摘要
在电力现货市场建设的初期,识别机组串谋竞价是有效防范发电商串谋行为真正发生的重要环节。依据市场份额将机组分为卡特尔类机组和竞争类机组,针对卡特尔类机组提出了基于排序多元Logit模型的串谋竞价识别方法。首先,依托博弈论原理,分析影响机组报价的因素,构建机组报价回归方程。然后,通过邹检验判断报价影响因素对卡特尔类机组和竞争类机组报价行为的影响程度是否存在差异。若存在差异,则建立卡特尔类机组报价的排序多元Logit模型,并利用似然比检验判断卡特尔类机组的高价序列和低价序列之间是否存在差异,论证卡特尔类机组串谋竞价的可能性。最后,将所提方法应用于某地区电力现货市场中,算例结果显示在5%的显著性水平下,卡特尔类机组存在串谋竞价的可能性。
In the early stage of the construction of electricity spot markets, identifying the collusion bidding of generators is an important link to effectively prevent the actual occurrence of collusion behaviors of power producers. Based on the market sharing,the generators are classified into Cartel-type generators and competitive-type generators. For the Cartel-type generators, a method of collusion bidding based on ordered Logit model is proposed. Firstly, based on the game theory, the factors affecting the bidding of generators are analyzed, and the bidding regression equation of generators is constructed. Then, the Chow test is used to determine whether there is a difference in the degree of the influence of the bidding influence factors on the bidding behaviors of Cartel-type generators and competitive-type generators. If there is a difference, the ordered Logit model of Cartel-type generators is established, and the likelihood ratio test is used to determine whether there is a difference between the high-priced sequence and the low-price sequence of the Cartel-type generators, thereby demonstrating the possibility of collusion bidding of Cartel-type generators. Finally, the proposed method is applied to an electricity spot market in a region, and the case study results show that at a significance level of 5%, which means a possibility of collusion bidding of Cartel-type generators.
作者
孙波
邓瑞林
谢敬东
孙欣
SUN Bo;DENG Ruilin;XIE Jingdong;SUN Xin(School of Economics and Management,Shanghai University of Electric Power,Shanghai 200090,China)
出处
《电力系统自动化》
EI
CSCD
北大核心
2021年第6期109-115,共7页
Automation of Electric Power Systems
基金
国家自然科学基金资助项目(71972127)
教育部人文社会科学研究基金资助项目(15YJCZH147)。
