引用本文:刘挺坚,刘友波,刁塑,刘俊勇,韩安兵,夏志军.连锁故障中负荷损失数值特征的非参关联分析[J].电力自动化设备,2018,(7):
LIU Tingjian,LIU Youbo,DIAO Su,LIU Junyong,HAN Anbing,XIA Zhijun.Nonparametric correlation analysis of numerical feature for load loss in cascading failure[J].Electric Power Automation Equipment,2018,(7):
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 4491次   下载 1891  
连锁故障中负荷损失数值特征的非参关联分析
刘挺坚1, 刘友波1, 刁塑2, 刘俊勇1, 韩安兵2, 夏志军2
1.四川大学 电气信息学院,四川 成都 610065;2.国网泸州供电公司,四川 泸州 646000
摘要:
从大量连锁故障仿真数据中观察负荷损失与电网特性指标的数值特征关系,建立负荷损失与阶段性电网特性指标的非参关联分析模型。基于多层时序演化连锁故障模型构建数据仿真平台,选取电气状态指标、网络结构指标以及有序分布指标这3类电网特性指标作为连锁故障负荷损失的关联对象,通过仿真得到连锁故障负荷损失与电网特性指标的大数据样本。采用非参数独立性筛选方法剔除弱关联特性指标,进一步利用Group Lasso非参数回归算法确定关联特性指标与负荷损失的非参数关联规则。西南某省级电网算例表明,非参关联分析方法可揭示与连锁故障负荷损失具有强关联性的电网特性指标,同时显性地给出电网特性指标与负荷损失的关联水平和趋势。
关键词:  连锁故障  负荷损失  电网特性指标  关联性分析  非参数独立性筛选  Group Lasso非参数回归
DOI:10.16081/j.issn.1006-6047.2018.07.020
分类号:TM711
基金项目:国家自然科学基金资助项目(51437003,51207098)
Nonparametric correlation analysis of numerical feature for load loss in cascading failure
LIU Tingjian1, LIU Youbo1, DIAO Su2, LIU Junyong1, HAN Anbing2, XIA Zhijun2
1.College of Electrical Engineering and Information Technology, Sichuan University, Chengdu 610065, China;2.State Grid Luzhou Power Supply Company, Luzhou 646000, China
Abstract:
The numerical correlation between the load loss and power grid performance indexes is studied by nonparametric correlation analysis from the massive simulation data of cascading failures. The simulation platform is firstly built based on the multi-layer time-series evolution model of cascading failures. Three kinds of power grid performance indexes, i. e. electric state indexes, network structural indexes and entropy-based indexes, are then selected as the correlation objects of the load loss caused by cascading failures. The nonparametric independence screening is adopted to screen out the weakly coupled indexes, and Group Lasso nonparametric regression algorithm is further used to determine the nonparametric correlation rule of the load loss against the power grid performance indexes. A case study on a provincial power grid in southwest China demonstrates that the proposed method can identify the performance indexes that have strong correlations with load loss caused by cascading failures. Moreover, the nonparame-tric correlation rule explicitly reveals the correlation trend between the load loss and power grid performance indexes.
Key words:  cascading failure  load loss  performance index of power grid  correlation analysis  nonparametric independence screening  Group Lasso nonparametric regression61

用微信扫一扫

用微信扫一扫