基于轨迹特征关键点匹配的电压暂降同源数据精确检测算法

Accurate Detection Algorithm for Homologous Voltage Sag Data Based on Matching of Feature Key Points for Trajectories

为弥补现有电压暂降同源检测算法的不足,提高同源检测算法的精度,提出了一种基于轨迹特征关键点匹配的电压暂降同源数据精确检测算法。首先,将电压暂降数据有效值(RMS)波形转换为灰度轨迹图片,基于尺度不变特征转换算法对电压暂降灰度轨迹图片进行分析,提取RMS轨迹的特征关键点;然后,利用特征关键点的梯度方向信息与待检测电压暂降数据的RMS轨迹进行匹配计算;最后,以匹配规则作为电压暂降同源数据判断的标准。所提算法可以很好地弥补现有同源检测算法中的不足,降低了对检测数据集的客观要求,具有更强的工程适用性及实际应用价值。

To make up for the shortcomings of the existing homologous voltage sag detection algorithms and improve the accuracy of the algorithms, this paper proposes an accurate detection algorithm of homologous voltage sag data based on the matching of feature key points for trajectories. Firstly, the root mean square(RMS) waveform of the voltage sag data is converted into a greyscale trajectory image. Then, the gray-scale trajectory image of the voltage sag is analyzed based on the scale-invariant feature transform(SIFT) algorithm, and the feature key points of the RMS trajectory are extracted. Secondly, the gradient direction information of feature key points is used to match the RMS trajectories of voltage sag data to be detected. Finally, the matching rule is used as the criterion for the judgment of homologous voltage sag data. The proposed algorithm can make up for the shortcomings of the existing homologous detection algorithms, reduce the requirements for the detection data set, and have better engineering applicability and practical application value.