基于小波包和支持向量机的电能质量扰动识别  

Power Quality Disturbance Signals Identification Based on Wavelet Packet and SVM

作　　者：李燕楠[1] 包广清[1] 

机构地区：[1]兰州理工大学电信学院,兰州甘肃730050

出　　处：《电力科学与工程》2012年第3期21-26,共6页Electric Power Science and Engineering

基　　金：国家自然基金项目(50877034)

摘　　要：电能质量扰动识别是电能质量检测系统的重要组成部分,也是进一步采取适当措施对其进行治理和控制的前提和依据。通过MATLAB仿真软件建立5种典型扰动信号的模型,包括电压突降、突升、中断、脉冲暂态及谐波;利用小波包分析方法对上述扰动信号进行特征向量提取;并采用粒子群算法对SVM核函数参数γ和惩罚参数C寻优,确定最优SVM分类模型,最终测试精度为98.125%,表明该算法实时性强、识别精度高,从而验证了所用方法的可行性。Power quality disturbance recognition is an important component of the power quality detection system, which also is the basis and premise for the management and control of the power quality system . For the pattern recognition of disturbance signal, this paper presents five benchmarks of disturbance signal by MATLAB; Then, feature vectors of the disturbance signal which are extracted by the wavelet packet transforms; The particle swarm optimization is used to choose optimal parameters. Numerical results shows this approach achieved a classification accuracy 0f 98. 125%, a few training samples and training time is short, a good real-time performance . It is an effective method for identify power disturbance signals.

关 键 词：电能扰动信号 小波包变换 支持向量机(SVM) 粒子群算法(PSO) 

分 类 号：TM712[电气工程—电力系统及自动化]