A Deep Reinforcement Learning Bidding Algorithm on Electricity Market
- PDF / 737,966 Bytes
- 10 Pages / 595.22 x 842 pts (A4) Page_size
- 16 Downloads / 243 Views
https://doi.org/10.1007/s11630-020-1308-0
Article ID: 1003-2169(2020)00-0000-00
A Deep Reinforcement Learning Bidding Algorithm on Electricity Market JIA Shuai1*, GAN Zhongxue2*, XI Yugeng1, LI Dewei1, XUE Shibei1, WANG Limin3 1. Department of Automation, Key Laboratory of System Control and Information Processing, Shanghai Jiao Tong University, Shanghai 200240, China 2. State Key Laboratory of Coal-based Low-carbon Energy, ENN Science and Technology Development Co. Ltd., Langfang 065001, China 3. ENN Energy Power Technology (Shanghai) Co. Ltd., Shanghai 201306, China © Science Press, Institute of Engineering Thermophysics, CAS and Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract: In this paper, we design a new bidding algorithm by employing a deep reinforcement learning approach. Firms use the proposed algorithm to estimate conjectural variation of the other firms and then employ this variable to generate the optimal bidding strategy so as to pursue maximal profits. With this algorithm, electricity generation firms can improve the accuracy of conjectural variations of competitors by dynamically learning in an electricity market with incomplete information. Electricity market will reach an equilibrium point when electricity firms adopt the proposed bidding algorithm for a repeated game of power trading. The simulation examples illustrate the overall energy efficiency of power network will increase by 9.90% as the market clearing price decreasing when all companies use the algorithm. The simulation examples also show that the power demand elasticity has a positive effect on the convergence of learning process.
Keywords: electricity market, reinforcement learning, energy efficiency, conjectural variation, bidding strategy
1. Introduction An electricity market refers to competitive market, which enables purchases bidding to buy, sales offering to sell and short-term trades, generally in the form of financial or obligation swaps. Power producers and users trade electricity through negotiation and bidding, and set the price and quantity through market competition using supply and demand principles. In electricity market, each electricity firm will adopt optimal bidding strategy to maximum its profit. Therefore, it is worth studying how power firms should construct bidding strategies. A class of competitive electricity market models can be divided into price competition models, production Received: Apr 02, 2019
competition models and supply function competition models according to different competitive variables [1−4]. Firms use strategic competition in pursuit of maximizing their own profits. It is well known that due to the market mechanism and asymmetric information, firms often ignore the changes of the market price (no response or delay even there is a response), resulting in the inelastic demand of electricity in the electricity market features. Therefore, using the price competition model to analyze the strategic behavior of power generation companies is far from the actual market operation.
Data Loading...
