2. | Mehmet Güray Güler
|
| Industrial Engineering Dep., İstanbul Technical University | |
| 3. | Mehmet Yasin Ulukus
|
| Industrial Engineering, Istanbul Technical University |
Abstract
In electricity markets, participants employ diverse strategies when submitting their bids. From a realistic perspective, it can be considered that participants aim to increase profitability and thus maintain their market presence. However, unlike profit maximization, participants' bidding behaviors are shaped by considerations like available capacity, market prices, market risk, etc. This research aims to investigate the bidding behaviors of electricity market participants using Inverse Reinforcement Learning (IRL). IRL aims to identify an objective function (reward function) from the historical bidding behaviors of an optimally behaving market participant. In this study, different IRL structures are utilized for linear or nonlinear reward function assumptions. In addition, the bidding behaviors of power plants using different energy sources, such as wind, gas, coal, etc., are comparatively analyzed. The research relies on data sourced from the Australian electricity market.
Keywords
- Electricity Markets
- Machine Learning
Status: accepted
Back to the list of papers