中国电力 ›› 2022, Vol. 55 ›› Issue (6): 33-41.DOI: 10.11930/j.issn.1004-9649.202007222

• 电网调度模型研究 • 上一篇    下一篇

基于贪婪算法的配网多阶段快速重构新方法

赵胜举1, 么莉1, 林济铿2, 张旭1   

  1. 1. 天津大学 电气自动化与信息工程学院,天津 300072;
    2. 同济大学 电子与信息工程学院,上海 201804
  • 收稿日期:2020-08-07 修回日期:2021-02-18 出版日期:2022-06-28 发布日期:2022-06-18
  • 作者简介:赵胜举(1995—),男,硕士,从事配网自动化研究,E-mail:1432080607@qq.com;林济铿(1967—),男,通信作者,教授,从事系统稳定性分析及控制、智能电网研究,E-mail:mejklin@126.com
  • 基金资助:
    国家自然科学基金资助项目(基于图论的智能电网电力孤岛的模型和优化,51177107)

Fast Multi-stage Reconfiguration Method for Distribution System Based on Greedy Algorithm

ZHAO Shengju1, YAO Li1, LIN Jikeng2, ZHANG Xu1   

  1. 1. School of Electronic Information Engineering, Tianjin University, Tianjin 300072, China;
    2. School of Electrical Engineering, Tongji University, Shanghai 201804, China
  • Received:2020-08-07 Revised:2021-02-18 Online:2022-06-28 Published:2022-06-18
  • Supported by:
    This work is supported by National Natural Science Foundation of China (Model and Optimization of Power Island in Smart Grid Based on Graph Theory, No.51177107)

摘要: 针对多阶段动态重构计算速度慢的问题,提出一种动态重构新方法。该方法基于给定的寻优尺度参数M,对以开关动作次数最多的两个相继阶段为中心的2M+2个阶段进行各种可能合并,分别对各个合并后的新阶段进行静态重构,然后选择目标函数值最小的合并方案为当前最优合并方案;如此反复迭代,直到目标函数值不再减少且当前解为可行解时,即得到近似最优解。方法在获得与全局最优解近似解的前提下,其计算速度都比全局最优解快近2个数量级。算例表明该方法因其求解的高效性而具有较好的工程应用前景。

关键词: 配网重构, 最优流法, Mayeda生成树, 贪婪算法, 多阶段动态重构

Abstract: Aiming at the slow-speed computation problem of the multi-stage dynamic reconfiguration, a new dynamic reconfiguration method is proposed. Based on the given optimization scale parameter M, the proposed method makes all possible merges for 2M+2 phases which are centered on the two successive phases with the highest number of switching actions, and carries out static reconfiguration for each new merged phase, and then selects the merging scheme with a minimum objective function value as the current optimal merging scheme; repeat the iteration until the objective function value is no longer reduced and the current solution is a feasible solution, thereby obtaining the approximate optimal solution. On the premise of obtaining an approximate solution to the global optimal solution, the calculation speed is nearly 2 orders of magnitude faster than the global optimal solution. Case study shows that the proposed method has good engineering application prospect because of its high efficiency in solution.

Key words: distribution network reconfiguration, optimal flow method, Mayeda spanning tree, greedy algorithm, multi-stage dynamic reconfiguration