基于混合监督学习的极端天气电力系统故障精准预警

王运; 蒙飞; 常鹏; 杨宏; 杨波; 乔咏田

doi:10.19725/j.cnki.1007-2322.2024.0054

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于混合监督学习的极端天气电力系统故障精准预警

智能电网 | 更新时间：2026-06-02

- 基于混合监督学习的极端天气电力系统故障精准预警
- Precise Early Warning of Extreme Weather-induced Power System Faults Based on Hybrid Supervised Learning
- 现代电力 2026年43卷第3期页码：444-454
- 作者机构：
  
  国网宁夏电力有限公司, 宁夏回族自治区,银川市,750001
- 作者简介：
  
  王运(1986)，男，学士，高级工程师，研究方向为电力调控运行，E-mail：275270864@qq.com
  蒙飞(1987)，男，学士，高级工程师，研究方向为电网调控运行，E-mail：474504618@qq.com
  常鹏(1986)，男，硕士研究生，高级工程师，研究方向为电网调控运行，E-mail：68344884@qq.com
  杨宏(1987)，男，硕士研究生，高级工程师，研究方向为电网调控运行，E-mail：421067115@qq.com
  杨波(1989)，男，学士，高级工程师，研究方向为电网调控运行，E-mail：295963198@qq.com
  乔咏田(1990)，男，硕士研究生，高级工程师，研究方向为电网调控人工智能，E-mail：759729997@qq.com
- 基金信息：
  
  国网宁夏电力有限公司科技项目(5229NX220027)。
- DOI：10.19725/j.cnki.1007-2322.2024.0054
  中图分类号： 3
- 纸质出版：2026
- 稿件说明：
移动端阅览
王运, 蒙飞, 常鹏, 等. 基于混合监督学习的极端天气电力系统故障精准预警[J]. 现代电力, 2026,43(3):444-454.

WANG Yun, MENG Fei, CHANG Peng, et al. Precise Early Warning of Extreme Weather-induced Power System Faults Based on Hybrid Supervised Learning[J]. 2026, 43(3): 444-454.
王运, 蒙飞, 常鹏, 等. 基于混合监督学习的极端天气电力系统故障精准预警[J]. 现代电力, 2026,43(3):444-454. DOI： 10.19725/j.cnki.1007-2322.2024.0054.

WANG Yun, MENG Fei, CHANG Peng, et al. Precise Early Warning of Extreme Weather-induced Power System Faults Based on Hybrid Supervised Learning[J]. 2026, 43(3): 444-454. DOI： 10.19725/j.cnki.1007-2322.2024.0054.

摘要

针对当前电力系统故障风险预警方法难以将耦合情况复杂的气象要素和故障精准映射，无法实现多类型极端天气下的多维气象特征提取、故障预警精度低的问题，提出结合有监督学习和无监督学习的极端天气故障精准预警方法。首先，分析极端天气场景故障时刻的气象要素和故障的关联度，利用层次分析法为气象要素赋权，实现气象要素和故障之间的风险自适应赋权。其次，构建无监督的变分自编码训练网络对数据降维，优化风险权重特征，以精准表征气象要素和故障的映射关系。然后，在无监督学习网络解码器后嵌入基于随机森林算法的有监督网络，用以强化风险权重特征和电力故障的相关性。最后，利用支持向量机构建预警模型，将风险权重特征、设备数据和环境数据作为输入，实现电力系统的故障预警。通过算例分析可知，故障预警准确度达到87.13%，可在事故发生前15.19 min发出预警。

Abstract

Aiming to address the challenges of the current power system fault risk warning methods in accurately mapping the meteorological elements and faults with complex coupling and its incapability in realizing multi-dimensional meteorological feature extraction under multiple types of extreme weather as well as the problem of low fault warning accuracy

in this paper we propose a combination of supervised and unsupervised learning as an accurate warning method for extreme weather-induced faults. First

the meteorological elements and fault correlations at the moment of failure in extreme weather scenarios are analyzed

and the hierarchical analysis method is utilized to assign weights to the meteorological elements

so as to realize the risk-adaptive assignments between the meteorological elements and the faults. Subsequently

a supervised network based on the random forest algorithm is embedded after the decoder of the unsupervised learning network

aiming to enhance the correlation between the risk weight features and power faults. Finally

the support vector mechanism is employed to construct an early warning model. The risk weight features

equipment data and environmental data are utilized as inputs to realize the fault early warning of the power system. Experimental analysis demonstrates that the fault warning accuracy reaches 87.13%

with the warning system capable of providing alerts 15.19 minutes prior to the occurrence of incidents

关键词

Keywords

references

WU Chen, SU Mingxin, XIE Yunyun, et al . BP neutral network based lightning fault prediction of transmission lines[J ] . Electric Power Engineering Technology, 2020, 39(5): 133−139(in Chinese).

