【摘要】 励磁涌流是变压器保护误动的主要原因之一,为此提出一种基于峭度系数的快速区分变压器励磁涌流和故障电流的新方法。对变压器差动电流采样数据进行一系列处理后,将其视为一变量的概率分布,并求取峭度系数。由于变压器故障波形呈正弦特性,所计算的峭度系数小于零;而励磁涌流因为含有非平稳波峰、间断角,对应的峭度系数大于零;以峭度是否大于零作为区别变压器涌流与故障电流的判据。仿真结果表明,该方法能可靠区分故障电流与励磁涌流(或和应涌流),可显著缩短空投故障变压器的闭锁时间,且具有良好的抗噪性能。更多还原

【Abstract】 The inrush current is one of the main reasons of maloperation in transformer protection. Based on kurtosis, a novel method to distinguish inrush current is proposed in this paper. After a series of processing to the sampled data of transformer differential current, the new data is treated as probability distribution of a variable and the kurtosis coefficient is calculated. The kurtosis coefficient of internal fault current is much smaller than 0 due to its sine characteristic; while the kurtosis coefficient of inrush current is much bigger than 0 for its non-stationary crest and dead angle. So, the kurtosis can be taken as the criterion distinguishing inrush current from internal fault current. The simulation results indicate that this method can reliably discriminate magnetizing inrush current(or sympathetic inrush current) from fault current. In addition, blocking time caused by no-load closing faulted transformers can be shorted significantly. In particularly, this approach has very strong ability to resist noise interference.更多还原