半月刊

ISSN 1000-1026

CN 32-1180/TP

+高级检索 English
基于卷积神经网络的暂态电压稳定评估及风险量化
作者:
作者单位:

1.华南理工大学电力学院,广东省广州市 510640;2.国网江苏省电力有限公司检修分公司,江苏省南京市 211102;3.南方电网科学研究院有限责任公司,广东省广州市 510663;4.广东电网有限责任公司电力调度控制中心,广东省广州市 510660

摘要:

提出了一种暂态电压稳定性评估及其风险量化方法。首先,探讨卷积神经网络(CNN)与暂态电压稳定评估的关联性和匹配性,建立了基于CNN的暂态电压稳定评估模型。其次,在可信度框架下引入四元评估结构,可有效解决CNN在稳定边界识别上对时域仿真依赖的难题。然后,通过评估结果获取暂态电压稳定裕度,并将其与可信度相结合来构建风险函数,从而实现对暂态电压稳定的风险量化分级。实际电网算例分析结果验证了该方法的有效性。

关键词:

基金项目:

国家自然科学基金资助项目(U1766213);中国南方电网有限责任公司科技项目(GDKJXM20198236)。

通信作者:

作者简介:

陈达(1994—),男,硕士,主要研究方向:电力系统稳定与控制、直流输电技术。E-mail:scut_ep_chenda@163.com
朱林(1979—),男,通信作者,博士,副教授,主要研究方向:电力系统稳定与控制、直流输电技术以及新能源并网。E-mail:zhul@scut.edu.cn
张健(1995—),男,硕士研究生,主要研究方向:电力系统稳定与控制。E-mail:519765449@qq.com


Transient Voltage Stability Assessment and Risk Quantification Based on Convolutional Neural Network
Author:
Affiliation:

1.School of Electric Power, South China University of Technology, Guangzhou 510640, China;2.Maintenance Branch Company of State Grid Jiangsu Electric Power Co., Ltd., Nanjing 211102, China;3.Electric Power Research Institute of China Southern Power Grid Co., Ltd., Guangzhou 510663, China;4.Power Dispatching Control Center of Guangdong Power Grid Co., Ltd., Guangzhou 510660, China

Abstract:

A method is proposed for assessing transient voltage stability and quantifying its risk. Firstly, the correlation and matching between the convolutional neural network (CNN) and transient voltage stability assessment are discussed and the CNN based assessment model for the transient voltage stability is built. Secondly, the quaternary assessment structure is introduced in the credibility framework, which can effectively solve the dependence of CNN on the time-domain simulation in the stable boundary recognition. Then, the transient voltage stability margin is obtained through the assessment results. Combined with the credibility, a risk function is constructed to realize the quantitative classification of the risk of transient voltage stability. Validity of the method is verified by the analysis results of an actual power grid example.

Keywords:

Foundation:
This work is supported by National Natural Science Foundation of China (No. U1766213) and China Southern Power Grid Company Limited (No. GDKJXM20198236).
引用本文
[1]陈达,朱林,张健,等.基于卷积神经网络的暂态电压稳定评估及风险量化[J].电力系统自动化,2021,45(14):65-71. DOI:10.7500/AEPS20200615002.
CHEN Da, ZHU Lin, ZHANG Jian, et al. Transient Voltage Stability Assessment and Risk Quantification Based on Convolutional Neural Network[J]. Automation of Electric Power Systems, 2021, 45(14):65-71. DOI:10.7500/AEPS20200615002.
复制
分享
历史
  • 收稿日期:2020-06-15
  • 最后修改日期:2021-01-15
  • 录用日期:
  • 在线发布日期: 2021-07-21
  • 出版日期:
相关附件