Internal Over-Voltage Identification Method of Distribution Network Based on AD-SVM Algorithm

2018 International Conference on Advanced Technologies in Energy, Environmental and Electrical Engineering

能源学;生态环境科学;无线电电子学

Liao, Yufei^1 ; Yang, Gengjie^1 ; Gao, Wei^1 ; Guo, Moufa^1

College of Electrical Engineering and Automation, Fuzhou University, Fuzhou

350116, China^1

关键词: Atomic decomposition;    Characteristic atoms;    High-accuracy;    Identification method;    Over-voltages;    SVM algorithm;    Three phase voltage;    Time frequency characteristics;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/223/1/012013/pdf
DOI  :  10.1088/1755-1315/223/1/012013

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

An internal over-voltage identification method based on atomic decomposition(AD) and Support Vector Machine(SVM) is proposed. Firstly, the three-phase voltages of bus are decomposed by AD algorithm to get the optimal atoms. According to the atom frequency, the optimal atoms are constructed as characteristic atom matrix, which can reflect the time-frequency characteristics of the over-voltage signal. Features are obtained from the characteristic atom matrix by singular value decomposition. Finally, the features are input into the SVM to achieve the identification of seven types of internal over-voltages. The proposed method is verified on the simulation platform. The results show that the proposed method has high accuracy.