采用奇异值梯度信息的暂态电能质量扰动自适应检测方法

Adaptive detection method of transient power quality disturbance based on singular value gradient information

为了满足对电网非平稳扰动信号快速、准确分析的要求,提出了一种采用奇异值梯度信息的暂态电能质量扰动检测新方法。通过滑动窗奇异值分解(SVD)方法提取信号的变化特征、降低噪声干扰,并通过奇异值梯度求取扰动指示信号,得到初步定位结果。提出无参自适应阈值,进一步抑制噪声干扰并实现对暂态扰动信号的检测定位。所提算法原理简单,无需进行前置滤波及参数调节。一系列仿真试验的对比分析结果表明,所提算法定位准确、抗干扰能力强,对过零点扰动也有较好的检测效果。通过对变电站实际暂态扰动数据的检测分析,进一步验证了所提算法的有效性。

In order to satisfy the requirement of fast and accurate analysis of non-stationary disturbance signals in power grid,a novel detection method of transient power quality disturbance based on singular value gradient information is proposed. The sliding window SVD(Singular Value Decomposition) method is used to extract the change characteristics of signals and reduce the noise interference. Then the disturbance indication signal is obtained by using the singular value gradient information,and the initial location result is obtained. A parameter-free adaptive threshold is proposed to further suppress the noise interference and detect and locate the transient disturbance signals. The proposed algorithm is simple in principle and does not require pre-filtering and parameter adjustment. The analysis and comparison results of a series of simulation tests show that the proposed algorithm has the advantages of accurate location,strong anti-interference ability and good detection effect on zero-crossing disturbance. The validity of the proposed algorithm is further verified through the detection analysis of the actual transient disturbance data of substation.