摘要
S变换具有很好的时频分析能力,能精确提取突变信号的关键特征信息。在分析S变换原理基础上,提出一种基于多分类支持向量机的电压暂降源识别方法。用S变换对电压暂降信号进行时频分析,提取各类暂降特征;用多分类支持向量机对特征进行训练与识别。通过仿真算例验证,该方法能有效识别电压暂降源,可应用于电能质量监测系统。
Due to its good performance of time-frequency analysis, S-transform is utilized to obtain the amplitude, time, frequency and phase of the transient signal. A new approach based on S-transform and multi-class support vector machine （SVM） is presented for identification of voltage sags source. Firstly, through S-transform time-frequency analysis, some characteristics related to voltage sags are extracted, and those features are trained and identified using multi-class SVM. The experiment results confirm that this approach is effective to identify the voltage sags source and can be applied to power quality monitoring system. This work is supported by National Natural Science Foundation of Zhejiang Province（No.Y1090182）.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2010年第22期151-155,共5页
Power System Protection and Control
基金
浙江省自然科学基金项目(Y1090182)~~
关键词
电能质量
S变换
多分类支持向量机
电压暂降源
识别
power quality： S-transform
multi-class support vector machine
voltage sags source~ identification
