基于感知哈希序列的电压暂降事件同源识别

Homology identification of multi voltage sag events based on perceptual Hash sequence

针对现有方法未能充分利用电压采样值信息且对特殊录波情况适应性较差的问题,提出基于感知哈希序列相似性的电压暂降事件同源识别方法。首先,提出一种基于突变点检测和录波情况识别的暂降数据段提取方法,将对录波文件中暂降事件的识别转化为对完整和不完整暂降数据段的分别识别。其次,利用格拉姆角场(Gramian angular field, GAF)将提取后的暂降数据段从一维时间序列形式转化为二维图像,并利用感知哈希算法将其转化为哈希序列。然后,通过欧氏距离刻画相似性,并根据选取的阈值完成电压暂降同源识别。最后,利用IEEE30节点系统生成的仿真数据以及北京、福建地区的实测数据进行验证,证明了该方法具有识别精度高、适应性好等优势。

Existing methods for identification of voltage sag, cannot make full use of the voltage sampling value information and have poor adaptability to special recording conditions. Thus a homologous identification method of voltage sag events based on perceptual Hash sequence similarity is proposed. First, this paper proposes a method for extracting sag data segments based on mutation point detection and recording situation recognition. This transforms the recognition of sag events in recording files into the recognition of complete and incomplete sag data segments respectively. Secondly, it uses the Gramian angle field(GAF) to convert the extracted sag data segment from one-dimensional time series to two-dimensional images, and applies the perceptual Hash algorithm to convert it into a Hash sequence. Then, by describing the similarity through Euclidean distance, it completes the homology identification of voltage sag according to the selected threshold. Finally, it applies the simulation data generated by the IEEE30 node system to the measured data in Beijing and Fujian. This shows that the method has advantages of high recognition accuracy and good adaptability.