引用本文:杨丽君,曹玉洁,何新,王伟浩,王晨.多代理系统下的主动配电网故障恢复博弈策略[J].控制理论与应用,2018,35(5):699~708.[点击复制]
YANG Li-jun,CAO Yu-jie,HE Xin,WANG Wei-hao,WANG Chen.Active distribution network service restoration using multiagent systems and game theory[J].Control Theory and Technology,2018,35(5):699~708.[点击复制]
多代理系统下的主动配电网故障恢复博弈策略
Active distribution network service restoration using multiagent systems and game theory
摘要点击 2046  全文点击 1219  投稿时间：2017-09-25  修订日期：2018-02-26
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2018.70704
  2018,35(5):699-708
中文关键词  主动配电网  故障恢复  多代理系统  博弈论  利益均衡
英文关键词  active distribution network  service restoration  multiagent systems  game theory  interest balance
基金项目  国家自然科学基金项目(61573302)资助.
作者单位邮编
杨丽君 电力电子节能与传动控制河北省重点实验室燕山大学 066004
曹玉洁* 电力电子节能与传动控制河北省重点实验室燕山大学 066004
何新 国网冀北电力有限公司 
王伟浩 电力电子节能与传动控制河北省重点实验室燕山大学 
王晨 电力电子节能与传动控制河北省重点实验室燕山大学 
中文摘要
      在保障主动配电网可靠恢复前提下, 为实现供电侧与用户侧利益均衡, 本文提出了多代理系统(multiagent system, MAS)下的主动配电网故障恢复博弈策略. 为充分考虑电网和电力用户对故障恢复的决策影响, 设计了由电 网代理、用户总代理和协调代理构成的MAS, 建立了MAS信息传输模型. 然后, 构建了供电侧和用户侧在故障恢复 中的利益函数、恢复策略空间, 以及以电网代理和用户总代理作为参与人的合作博弈恢复模型. 电网代理和用户总 代理分别以改进的蚁群算法和统计方法进行分布并行计算, 协调代理以双方共同的利益函数作为寻优目标通过迭 代算法求得满足纳什均衡的恢复策略. 本文以IEEE69节点模型为例, 分别对单故障和连锁故障恢复情景求解, 验证 了本文所提策略的有效性.
英文摘要
      On the premise of ensuring the reliable service restoration of the active distribution network and in order to realize the interest balance between the power supply side and the user side, a game strategy for active distribution network service restoration using multiagent systems (MAS) is proposed. In order to consider the influence of the power grid and power consumers to service restoration, an MAS composed of the power grid agent, the user general agent and the coordination agent is designed, and an information transmission model of MAS is built. Then, the interest functions of the power supply side and the user side in the recovery processes, the space of recovery strategies, and a cooperative game recovery model taking the power grid agent and the user general agent as participants are established. The power grid agent and the user general agent are performed by distributed parallel computing with improved ant colony algorithm and the statistical method respectively. The coordination agent takes the common interest function of both sides as the optimization objective, and the recovery scheme satisfying the Nash equilibrium is obtained by iteration algorithm. This paper takes IEEE69 node model as an example to solve single fault and cascading failures, and the effectiveness of the proposed strategy is verified.