主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2023, Vol. 31 ›› Issue (10): 30-39.doi: 10.16381/j.cnki.issn1003-207x.2020.2071

• • 上一篇    下一篇

电力市场竞价上网的拍卖机制设计

宫汝凯()   

  1. 上海财经大学数字经济系,上海 200433
  • 收稿日期:2020-11-06 修回日期:2021-05-28 出版日期:2023-10-15 发布日期:2023-11-03
  • 通讯作者: 宫汝凯 E-mail:gong.rukai@dhu.edu.cn
  • 基金资助:
    国家自然科学基金资助面上项目(71873028);国家自然科学基金重点资助项目(71832001);上海市浦江人才计划(21PJC001)

Market Design of Electricity Bidding Trading

Ru-kai GONG()   

  1. Department of Digital Economics,Shanghai University of Finance and Economics,Shanghai 200433,China
  • Received:2020-11-06 Revised:2021-05-28 Online:2023-10-15 Published:2023-11-03
  • Contact: Ru-kai GONG E-mail:gong.rukai@dhu.edu.cn

摘要:

竞价上网是电力行业实现市场化的基础条件,竞价机制设计是深化市场化改革亟需探讨的议题。基于离散多物品逆向拍卖的理论框架对电力市场竞价上网的交易机制进行建模,在考虑需求状态、市场结构和报价区间等现实情形下刻画出统一价格(UPA)和差别价格(DPA)两种常用竞价机制下发电厂商的均衡报价策略,并从期望收益和生产效率两个视角进行机制比较。结果表明:(1)在低需求状态下,两种机制的市场均衡价格均为未被调用的最高效率发电厂商(机组)的边际成本;在高需求状态下,UPA存在多重纯策略均衡,DPA出现混合策略均衡。(2)在需求确定(较短报价区间)的情形下,DPA比UPA更可能形成更低的均衡价格;在需求不确定(较长报价区间)的情形下,对称发电厂商在两种机制下具有相同的期望收益。(3)两种机制在生产效率方面的比较是不确定的,具体取决于模型参数和UPA多重均衡的选择。本文的研究结论为完善电力竞价上网交易机制和深化电力体制改革提供理论基础和决策借鉴。

关键词: 竞价上网, 拍卖机制设计, 统一价格, 差别价格

Abstract:

“Electricity bidding on the grid” is the basic condition for the electricity industry to achieve market-oriented operation, and the bidding mechanism design is an important topic that needs to be discussed in deepening the reform of the electricity market. The existing theory and practice show that uniform price auction (UPA) and discriminatory price auction (DPA) are the two main bidding mechanisms usually used in electricity market transactions. Then, based on the commodity attributes and industry characteristics of electricity, a discrete multi-item auction model framework is adopted to model the transaction mechanism of electricity bidding on the grid.First, a benchmark model of duopoly is established, considering that there are two power generation companies with asymmetric production capacity and marginal cost “single-goods” quotation function. The equilibrium bidding behavior of power generation companies is depicted under the UPA and DPA mechanisms, and they are compared from the expected revenue and production efficiency. Then, considering more realistic situations, the model is expanded from aspects of multivariate oligopoly, multi-unit quotation, more flexible demand function and longer quotation interval, and the following conclusions are obtained:(i) The equilibrium quotation of power generation manufacturers depends on the state of demand: In the state of low demand, the market equilibrium price under the two mechanisms does not exceed the marginal cost of the maximum efficiency generator that has not been invoked; in the state of high demand, there are multiple pure strategy equilibriums under the UPA mechanism; and a mixed strategy equilibrium under DPA mechanism. (ii) The comparison of the expected revenue is affected by the quotation interval: when the demand is determined (the quotation period is short), compared with the UPA mechanism, the power generation company obtains a lower revenue under the DPA; Under the situation of uncertain demand (long quotation period), symmetrical power generation companies face the same expected return under the two mechanisms. (iii) The comparison of the UPA and DPA mechanisms in terms of production efficiency is not clear, which depends on the model parameters and the choice of multiple pure strategy equilibrium under the UPA mechanism.It attempts to use the auction theory framework to model the power market bidding online. By comparing the market performance of the two bidding mechanisms, the boundary conditions for mechanism design and selection are obtained. It is hoped that these conclusions will provide a theoretical basis and decision-making reference for improving the design of the power market bidding online transaction mechanism and deepening the power system reform in China.

Key words: bidding trading, auction design, uniform price, discriminatory price

中图分类号: