中文摘要：

城市轨道交通大多采用弓网（受电弓与接触网）接触方式向电力机车供电。当弓网分离时,将造成燃弧发生和牵引电流扰动,恶化机车受流。针对地铁直流牵引电流特征,提出了一种改进的小波包奇异熵算法,并用于燃弧牵引电流处理分析,得出牵引电流扰动指标Γ。根据非接触式弓网燃弧检测系统采集广州地铁现场试验数据,分析结果表明：改进后小波包奇异熵能直观、有效地提取牵引电流扰动特征;扰动指标Γ能有效反映牵引电流扰动程度,衡量机车受流质量;牵引电流扰动指标Γ与燃弧紫外能量呈幂函数关系,燃弧紫外能量越大,扰动越严重,特别当燃弧紫外能量超过400 mJ时,牵引电流扰动Γ急剧上升,机车受流受到严重影响。

英文摘要：

In urban railway traffic, electric locomotives get power through the contact between the overhead pantograph and catenary. When the pantograph and catenary separate, arc discharges will occur, which causes traction current distortion and deteriorates locomotive current collection. In accordance with characteristics of metro DC traction current, we proposed an improved wavelet packet singular entropy algorithm. The algorithm was used to process and analyze arc traction current, and traction current disturbance degree index Γ was given. Test data were collected by using non-contact pantograph-catenary arc detection device in Guangzhou Metro. The results show that the improved wavelet packet singular entropy can be intuitive and effective extraction of traction current disturbance characteristics. Disturbance index Γ can effectively reflect the disturbance degree of traction current, measuring the quality of locomotive current collection. There is a power function relationship between disturbance index of traction current and arc ultraviolet energy, and the greater the arc ultraviolet energy, the more serious the disturbance. Especially when the arc ultraviolet energy is greater than 400 m J, traction current disturbance Γ sharply increases and the current receiving quality of locomotive is seriously affected.