| 基于ESMD与SVM的电能质量混合扰动识别 |
| |
| 引用本文： | 杨晓楠,吕国强,侯鹏飞,毕贵红.基于ESMD与SVM的电能质量混合扰动识别[J].教育技术导刊,2019,18(11):42-47. |
| |
| 作者姓名： | 杨晓楠 吕国强 侯鹏飞 毕贵红 |
| |
| 作者单位： | 1. 昆明理工大学 冶金与能源工程学院，云南 昆明 650500,2. 西南交通大学 电气工程学院，四川 成都 610000,3. 昆明理工大学 电力工程学院，云南 昆明 650500 |
| |
| 摘 要： | 针对实际电能质量扰动种类繁多、扰动信号差异不明显、存在多种混合扰动，导致识别电能质量非常困难的情况，提出一种基于极点对称经验模式分解方法（ESMD）和支持向量机（SVM）的电能质量混合扰动信号分类识别新方法。首先，对加入白噪声的混合扰动信号利用小波软阈值去噪处理|其次，利用ESMD将信号分解为不同信号分量，对每类扰动的不同信号分量分别提取样本熵和互样本熵特征值，所有分量特征值构成特征向量|最后利用SVM对扰动信号特征向量进行分类和混合扰动识别。研究表明，该方法对混合扰动识别正确率很高，是一个有效的方法。
|
| 关 键 词： | 样本熵 互样本熵 电能质量混合扰动 极点对称经验模式分解方法 支持向量机 |
| 收稿时间： | 2019-01-16 |
Identification of Power Quality Complex Disturbances Based on ESMD and SVM |
| |
| YANG Xiao-nan,LV Guo-qiang,HOU Peng-fei,BI Gui-hong.Identification of Power Quality Complex Disturbances Based on ESMD and SVM[J].Introduction of Educational Technology,2019,18(11):42-47. |
| |
| Authors: | YANG Xiao-nan LV Guo-qiang HOU Peng-fei BI Gui-hong |
| |
| Institution: | 1. Faculty of Metallurgical and Energy Engineering， Kunming University of Science and Technology， Kunming 650093， China| 2. Faculty of Electrical Engineering Southwest Jiaotong University，Chengdu 610000，China| 3. Faculty of Electrical Engineering Kunming University of Science and Technology，Kunming 650500，China |
| |
| Abstract: | For the problem of a wide variety of power quality disturbances in actual power quality disturbance，unobvious disturbance signal， a variety of mixed disturbance phenomenon and difficult power quality identification，this paper presents a new approach based on extreme-point symmetric mode decomposition（ESMD） and support vector machine （SVM） for identification of power quality complex disturbances based on extreme-point symmetric mode decomposition（ESMD） and support vector machine （SVM）. Firstly， the complex disturbances signals with white noise are processed by the wavelet-based soft-threshold de-noising，then the signals are decomposed into different signal components by using ESMD， the sample entropy and the cross sample entropy eigenvalues are extracted for different signal components of each kind of disturbance，all eigenvalues of the components constitute the feature vector， and finally the SVM is used to classify the disturbed signal feature vector and to identify the complex disturbances. The results show that the method can be used to identify the mixed perturbation with high accuracy and can provide an effective method for this research. |
| |
| Keywords: | sample entropy cross sample entropy power quality complex disturbances extreme-point symmetric mode decomposition support vector machine |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
