随着可再生能源渗透水平的不断提高，现代电力系统面临着更多不可避免的不确定性，这些不确定性可能导致系统的弱阻尼振荡问题。对于可再生能源渗透率很高的电力系统，检测同步发电机之间的相干性是态势感知的关键环节。为此，文中提出了一种基于广域测量系统（Based Area Measurement System，WAMS）的相干检测算法，该方法采用了核主成分分析（Kernel Principal Component Analysis, KPCA）和聚类(Affinity Propagation，AP)分析法可应用于可再生能源广泛渗透的电力系统。文中提出了几种轨迹相似度指标，用于确定惯性中心（Center of Inertia,COI）坐标中任意两个发电机轨迹之间的相似性；提出了一种基于KPCA方法的集成轨迹相似度指标，以解决多个指标之间的相干性问题；随后采用AP聚类分析方法检测同步发电机之间的相干性，可无需预先指定聚类的数量；利用高可再生能源发电渗透率的华南电力系统和包括张北风电场的华北电力系统的一部分进行仿真分析，结果证明了所提方法的适用性和实用性。
英文摘要:
With the continuous improvement of renewable energy penetration level, modern power system is faced with more inevitable uncertainties, which may lead to the problem of weak damping oscillation of system. For the power system with high renewable energy penetration, detecting the coherence between synchronous generators is one of the key steps of situation awareness. Therefore, this paper presents a coherent detection algorithm based on the wide area measurement system (WAMS). This method adopts kernel principal component analysis (KPCA) and affinity propagation (AP) clustering analysis, which can be applied to the power system with extensive renewable energy penetration. In this paper, several trajectory similarity indices are proposed to determine the similarity between any two generator trajectories in the center of the inertia (COI). Secondly, an integrated trajectory similarity index based on KPCA is proposed to solve the problem of the coherence between multiple indexes. Then, AP clustering analysis method is used to detect the coherence between synchronous generators without specifying the number of clusters in advance. Finally, the simulation analysis of the South China power system with high renewable energy generation permeability and the North China power system including Zhangbei wind farm is carried out, and the results show the applicability and practicability of the proposed method.