城轨直流牵引系统短路故障早期检测研究

Research on Short-Circuit Fault Early Detection for Urban Railway DC Traction System

城轨直流牵引系统分布狭长密集,一旦发生故障,快速上升的电流和所产生的短路能量极易引起火灾事故,威胁列车安全运行和乘客人身安全。目前,直流短路检测研究存在短路故障特性研究不足且缺乏实验验证等问题。鉴于此,从电力电子换流器内部电路结构出发,分析换流器发生短路故障时,其内部暂态过程及其特性,并以此特性开发相应的直流牵引系统短路实验平台。引入Mallat快速小波算法,提出电压、电流双小波分解的直流短路故障早期检测方法,经实验验证表明,该方法不仅可区分正常工况直流暂态过程与短路故障,且可在短路故障发生300μs内实现有效的故障早期检测及辨识。对比传统直流馈线电流上升率及电流增量(DDL)保护法,该方法故障检测速度提高10余倍,检测故障时短路电流不足DDL方法的10%,可为早期检测后续的故障早期抑制与快速开断的深入研究奠定良好的技术前提与基础。

The urban railway DC traction system is distributed in a narrow and dense manner.Once a fault occurs,the rapidly rising short circuit current and the short-circuit energy generated are susceptible to fire accidents,threatening the safe operation of the train and the personal safety of the passengers.At present,the research on DC short circuit detection technology is lack of sufficient understanding of the characteristics of DC short circuit and practical experimental verification.In view of this,the transient process and transient characteristics of the converter are analyzed in detail starting from the internal circuit structure of the power electronic converter;the corresponding DC short circuit experiment platform is set up at the same time.Then the Mallat fast wavelet algorithm is introduced,and an early detection technique for DC fault is presented based on voltage and current dual wavelet decomposition.Finally,the experimental results show that the proposed technology can distinguish the normal working condition DC transient process and short circuit fault effectively,and realize an effective early detection and identification of faults within 300 μs of short-circuit faults.Compared with the traditional DC feeder current rate and current increment(DDL) protection method,the fault detection speed of the method is increased by more than 10 times,and the short-circuit current is less than 10% of the DDL method when detecting the fault,which can lay a good technical premise and foundation for intensive research on fault early suppression and fast breaking.