YU Xiaoxiao, SONG Fulong, LI Juan, et al . Power supply security improvement of power grid wit h high proportion of renewable energy under extreme weather events[J ] . Modern Electric Power, 2023, 40(3): 303−313(in Chinese).

WANG Jian, YAO Jiangning, LIU Zeqing, et al . Fault statistical analysis and probability distribution fitting for a power distribution network in adverse weather condition[J ] . Power System Protection and Control, 2022, 50(17): 143−153(in Chinese).

LI Zhenkun, CHEN He, LI Xincong, et al . Emergency maintenance and recovery strategy for failures in regional internet of energy in extreme weather[J ] . Modern Electric Power, 2024, 41(6): 1109−1118(in Chinese).

YANG B, GAO X, HAN Y, et al . A thunderstorm identification method combining the area of graupel distribution region and weather radar reflectivity[J ] . Earth and Space Science, 2020, 7(1): 1−10.

HUANG Wenxin, WU Jun, GUO Zihui, et al . Grid resilience assessment and differentiated planning against extreme typhoon disasters[J ] . Automation of Electric Power Systems, 2023, 47(05): 84−91(in Chinese).

MOON S H , KIM Y H. Forecasting lightning around the Korean Peninsula by postprocessing ECMWF data using SVMs and undersampling[J]. Atmospheric Research, 2020, 243: 105026.

WANG Xiaoxue, CHI Linlin, CHEN Haiwen. A distribution network resilience enhancement method considering mobile energy storage under typhoon storm[J]. Proceedings of the CSU-EPSA, 2024, 36(5): 69−79, 89(in Chinese).

孙青松. 计及线损的电力系统抗冰与融冰策略研究[D]. 重庆: 重庆大学, 2022.

刘晓楠. 台风灾害影响下面向规划的输电网韧性评估方法研究[D]. 天津: 天津大学, 2021.

LI Jiayu, QIN Wenping, JING Xiang, et al . Power system resilience assessment considering cascading failure under typhoon disaster[J ] . Proceedings of the CSU-EPSA, 2023, 35(01): 14−22(in Chinese).

BAO R, ZHANG Y, MA B J, et al . An artificial neural network for lightning prediction based on atmospheric electric field observations[J ] . Remote Sensing, 2022, 14(17): 1−27.

WANG Gouming, KIM W H, KIL G S, et al . An intelligent lightning warning system based on electromagnetic field and neural network[J ] . Energies, 2019, 12(7): 1−11.

SUN Liqiang, LI Peng, LIU Xuanting, et al . Analysis model of transmission line under ice c oating disaster considering cascading fault[J ] . High Voltage Apparatus, 2022, 58(01): 155−161+169(in Chinese).

XIANG Shaobin, TU Shuiyuan. Distribution network fault distance estimation method based on isolated forest algorithm and big data mining[J]. Journal of Electrical Engineering, 2022, 17(01): 179−185(in Chinese).

REN Yirui, LIN Jikeng, NING Jian, et al . A comprehensive evaluation strategy of a power system fault diagnosis method based on dominance and the Choquet integral[J ] . Power System Protection and Control, 2021, 49(06): 145−154(in Chinese).

ZHANG Q, ZHANG X, POLYCAIPOU M M, et al . Distributed sensor fault detection and isolation for multimachine power systems[J ] . International Journal of Robust Nonlinear Control, 2014, 24(8-9): 1403−1430.

HU Wei, ZHENG Le, MIN Yong, et al . Research on power system transient stability assessment based on deep learning of big data technique[J ] . Power System Technology, 2017, 41(10): 3140−3146(in Chinese).

ZHANG Y, YANG S, LU W, et al . Experiments of artificially triggered lightning and its application in Conghua, Guangdong, China[J ] . Atmospheric Research, 2014, 135: 330−343.

JIAO Jianing, LIU Lu, ZHANG Tianyu, et al . Resilience enhancement strategy with multi-stage collaboration for receiving-end grid under typhoon disaster[J ] . Automation of Electric Power Systems, 2023, 47(12): 9−18(in Chinese).

浏览量

181

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于电压序列相似性的户变关系与相别识别

计及季节特性与日特性的多源–荷联合场景生成方法研究

基于分布式聚类模型的电力负荷特性分析

基于深度学习的电网短期负荷预测方法研究

基于混合神经网络和遗传算法的故障诊断系统