中国电力 ›› 2018, Vol. 51 ›› Issue (1): 179-184.DOI: 10.11930/j.issn.1004-9649.201605027

• 技术经济 • 上一篇    

基于高维赋范与SGHSA算法的用电峰谷时段划分模型

刘树勇1, 李娜1, 符景帅2   

  1. 1. 国网天津市电力公司经济技术研究院, 天津 300171;
    2. 华北电力大学 经济与管理学院, 北京 102206
  • 收稿日期:2016-05-04 修回日期:2016-12-15 出版日期:2018-01-05 发布日期:2018-02-28
  • 作者简介:刘树勇(1978—),男,天津人,高级工程师(教授级),从事电力技术经济研究,E-mail:39230362@qq.com。
  • 基金资助:
    国家电网公司科技项目(SGTYHT/14-JS-188)。

An Electricity Peak and Valley Time Period Division Model Based on High Dimensional Normed Space and SGHSA Algorithm

LIU Shuyong1, LI Na1, FU Jingshuai2   

  1. 1. State Grid Tianjin Economic Research Institute, Tianjin 300171, China;
    2. School of Ecomonic and Management, North China Electric Power University, Beijing 102206, China
  • Received:2016-05-04 Revised:2016-12-15 Online:2018-01-05 Published:2018-02-28
  • Supported by:
    This work is supported by Science and Technology Project of State Grid Corporation of China(No.SGTYH/14-JS-188).

摘要: 科学划分用电峰、平、谷时段是合理制定峰谷分时电价的基础,时段划分结果应满足长期适用性,并且最大限度地反映出不同时段间负荷的差异。对峰谷时段划分的建模问题展开研究,首先,通过定义涵盖较长时间周期信息的各时点上高维负荷向量及其范数构造时段划分的数据样本集;其次,在范数的基础上引入半梯形隶属函数来确定各时点属于峰、平、谷时段的隶属度;再次,建立阈值优化模型,通过自适应全局寻优搜索算法(SGHSA)对分类阈值进行寻优,并完成时段划分模型的构建;最后,结合某地区实例进行分析,在验证模型合理性的基础上,输出该地区居民用电峰谷时段划分的参考结果。

关键词: 分时电价, 时段划分, 高维负荷向量, 模糊聚类, SGHSA

Abstract: Scientific division of peak, flat and valley time period is fundamental for rational development of time-of-use tariff. Time division results should meet long-term applicability, and maximally reflect load differences in different time periods. Modeling problem of peak and valley time division is discussed. Firstly, data sample set of division model is founded by defining high dimensional load vectors which embody information of a long period and its norm at each time point. Secondly, by introduction of semi gradient membership function, membership degree of each time point to peak, flat and valley period are defined based on norm of load vectors. Thirdly, threshold optimization model is established to optimize threshold by using SGHSA searching algorithm. This completes the construction of period division model. Finally, exemplary application on a certain area proves reasonability of proposed model.

Key words: time-of-use tarrif, time period division, high dimensional load vector, fuzzy clustering, SGHSA

中图分类号: