中文摘要：

为实现多负载同时工作时能准确诊断某一支路发生的电弧故障,按照UL 1699和GB 14287.4标准搭建了碳电极电弧故障试验平台。依据试验结果,基于相空间重构将电弧电流引入高维空间,分别采用容积维数和关联维数提取故障电弧特征向量。运用最小二乘支持向量机构造电弧故障诊断器,实现了串联电弧故障准确诊断。研究结果可为电弧式电气火灾监控探测器提供理论基础。

英文摘要：

In order to realize the accurate diagnosis of the arc fault in a slip road in the condition of a variety of load at the same time working, a carbon electrode arc fault experimental platform was built according to the UL 1699 and GB 14287.4 standards. Depending on the experimental results ,the arc current signal is introduced into the high dimensional space by using the phase space reconstruction algorithm. Moreover, the characteristic vector of the fault arc is extracted by the volume dimension and the correlation dimension respectively. A new arc fault diagnosis method, based on the construction of are fault diagnosis device using least squares support vector machine （LSSVM） was proposed. The research result can provide the theoretical basis for arc type electrical fire monitoring detector.