Open Access

ARTICLE

Detecting Market Power Risk via Progressive Bilateral Matching in Electricity Markets

Siting Dai¹, Wenyang Deng^2,*, Dongliang Xiao², Haiqing Cai³
1 International Business School, Guangzhou City University of Technology, Guangzhou, China
2 School of Automation, Guangdong University of Technology, Guangzhou, China
3 CSG Electric Power Research Institute Co., Ltd., Guangzhou, China
* Corresponding Author: Wenyang Deng. Email:
(This article belongs to the Special Issue: Grid Integration of Intermittent Renewable Energy Resources: Technologies, Policies, and Operational Strategies)

Energy Engineering https://doi.org/10.32604/ee.2026.080534

Received 11 February 2026; Accepted 20 April 2026; Published online 27 May 2026

Abstract

With the increasing complexity of bilateral transactions in liberalized electricity markets, vertical collusion between generators and retailers can distort price signals and undermine market fairness, while remaining difficult to identify using single-stage or single-feature screening. This paper proposes a progressive bilateral matching framework for market power risk detection that links transaction-stage behavioral preference alignment with clearing-stage outcome concentration verification to form a closed-loop identification mechanism. Specifically, a two-stage model is developed by integrating (i) a Borda count-based preference ranking module to quantify multidimensional transaction behavior and produce an initial high-risk matching set, and (ii) an entropy-balancing-based concentration assessment module to verify abnormal clearing concentration patterns and suppress spurious behavioral flags. The proposed framework is validated using real centralized trading data from a provincial electricity market, involving five generation firms and eight retail firms. Empirical results show that jointly leveraging behavioral and structural signals improves interpretability and reduces false-positive alarms compared with behavior-only screening, while accurately isolating trading pairs exhibiting both high preference compatibility and abnormal concentration characteristics. The proposed approach provides regulators with a transparent and operational tool for preemptive collusion surveillance and targeted investigation in complex electricity trading environments.

---

Keywords

Market power risk; collusion detection; borda count; entropy balancing; electricity market regulation

---