摘要
故障电弧是引发电气火灾事故的主要原因之一。本文将支持向量机引入故障电弧研究领域,进行不同负荷情况下的故障电弧识别检测。首先参照美国UL1699标准进行实验采集电流数据,然后利用支持向量机实现故障电弧的训练、检测识别,并对训练、识别结果进行了分析,实验证明本文的检测方法具备一定的泛化能力。
Fault arc is one of the prime reasons causing electrical fire accidents. This paper has introduced support vector machine into fault arc research field and recognized fault arcs under different loads. Firstly, tests were made to collect data based on American Standard UL1699. And then training, detection and recognition of fault arcs are made by support vector machine. The analysis of the results has proved the test method is of certain ability of generalization.
出处
《有色冶金设计与研究》
2011年第4期58-61,共4页
Nonferrous Metals Engineering & Research
关键词
支持向量机
核函数
故障电弧
support vector machine
kernel function
fault arc
