基于暂态残压计算的接触网故障性质智能辨识

摘要

In order to avoid the permanent fault making the second short-circuit current shock impact on overhead contact line due to the reclosing of circuit breaker,we designed the catenary fault intelligent recognition to identify the fault characteristics of circuit breaker's automatic reclosing lock.According to the physical process of voltage's and current's development in fault and the data collected during the tripping of circuit breaker to the relock in catenary,we analyzed the identification mechanism of the fault properties of the catenary.According to the actual situation of the up and down feeder circuit breakers tripping at the same time in contact wire fault,we established the equivalent circuit and deduced the calculation formula for the catenary residual voltage (transient residual voltage for short) when instantaneous fault occurred in the first half part,middle and the latter half part of catenary in the transient process of residual voltage.A new intelligent recognition method was presented to recognize catenary fault.When the transient residual voltage tended to be 0,the catenary fault was permanent fault.When the transient residual voltage was over a threshold and there was a turning point in exponential decay curve,the catenary fault was instantaneous fault.In order to verify the effectiveness of the automatic reclosing of circuit breaker by the intelligent identification method based on the fault characteristics of the catenary,we used artificial grounding method to simulate the actual permanent fault and instantaneous fault of catenary on a special railway line.The tested values and calculated values of transient residual voltage were compared.Results show that the relative error between the tested and calculated values of transient residual voltage at the turning point is 1.5 ％.The proposed identification method can correctly recognize the fault properties of the catenary.The research method has no artificial hypothesis in the process of derivation.The research result has universal applicability,and the comprehensive criteria can meet the requirements of engineering applications.%为了避免接触网发生永久性故障时断路器因重合闸造成其二次受短路电流冲击而损坏,实现人工智能控制断路器的自动重合闸,根据接触网故障时电压、电流发展的物理过程和在变电所馈线断路器的跳闸至重合闸期间采集的接触网故障信息,分析接触网故障性质的辨识机理;按照接触网故障时上下行馈线断路器同时跳闸的实际情况,建立等值电路,推导接触网前半段、中间和后半段发生瞬时性故障时暂态过程中的残压(简称暂态残压)计算式.据此提出接触网故障性质的智能辨识方法:暂态残压趋近于0的接触网故障为永久性接触网故障,暂态残压超过某一阈值而且指数衰减曲线有拐点的接触网故障为瞬时性接触网故障.在某铁路专用线采用人工接地方式分别模拟接触网发生永久性故障和瞬时性故障的情况,验证基于接触网故障性质智能辨识方法控制断路器自动重合闸的有效性,并对比暂态残压的试验值与计算值.结果表明:暂态残压拐点处的计算值与试验值的相对误差为1.5％;提出的辨识方法能够正确判断接触网的故障性质,也没有人为的假设条件,具有普遍适用性,综合判据能够满足工程应用要求.