发电技术 ›› 2023, Vol. 44 ›› Issue (6): 896-908.DOI: 10.12096/j.2096-4528.pgt.23110

• 智能电网 • 上一篇    

考虑用户诉求差异的居民可控负荷资源优化控制

黄河1, 王燕2, 姜念3, 吴强1, 张雅静4, 杨秀媛4   

  1. 1.国网江苏省电力有限公司,江苏省 南京市 210000
    2.国网北京延庆供电公司,北京市 延庆区 102100
    3.天地电研(北京)科技有限公司,北京市 昌平区 102206
    4.北京信息科技大学自动化学院,北京市 海淀区 100192
  • 收稿日期:2023-09-10 出版日期:2023-12-31 发布日期:2023-12-28
  • 通讯作者: 张雅静
  • 作者简介:黄河(1984),男,硕士,高级工程师,主要研究方向为配电网规划与管理,992708605@qq.com
    王燕(1996),女,硕士,主要研究方向为控制工程、新能源发电,1049698158@qq.com
    张雅静(1984),女,博士,副教授,主要研究方向为电力电子、电力传动与新能源发电,本文通信作者,zhangyajing@bistu.edu.cn
  • 基金资助:
    国家自然科学基金项目(52107176)

Optimal Control of Residents’ Controllable Load Resources Considering Different Demands of Users

He HUANG1, Yan WANG2, Nian JIANG3, Qiang WU1, Yajing ZHANG4, Xiuyuan YANG4   

  1. 1.State Grid Jiangsu Electric Power Co. , Ltd. , Nanjing 210000, Jiangsu Province, China
    2.State Grid Beijing Yanqing Power Supply Company, Yanqing District, Beijing 102100, China
    3.Beijing T&D Power Research Co. , Ltd. , Changping District, Beijing 102206, China
    4.School of Automation, Beijing Information Science & Technology University, Haidian District, Beijing 100192, China
  • Received:2023-09-10 Published:2023-12-31 Online:2023-12-28
  • Contact: Yajing ZHANG
  • Supported by:
    National Natural Science Foundation of China(52107176)

摘要:

对需求侧居民可控负荷资源进行优化控制,是虚拟电厂(virtual power plant,VPP)的一个主要方面,是提高风电和光电等新能源利用率、实现碳中和的重要手段。控制可控负荷的关键是满足用户基本用电需求,而不同用户的用电需求不尽相同。因此,提出了考虑用户需求的居民可控负荷资源优化控制方法。考虑到不同用户的用电诉求差异性,VPP作为电网与负荷群的中间环节,对可控负荷进行管理。首先,提出改进最优k值选取的K-Means聚类算法,对居民可控负荷进行聚类;其次,依据诉求差量函数,对聚类的每组负荷调控后用电情况与用电原始需求偏差进行量化描述,提出基于诉求差量的优先级划分规则;最后,在进行调控时VPP通过对不同优先级的组别补偿不同费用的方式,实现VPP收益最大和用户用电行为改变最小的调控目标。仿真验证结果表明,所提优化控制策略可有效实现负荷曲线的削峰填谷,并且在对用户用电需求影响较小的情况下保证VPP收益。

关键词: 虚拟电厂(VPP), 可控负荷, 负荷聚类, 用户诉求, 削峰填谷

Abstract:

Optimizing the controllable load resources of residents on the demand side is a major aspect of virtual power plant (VPP). It is an important means to improve the utilization rate of new energy such as wind power and photovoltaic power, to achieve carbon neutrality. The key to control controllable load is to meet the basic power demand of users, and different users have different power demand. Therefore, an optimal control method of residents’ controllable load resources considering user demand was proposed. Considering the different demand of different users, as an intermediate link between the power grid and the load group, VPP manages the controllable load. Firstly, an improved K-means clustering algorithm for optimal k selection was proposed for clustering. Then, according to the demand difference function to quantify the deviation between the power consumption and the original demand of each group after load regulation, a priority division rule based on the demand difference was proposed. Finally, in the regulation, VPP compensated different costs for different priority groups to achieve the regulation goal of maximizing VPP revenue and minimizing changes in user electricity consumption behavior. The simulation results show that the proposed optimal control strategy can effectively cut the peak and fill the valley of the load curve, and ensure the VPP revenue with less impact on the power demand of users.

Key words: virtual power plant (VPP), controllable load, load clustering, user demand, peak shaving and valley filling

中图分类号: