Conjectural variation based learning of generator´s behavior in electricity market

Author

Song, Yiqun ; Hou, Zhijian ; Wen, Fushuan ; Ni, Yixin ; Wu, Felix F.

Author_Institution

Dept. of Electr. Eng., Shanghai Jiao Tong Univ., China

Volume

3

fYear

2003

fDate

25-28 May 2003

Abstract

In this paper, a conjectural variation based learning method is proposed for generation firms to improve their strategic bidding performance in a spot electricity market taking account of the expected reaction of their rivals. With the application of conjecture, each firm can make its optimal generation decision in the learning process according to available information published in the electricity market. Examples are used to illustrate that motivation is existed for each firm to start learning, and learning of all firms will decrease the market clearing price of electricity and improve the total social welfare.

Keywords

learning (artificial intelligence); modelling; power generation economics; power markets; conjectural variation based learning method; generation firms; market clearing price; optimal generation decision; spot electricity market; strategic bidding performance; Consumer electronics; Economic forecasting; Electricity supply industry; Flexible manufacturing systems; Learning systems; Power generation; Power generation economics; Power industry; Power system economics; Predictive models;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 2003. ISCAS '03. Proceedings of the 2003 International Symposium on

Print_ISBN

0-7803-7761-3

Type

conf

DOI

10.1109/ISCAS.2003.1205041

Filename

1205041