考虑高阻接地的配电网故障检测方法

Fault Detection of Distribution Network Considering High Impedance Faults

实际配电系统中存在一些高阻态的故障,表现为微弱、非线性、随机和不稳定的现象,这对于故障的检测提出了更高的挑战。为此,该文提出一种考虑高阻接地的配电网故障检测方法。首先,通过分析大量的现场波形数据,得到了故障发生前后的波形特性。并且,为减弱噪声的干扰,对现场数据进行预处理。进一步,提出了一种基于数学形态学的故障特征增强方法,以放大故障在发生时刻的响应,同时减弱故障暂态过程以及正常运行状态下的畸变响应。接着,提出了一种基于狄利克雷过程高斯混合模型的故障时刻检测方法,通过对增强后的故障特征进行自适应判断,实现故障时刻的快速准确检测。基于实际配电网现场数据,进一步验证了该文所提方法的优势。实验结果表明,所提方法通用性强,仅需要配电网零序电流数据。同时,该文方法具有较高的检测精度和检测效率,满足配电网的可靠性和安全性要求。

There are some high-resistance faults in the actual distribution systems with weak,nonlinear,random and unstable characteristics,which pose a higher challenge to the detection of faults.Therefore,a new method of fault detection in the distribution networks considering high impedance faults is proposed in this paper.Firstly,the waveform characteristics before and after the fault are obtained by analyzing large amounts of the field waveform data.The field data is denoised to reduce the interference of noise.Furthermore,a fault feature enhancement based on the mathematical morphology is proposed to amplify the fault response at the moment of occurrence,while weakening the distortion response in both the fault transient processes and the normal operating conditions.Then,a fault moment detection based on the Dirichlet process Gaussian mixture model is proposed.Through adaptively judging the enhanced fault features,the fast and accurate detection of fault moment is realized.Based on the field waveform data,the superiority of this proposed method is further verified.The experimental results show that the proposed method has stronger universality,only needing the zero-sequence current data.At the same time,it has high detection accuracy and efficiency which meets the reliability and safety requirements of the distribution networks.