基于改进的模糊c-均值聚类算法的负荷特性指标分析与分类

Index analysis and classification of load characteristics based on improved fuzzy c-means algorithm

负荷侧管理是利用不同特性负荷作为需求响应资源,对具有多样性和复杂性的负荷种类进行较高精度的分类,提高负荷曲线的相似度,便于组合优化。为了对实际负荷更加精确地分类,提出一种基于改进的模糊c-均值聚类算法的负荷特性指标分类方法。该算法在聚类分割迭代中采用加权欧氏距离,对每种聚类中心进行负荷特性指标分析,并对各类别中的负荷采用不同控制方法。最后分类结果表明,所提方法使各分类中负荷具有较高相似性,为后续负荷的预测和控制奠定了基础。

Load side management is a method that utilizes the load with different characteristics as the demand response resources to classify the various and complex load in a higher accuracy, which improves the similarity of the load curve to facilitate the combination and optimization. In order to get a more accurate classification of the actual load, a load performance index based on the improved fuzzy c-mean clustering（ FCM） algorithm is proposed. The weighted Euclidean distance is adopted in the clustering segmentation iteration in this algorithm, then the load performance index of each clustering center is analyzed, so the different control methods can be taken to control the loads which belong to different category. The final classification results show that the proposed method has a higher accuracy to the load classification, which lays the foundation for the prediction of the load and the combination and optimization of the control.