基于移动时间窗的直流局部放电特征提取方法

电气设备局部放电信号的特征提取是电气设备绝缘在线监测及故障诊断技术领域的前沿课题。当前的研究大多集中在交流局部放电的特征提取方法上,很少有人研究直流局部放电的特征提取方法。文中采用移动时间窗技术,提出一种新的直流局部放电特征提取方法。该方法通过研究时间窗内的局部放电次数、最大视在放电量、平均视在放电量等物理量与电气设备绝缘状态之间的关系,提取它们作为表征电气设备绝缘性能的特征量,用来进行状态评估与故障诊断。实验结果表明,该方法所提取的特征量能够很好地表征电气设备的绝缘性能,为电气设备绝缘的在线监测提供了理论依据。把它应用于地铁牵引变电站馈线电缆绝缘的在线监测中,效果良好。

Features Extraction Method of DC Partial Discharge Based on Mobile Time Window

The feature of partial discharge (PD) signal of electrical devices is a frontier research subject in the online monitoring and diagnosing high voltage insulation failure. In this paper, using the technique of mobile time window (MTW) , a new feature extraction method for DC partial discharge is proposed. By studying the relationship between the insulating condition of electrical devices and the feature parameters in the time window, the discharge number, maximum magnitude and mean magnitude of partial discharges are extracted as the feature parameters which represent the insulating performance of electrical devices. The experimental results show that the insulating performance of electrical devices can be represented very well by the feature parameters which are extracted by the proposed method. Hence that provides a theoretic method for the online monitoring of electrical devices. Finally, the proposed method is applied to the online monitoring of feeder line cable in a metro substation and good effect is obtained.