Abnormal Electricity Price Warning Method for Electricity Spot Markets Based on Non-mechanistic Modeling

SONG Jifeng; ZHENG Wenbin; CHEN Mingyuan; QI Le; ZOU Qi

doi:10.19725/j.cnki.1007-2322.2024.0194

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Abnormal Electricity Price Warning Method for Electricity Spot Markets Based on Non-mechanistic Modeling

更新时间：2026-10-19

- Abnormal Electricity Price Warning Method for Electricity Spot Markets Based on Non-mechanistic Modeling
- Vol. 43, Issue 5, Pages: 952-960(2026)
- 作者机构：
  
  广西电网电力调度控制中心, 广西壮族自治区,南宁市,530000
- 作者简介：
- 基金信息：
- DOI：10.19725/j.cnki.1007-2322.2024.0194
  CLC： 5
- Published：2026
- 稿件说明：
移动端阅览
SONG Jifeng, ZHENG Wenbin, CHEN Mingyuan, et al. Abnormal Electricity Price Warning Method for Electricity Spot Markets Based on Non-mechanistic Modeling[J]. 2026, 43(5): 952-960.
DOI：

SONG Jifeng, ZHENG Wenbin, CHEN Mingyuan, et al. Abnormal Electricity Price Warning Method for Electricity Spot Markets Based on Non-mechanistic Modeling[J]. 2026, 43(5): 952-960. DOI： 10.19725/j.cnki.1007-2322.2024.0194.

摘要

大量新型市场主体在电力现货市场中兴起并享有一定的报价权，大幅增加了出清电价的灵活性、波动性、复杂性，提高了出现异常电价的风险。为保障电力现货市场安全稳定运营，有必要提出异常电价预警方法，提前防范和降低异常电价风险。首先，分析出清电价的异常表现与异常来源，从价格水平、价格波动、市场结构、成员行为、网络运行、系统管理、外部因素等7个方面构建出清电价状态预评估指标体系。其次，基于出清价格状态预评估指标体系，提出线性判别分析法(linear discriminant analysis

LDA)降维的BP神经网络预警方法和局部切空间排列法(local tangent space alignment

LTSA)降维的BP神经网络预警方法等两种不同的预警方法，在数据降维的基础上应用BP神经网络判断后续电价异常的可能性。最后，将所提出的异常电价预警方法应用于某地区电力现货市场的实际运营，结果显示，两种预警方法在异常电价预警中均有效，降维方法在给定的精度损失下能显著提高计算效率。

Abstract

A large number of new market participants have emerged in the electricity spot market and have been granted certain pricing rights. This development significantly improves the flexibility

volatility

and complexity of the clearing price

thereby elevating the risk of abnormal prices. To ensure the safe and stable operation of the electricity spot market

it is necessary to develop an abnormal price warning method to mitigate potential abnormal price risks in advance. Firstly

the abnormal performance and sources of the clearing price are analyzed. A pre-assessment indicator system for the state of the clearing price evaluation is constructed from seven aspects: price level

price volatility

market structure

participant behavior

network operation

system management

and external factors. Secondly

based on the pre-assessment indicator system

two different warning methods are proposed: a BP neural network warning method with dimensionality reduction using the linear discriminant analysis method and a BP neural network warning method with dimensionality reduction using the local tangent space alignment method. These methods apply BP neural networks to assess the possibility of subsequent price abnormalities after dimensionality reduction of the data. Finally

the proposed abnormal price warning methods are applied to the actual operational data of an electricity spot market in a certain region. The results indicate that both warning methods are effective for abnormal price warnings

and the dimensionality reduction methods significantly improve computational efficiency within a given accuracy loss.

关键词

Keywords

references

BAO Minglei, DING Yi, ZHOU Xiaoxin, et al . Risk assessment and management of electricity markets: A review with suggestions[J ] . CSEE Journal of Power and Energy Systems, 2021, 7(6): 1322−1333.

ZHANG Zhixiang, JIANG Long, YAN Xu, et al . Research on comprehensive evaluation of electricity market risk based on subjective and objective weighting[J ] . Southern Energy Construction, 2023, 10(06): 160 −169(in Chinese).

