• CSCD核心库收录期刊
  • 中文核心期刊
  • 中国科技核心期刊

电力建设 ›› 2021, Vol. 42 ›› Issue (11): 100-107.doi: 10.12204/j.issn.1000-7229.2021.11.011

• 智能电网 • 上一篇    下一篇

基于空调负荷数据挖掘的配电网扩展规划方法

苏宁1, 李笑彤1, 梁惠施2, 贡晓旭2(), 吕风波2, 席嫣娜1, 齐步洋3   

  1. 1.国网北京市电力公司经济技术研究院,北京市 100055
    2.清华四川能源互联网研究院,成都市 610213
    3.清华大学电机工程与应用电子技术系,北京市 100084
  • 收稿日期:2021-04-22 出版日期:2021-11-01 发布日期:2021-11-02
  • 通讯作者: 贡晓旭 E-mail:gongxiaoxu01@163.com
  • 作者简介:苏宁(1983),男,硕士,工程师,研究方向为电力系统规划等;
    李笑彤(1993),女,硕士,工程师,研究方向为电力系统规划与分析等;
    梁惠施(1983),女,博士,高级工程师,研究方向为智能配电网规划运行、分布式能源接入、能源大数据挖掘、能源区块链等;
    吕风波(1990),男,硕士,工程师,研究方向为电网规划等;
    席嫣娜(1992),女,硕士,工程师,研究方向为电力系统规划等;
    齐步洋(1993),男,博士,助理研究员,研究方向为电力系统规划运行。
  • 基金资助:
    国网北京市电力公司科技项目“基于价值驱动的两网融合背景下配电网协调规划技术研究项目”(52023419000A)

Distribution Network Expansion Planning Method Based on Air-Conditioning Load Data Mining

SU Ning1, LI Xiaotong1, LIANG Huishi2, GONG Xiaoxu2(), LÜ Fengbo2, XI Yanna1, QI Buyang3   

  1. 1. State Grid Beijing Electric Power Company Economic and Technical Research Institute, Beijing 100055, China
    2. Tsinghua Sichuan Energy Internet Research Institute, Chengdu 610213, China
    3. Department of Electrical Engineering, Tsinghua University, Beijing 100084, China
  • Received:2021-04-22 Online:2021-11-01 Published:2021-11-02
  • Contact: GONG Xiaoxu E-mail:gongxiaoxu01@163.com
  • Supported by:
    Science and Technology Project Fund of State Grid Beijing Electric Power Company(52023419000A)

摘要:

针对配电网规划领域负荷数据挖掘问题,基于实际数据,应用隐马尔科夫模型算法,建立空调负荷挖掘及需求侧响应评估模型,将空调负荷从用电数据中分离,准确描述需求侧响应限值。以配电系统折算到年的总成本最小为目标函数,提出扩展规划模型并进行求解,最后通过IEEE 33节点算例对模型有效性进行验证。结果表明,所提数据挖掘算法能够有效挖掘用户需求侧响应潜力,有利于优化配电网规划方案,降低系统总成本,延缓配电网扩展投资,提高配电系统整体经济性,对配电网规划具有重要的现实意义和作用。

关键词: 数据挖掘, 需求侧响应, 配电网规划

Abstract:

This paper focuses on load data mining in the field of distribution network planning, on the basis of actual data, the model of response evaluation based on air-conditioning load data mining is established, separating air-conditioning load from power consumption data, which can describe the demand response limits accurately, and the distribution system expansion planning model is also proposed and solved by aiming at minimizing the total cost of the distribution system. The validity of the model is verified by an IEEE 33-node example. The results show that the data mining methodology can be used to evaluate the demand response potential, which enables the distribution network planning scheme to be optimized, the total system cost to be reduced, the investment of distribution system expansion to be delayed, and the overall economic efficiency of distribution system to be improved. The proposed method has important practical significance and function for distribution network planning.

Key words: data mining, demand response, distribution network planning

中图分类号: