| 引用本文: | 冯志颖,唐文虎,吴青华,陆国俊,栾乐.考虑负荷纵向随机性的用户用电行为聚类方法[J].电力自动化设备,2018,(9): |
| FENG Zhiying,TANG Wenhu,WU Qinghua,LU Guojun,LUAN Le.Users’ consumption behavior clustering method considering longitudinal randomness of load[J].Electric Power Automation Equipment,2018,(9): |
|
| 摘要: |
| 针对不考虑负荷纵向随机性所导致的数据损失和用户误分类的问题，提出了一种考虑负荷纵向随机性的基于推土机距离(EMD)的用户用电行为识别新方法。该方法通过统计电力用户同一时刻多天的负荷分布情况，从横向和纵向2个角度全面表征用户的用电行为。并结合EMD和欧氏距离度量不同用户用电行为的差异程度。以一组国际通用的居民用电负荷作为算例进行分析，算例结果表明，在横向特性较为相似的用户中，该方法能够很好地提取用户的纵向特性。定性和定量分析均表明，该方法对用户负荷的聚类效果精细合理。 |
| 关键词: 智能电网 用电行为 负荷曲线 纵向随机性 聚类方法 推土机距离 |
| DOI：10.16081/j.issn.1006-6047.2018.09.007 |
| 分类号:TM714 |
| 基金项目:国家高技术研究发展计划(863计划)项目(2015AA050201)；国家自然科学基金资助项目(51477054) |
|
| Users’ consumption behavior clustering method considering longitudinal randomness of load |
|
FENG Zhiying1, TANG Wenhu1, WU Qinghua1, LU Guojun2, LUAN Le2
|
|
1.School of Electric Power, South China University of Technology, Guangzhou 510640, China;2.Guangzhou Power Supply Co.,Ltd.,Guangzhou 510000, China
|
| Abstract: |
| To address the problem of data loss and misclassification caused by not considering the longitudinal randomness of load, a novel user’s consumption behavior recognition method based on EMD(Earth Mover’s Distance),which considers the longitudinal randomness of load, is proposed. The approach takes both the transverse and longitudinal perspectives into account to fully characterize the power consumption behaviors of electricity consumers by collecting the load distribution of electricity users at the same moment for a few days. The EMD is combined with the Euclidean distance to measure the difference degrees among different user’s electricity consumption behaviors. A group of international general household electricity load is employed as a benchmark example. Case study results indicate that the proposed method can better capture the longitudinal characteristics of customers with similar transverse characteristics. Both quantity and quality analysis show that the method is accurate and feasible for clustering the users’ load. |
| Key words: smart grid consumption behavior load curve longitudinal randomness clustering algorithms earth mover’s distance |
