摘要
将小波分析与支持向量机(SVM)相结合对供电线路故障进行自动诊断。运用变尺度多分辨小波分析方法对供电线路故障信号进行特征分析,SVM能够对小样本数据进行模式识别并且具有很好的分类推广能力。在小波分析信号特征提取的基础上,利用分布式多SVM分类器识别故障。实验表明,该方法能有效地对供电线路故障进行准确识别和诊断。
The combination of wavelet analysis and support vector machine is introduced to automatically diagnose power supply line fault. Multi-resolution and multi-scale wavelet analysis is used to perform characteristic analysis of fault signals. Support vector machine （SVM） can perform pattern recognition on small-samples and has excellent capability of classification and generalization. Based on wavelet analysis of signal characteristics, distributed multi-SVM classifier is utilized to identify the power supply line fault. The experimental results show that this method can effectively identify and diagnose power supply line fault.
出处
《北京石油化工学院学报》
2005年第3期36-40,共5页
Journal of Beijing Institute of Petrochemical Technology
关键词
小波分析
支持向量机
信号特征
线路诊断
wavelet analysis
support vector machine
signal feature
power line diagnosis
