| 引用本文: | 郭壮志,陈涛,徐其兴,陈璟华.配电网故障区段定位的互补松弛约束新模型与算法[J].电力自动化设备,2020,40(5): |
| GUO Zhuangzhi,CHEN Tao,XU Qixing,CHEN Jinghua.Novel fault section location model for distribution network with complementary relaxation constraints and its algorithm[J].Electric Power Automation Equipment,2020,40(5): |
|
| 摘要: |
| 研究快速高容错性馈线故障区段辨识技术对于提升配电网运行安全可靠性具有重要作用。基于故障电流信号并联叠加特性和逼近关系理论，首次提出代数建模机制下馈线故障区段定位的互补松弛约束模型，其优势为：克服了逻辑模型对群体智能算法的依赖，可采用数值稳定性好的梯度算法求解；利用互补松弛约束条件将非线性整数规划模型转化为具有光滑特性的非线性规划模型，降低了决策求解复杂性；无需对配电网进行分层解耦，即可实现单一和多重故障区段的有效辨识。为有效求解互补松弛故障定位模型，基于模型等价转换思想和松弛因子压缩策略，提出具有全局收敛特性的交互式优化决策算法。仿真算例验证新方法可实现配电网馈线单一或多重故障区段的准确高容错性辨识，且数值稳定性好、故障定位效率高。 |
| 关键词: 配电网 故障区段定位 高容错性 互补松弛约束模型 交互式优化算法 |
| DOI：10.16081/j.epae.202004008 |
| 分类号:TM76 |
| 基金项目:国家自然科学基金青年基金资助项目(51707057)；河南省科技攻关项目(202102210091)；河南省教育厅项目(17A470007) |
|
| Novel fault section location model for distribution network with complementary relaxation constraints and its algorithm |
|
GUO Zhuangzhi1, CHEN Tao1, XU Qixing1, CHEN Jinghua2
|
|
1.School of Electrical Information Engineering, Henan University of Engineering, Zhengzhou 451191, China;2.School of Automation, Guangdong University of Technology, Guangzhou 510006, China
|
| Abstract: |
| Fast and high tolerance feeder fault section location method plays an important role in enhancing safety and reliability of distribution network. Based on parallel superposition characteristics of fault current signals and approximation relation theory, a novel fault section location optimization model with complementary relaxation constraints described by algebraic relation for distribution network is proposed at the first time. The advantages are as follows: the dependence of fault identification logical optimization model on swarm intelligence algorithm is overcame, and it can be solved by gradient algorithm with good numerical stability; the complementary relaxation constraint conditions are employed to transform the nonlinear integer programming model into a nonlinear programming model with smooth characteristics, which reduces the complexity of model decision-making; the single and multiple fault sections can be effectively identified without hierarchical decoupling of distribution network. To solve the novel model effectively, an interactive optimization decision-making algorithm with global convergence is put forward based on model equivalence transformation and relaxation factor compression strategy. The simulative results show that the proposed method can accurately realize single or multiple feeder fault section identification with high fault tolerance for distribution network, and has good numerical stability and high fault location efficiency. |
| Key words: distribution network fault section location high fault tolerance complementary relaxation constraint model interactive optimization algorithm |
