基于FCM和DTW的多分区电压支撑能力评估

电网规模的发展和负荷的增长使得不同分区无功电压情况的差异性增强,负荷中心区域电压稳定问题日益突出,而负荷较轻区域部分时段易出现高电压和无功过剩,亟需全面评估电网中多分区的无功电压状态,为全局无功协调控制及分区间互动支撑提供数据支持。鉴于此,文中提出一种基于模糊C均值聚类(FCM)和动态时间归整算法(DTW)的评估方法。该方法采用FCM对分区多个时间断面的指标数据进行模糊聚类,形成聚类中心并提取样本的平均特征;同时,利用DTW对聚类中心数据与参考样本进行模式识别,可评估出电网多分区的电压支撑能力。以某典型电网为例,验证所提评估方法的有效性。

Multi-partition voltage support capability evaluation based on FCM and DTW

With the development of power grid scale and the increase of load, the difference of reactive voltage in different regions is enhancing, and the problem of voltage stability in the center region of the load is increasingly prominent, while high voltage and reactive power surplus are likely to occur in some periods in the lighter load region. It is urgent to comprehensively evaluate the reactive voltage status in multiple regions of the power grid, so as to provide data support for the coordinated control of global reactive power and the interaction between regions. In view of this, an evaluation method based on fuzzy C-means clustering (FCM) and dynamic time warping (DTW) is proposed. In this method, FCM is used to fuzzy cluster the index data of multiple time sections to form the cluster center and extract the average characteristics of the samples. At the same time, DTW is used for pattern recognition of the cluster center data and the reference samples, which can evaluate the voltage support ability of multiple time sections. Taking a typical power grid as an example, the effectiveness of the proposed evaluation method is verified.