电网技术Issue(4)：1026-1032,7.DOI:10.13335/j.1000-3673.pst.2015.04.023
基于数据挖掘的区域暂态电压稳定评估
Data Mining Based Regional Transient Voltage Stability Assessment
摘要
Abstract
In allusion to the imperfection of the theory related to regional transient voltage stability assessment and insufficient reliability of engineering criteria, a data mining based method to assess regional transient voltage stability is proposed and a two-layer assessment framework, in which the interaction between single bus load stability and multi bus voltages is synthetically considered, is constructed. Utilizing the measure index of nodal stability and the identification based voltage/reactive power sensitivity matrix the original features of the power network is extracted. To cope with the difficult problem that there is not yet reliable standard to delimitate the regional transient voltage instability, a constraint based semi-supervised learning method is used to reliably classify and label the data sets. Based on decision tree algorithm a step-by-step updated classification model is established to generate the criterion of regional transient voltage stability, and by use of this classification model the inherent law related to the voltage partition and representative buses can be mined. The validity of the proposed assessment scheme as well as the adaptability and the accuracy of the classification model are verified by simulation results of EPRI 36-bus system.关键词
区域暂态电压稳定评估/数据挖掘/灵敏度辨识/半监督学习/决策树Key words
regional transient voltage stability assessment/data mining/sensitivity identification/semi-supervised learning/decision tree分类
信息技术与安全科学引用本文复制引用
朱利鹏,陆超,孙元章,黄河,苏寅生,李智欢..基于数据挖掘的区域暂态电压稳定评估[J].电网技术,2015,(4):1026-1032,7.基金项目
国家重点基础研究发展计划资助项目(2012CB215206)；国家自然科学基金资助项目(51037002，51107061)。Project Supported by the National Basic Research Program of China (2012CB215206) and National Natural Science Foundation of China (51037002，51107061) （2012CB215206）
