摘要
电力系统的高维性是理论研究和工程实践中遇到的重要难题,特征选择是属性空间降维的有效手段,因而研究电力系统特征选择有着重要的理论价值和实际意义。为此提出了特征选择的混合互信息法(HMI)。HMI既考虑了候选属性所能提供的关于目标属性的新信息量,又考虑了候选属性与目标属性之间的相关性。文中针对静态电压稳定评估问题,将混合互信息法和其他几种基于互信息的特征选择方法进行了比较,实验结果表明:混合互信息法是一种有效的特征选择手段;数据挖掘技术在电力系统是有应用潜力的。
High dimensionality is one of the most troublesome difficulties encountered in power system analysis and application, Feature selection is an effective dimension reduction method in data mining field. This paper proposes a new feature selection method, called HMI （Hybrid Mutual Information）, which considers both the novel information a feature can provide and the relevance of the feature with respect to the target feature. In static voltage stability analysis application, HMI and other mutual information based feature selection methods are compared, the results show that HMI is an effective feature selection method and also show the potential application of data mining methods in power system.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2006年第7期77-81,共5页
Proceedings of the CSEE
基金
国家自然科学基金项目(50107005
50595414)
国家重点基础研究发展规划项目(2004CB217904)
新世纪优秀人才支持计划项目。~~
关键词
数据挖掘
特征选择
信息理论
互信息
静态电压稳定
电力系统
data mining
feature selection
information theory
mutual information
static voltage stability assessment
power system
