电能质量复合扰动分类识别  

Classification of power quality complex disturbances

作　　者：占勇[1,2] 程浩忠[2] 

机构地区：[1]安徽省电力公司系统研究中心,安徽合肥230022 [2]上海交通大学电气工程系,上海200240

出　　处：《电力自动化设备》2009年第3期93-97,共5页Electric Power Automation Equipment

摘　　要：电能质量扰动的分类分为信号特征提取和分类器2个阶段,采用S变换和支持向量机构造电能质量复合扰动的分类识别方案。利用S变换进行扰动信号特征提取,构造支持向量机静态分类树,再通过基于Mercer核的聚类方法对静态分类树进行动态扩展,形成动态分类树,实现对复合扰动的识别。给出了电能质量复合扰动分类算法的4个步骤:构建静态分类树;用基于Mercer核的聚类方法进行聚类分析;构建动态分类树;对新发现的扰动确定其具体类型,并给其命名。算例表明该方法不仅可以有效分类识别电压突降、电压突升、电压中断、暂态振荡、电压尖峰、电压缺口和谐波等7种电能质量扰动,还可以识别由其组合而成的电能质量复合扰动。The classification of power quality disturbances has two sequential stages： signal feature extraction and classifier design. An approach based on SVM（Support Vector Machines） and S-transform to detect and classify power quality complex disturbances is presented. The features of disturbances are extracted by S-transform to construct the static SVM classification tree, which is then dynamically expanded based on Mercer Kernel clustering algorithm to classify power quality complex disturbances. Four steps of power quality complex disturbance classification are presented： static classification tree construction, clustering analysis, dynamic classification tree construction and new found complex disturbance definition. Example shows that, the proposed method effectively classifies seven disturbance patterns of sag, swell, interruption, oscillatory transient, spike, notch and harmonics, as well as the power quality complex disturbances.

关 键 词：电能质量 复合扰动 S变换 支持向量机 分类 特征提取 

分 类 号：TM712[电气工程—电力系统及自动化]