XIE Jingdong, LU Chixin, LU Siwei, et al . Electricity market risk evaluation based on order relation-entropy weight method[J ] . Electric Power, 2021, 54(06): 71−78(in Chinese).

刘沆. 气电耦合虚拟电厂运营优化及风险评价模型研究[D]. 北京: 华北电力大学.

WANG Qin, WEN Fushuan, LIU Min, et al . Combined use of fuzzy set theory and analytic hierarchy process for comprehensive assessment of electricity markets[J ] . Automation of Electric Power Systems, 2009, 33(07): 32−37(in Chinese).

WU Zhongqun, ZHENG Ruijin, XU Feiyang, et al . Research on weather derivatives-based electricity price risk management under extreme weather scenarios[J ] . Journal of Global Energy Interconnection, 2024, 7(01): 66−78(in Chinese).

HE Jie, JIN Luosong, ZHAO Wen, et al . Market price risk analysis under background of renewable energy assumption guarantee[J ] . Modern Electric Power, 2022, 39(06): 631−639(in Chinese).

ZHENG Wei, WANG Xuanding, LIANG Zhiyuan, et al . Renewable energy electricity market risk estimating model based on Vine-Copula and Quasi-Monte Carlo method[J ] . Re newable Energy Resources, 2023, 41(12): 1642−1649(in Chinese).

ZHANG Zhengli, BI Zhaodong, LIU Meng, et al . Research on protection setting risk assessment system in power spot market[J ] . Electric Engineering, 2023(14): 132−136(in Chinese).

LUAN Jingjing, XIE Jingdong. Credit risk assessment of power market entities based on combined weighting-cloud model[J]. Power Energy, 2024, 45(03): 292−299(in Chinese).

HE Xiqing, ZHAO Ermin, WANG Feng, et al . Research on credit risk management system for electricity retail market[J ] . Distribution Utilization, 2024, 41(02): 81−88(in Chinese).

ZHENG Wei, TIAN Lin, HUANG Yuanming, et al . Improved COPOD identification method for performance risk in electricity market considering flexible regulatory strategy[J ] . Price: Theory Practice, 2023, (05): 59−64(in Chinese).

白杨, 王锦云, 李瑞. 基于电网企业视角的电力市场交易风险防控策略研究[C]//中国电力企业管理创新实践(2022). 云南电网有限责任公司, 2024.

LI Zhifeng, SHI Hang, WU Jingying, et al . Risk identification of electricity market operation management based on fuzzy-delphi and HFLTS-AHP methods[ J ] . Control Engineering of China, 2019, 26(05): 1003−1010(in Chinese).

ZHAO Yue, BAI Yang, LIU Sijie, et al . Research on dynamic rate risk assessment in my country's power market construction[J ] . Price: Theory Practice, 2021, (11): 167−172(in Chinese).

徐可琪. 高比例新能源下的市场电价预测及风险评估与分析[D]. 北京: 华北电力大学, 2023.

LI Zhengxin, ZHAO Lindu. Study of price spikes in power market[J]. Relay, 2008(01): 52−56(in Chinese).

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Short-term Load Forecasting Based on SOM-BP and Prophet Hybrid Model

Two-stage Market Bidding Model of Virtual Power Plant Considering Demand Response

An Improved Power Load Forecasting Method Based on Transformer

Short-Term Photovoltaic Power Prediction Based on Back Propagation Neural Network Improved by Hybrid Sparrow Algorithm

Discussion on Key Issues of Renewable Energy Participation in the Electricity Spot Market and Its Application on A Case Study

Related Author

Yuhang CHEN

Yuhong WANG

Lu NAN

Chuan HE

Tengxin WANG

Min ZHANG

Dongmei ZHAO

Yucun GUO

Related Institution

State Grid Shanxi Electric Power Research Institute

College of Electrical Engineering, Sichuan University

School of Electrical and Electronic Engineering, North China Electric Power University,Changping District

College of Computer Science, Sichuan University

Aostar Information Technologies Co., Ltd.

AI问答

⁰