基于S变换与特征优选的电能质量扰动识别  

Classification of Power Quality Disturbance Utilizing S-transform and Optimal Feature Selection

作　　者：朱勇 陶用伟 李泽群 ZHU Yong;TAO Yongwei;LI Zequn(Kaili Power Supply Bureau of Guizhou Power Grid Co.,Kaili 556000,China)

机构地区：[1]贵州电网有限责任公司凯里供电局,贵州凯里556000

出　　处：《电工技术》2023年第21期97-100,共4页Electric Engineering

基　　金：南方电网公司项目(编号060800KK52210005)。

摘　　要：针对S变换提取电能质量扰动信号的时频特征存在冗余,影响识别的精度和实时性的问题,提出了一种基于S变换与特征优选的电能质量扰动识别方法,使用S变换对11种电能质量扰动信号进行分析后提取时-频域特征,与原始扰动信号的幅值特征构成原始特征向量。然后提出了一种基于二进制粒子群优化算法和K近邻的扰动信号特征选择和分类方法。仿真结果表明该方法对单一扰动、复合扰动及叠加噪声的情况均有较好的分类性能。The time-frequency features extracted by S-transform are redundant,which affects the classification accuracy and real-time performance.To solve these problems,this work proposes a PQD identification method based on S-transform and optimal feature selection.Firstly,the time-frequency domain features of 11 kinds of PQD signals are extracted using S-transform,and the feature vector is formed with the amplitude features of the original voltage signals.Then,a PQD feature selection and classification method based on binary particle swarm optimization(BPSO)and K-nearest neighbor(KNN)is proposed.The simulation results show that the proposed method has better classification performance in classifying single and compound disturbances.

关 键 词：电能质量扰动 S变换 粒子群优化 最优特征选择 

分 类 号：TM712[电气工程—电力系统及自动化]