< back 日本語版

Demand Responce Effectiveness Analysis Using Power Market Multi-Agent Simulation and Plant Power Consumption Model [in Japanese]

Ryo Wakasugi, Kiyoshi Izumi, Masanori HIRANO

The 35th Annual Conference of the Japanese Society for Artificial Intelligence, p. 2I3-GS-5b-04, June 9, 2021


Conference

The 35th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI2021)

Abstract

With the deregulation of electric power, businesses with large demand for electric power need to newly consider uncertainties such as fluctuations in market prices and supply and demand when procuring electric power. In this study, we analyzed the effect of demand response (DR) in the electricity market. In our experiment, first, the characteristics were extracted from the factory power consumption time series data by principal component analysis, and the power consumption model of factory was constructed. Next, a simulation experiment was conducted using a multi-agent model that imitated the day-ahead market of JEPX (Japan Electric Power Exchange) in which a power supply agent and a demand agent participate in addition to the factory agent. The scale of each factory DR scenario was changed and the effect of DR was analyzed using two evaluation indexes in terms of cost effectiveness. As a result, it was confirmed that the larger the factory scale, the greater the DR effect due to its own market impact. On the other hand, the DR effect may vary regardless of the size of the factory or the scale of the DR, suggesting that the factors of the DR effect may be complex.

Keywords

Multi agent simulation; Power market; Demand responce; Electricity data;

doi

10.11517/pjsai.JSAI2021.0_2I3GS5b04


bibtex

@inproceedings{Wakasugi2021-jsai35,
  title={{Demand Responce Effectiveness Analysis Using Power Market Multi-Agent Simulation and Plant Power Consumption Model [in Japanese]}},
  author={Ryo Wakasugi and Kiyoshi Izumi and Masanori HIRANO},
  booktitle={The 35th Annual Conference of the Japanese Society for Artificial Intelligence},
  pages={2I3-GS-5b-04},
  doi={10.11517/pjsai.JSAI2021.0_2I3GS5b04},
  year={2021}
}