结合噪声滤波的电力系统暂态稳定预测  

Transient Stability Prediction of Power System Combined withNoise Filtering

作　　者：刘颂凯 党喜 陈萍 周倩[1,2] 杨超 王秋杰[1,2] 张雅婷 LIU Songkai;DANG Xi;CHEN Ping;ZHOU Qian;YANG Chao;WANG Qiujie;ZHANG Yating(College of Electrical Engineering andNew Energy,China ThreeGorgesUniversity,Yichang 443002,China;Hubei Provincial Collaborative Innovation CenterforNew Energy Microgrid,Yichang 443002,China;StateGrid Taiyuan Power Supply Company,Taiyuan 030000,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002 [2]新能源微电网湖北省协同创新中心,湖北宜昌443002 [3]国网太原供电公司,山西太原030000

出　　处：《智慧电力》2024年第7期32-39,共8页Smart Power

基　　金：国家自然科学基金资助项目(52307109);湖北省自然科学基金项目(2022CFB825)。

摘　　要：为了解决电力系统实际运行过程中存在大量的输出噪声,提出了一种结合噪声滤波的电力系统暂态稳定预测框架。本框架基于最优样本选择(OSS)方法和覆盖距离(CD)结合生成的覆盖距离滤波器(CDF)进行噪声滤波,然后使用轻量级梯度提升机(LightGBM)模型进行电力系统暂态稳定预测。在仿真软件提供的23节点系统和美国南卡罗莱纳州500节点实际系统上进行的测试结果表明,所提出的电力系统暂态稳定预测框架能够有效过滤噪声,并具有较强的鲁棒性。In order to solve the problem that there is a lot of output noise in actual operation of power system,a transient stability prediction framework for power systems is proposed combined with noise filtering.This framework is based on the generated Covering Distance Filter(CDF)which is combined with the Optimal Sample Selection(OSS)method and Covering Distance(CD)for noise filtering.Then the Light Gradient Boosting Machine(LightGBM)model is used for power system transient stability prediction.The test are conducted on 23-bus system provided by the simulation software and 500-bus real grid in South Carolina,USA.It is shown that the proposed power system transient stability prediction framework can effectively filter noise and has strong robustness.

关 键 词：噪声滤波 LightGBM 暂态稳定 电力系统 数据驱动 

分 类 号：TM716[电气工程—电力系统及自动化]