| 引用本文: | 杨睿,高红均,刘俊勇.基于有限PMU配置的配电网故障定位[J].电力自动化设备,2022,42(4): |
| YANG Rui,GAO Hongjun,LIU Junyong.Distribution network fault location based on limited PMU configuration[J].Electric Power Automation Equipment,2022,42(4): |
|
| 摘要: |
| 提出一种仅需要有限的相量测量单元(PMU)便能实现单故障和多重故障定位的配电网故障定位方法。首先通过叠加原理和等效变换推导故障网络节点电压方程，基于此建立压缩感知模型。然后利用[?1]正则化最小二乘法求解近似故障电流并进行归一化处理，确定疑似故障区间。通过电压残差公式和故障测距公式依次计算可能故障数目下对应的电压残差和故障距离，根据计算结果判定真实故障区间和故障位置。IEEE 33节点配电系统仿真实验表明：在有限PMU量测信息下，所提方法受故障类型、过渡电阻影响较小，适用于单故障和多重故障的场景，且具有一定的抗噪能力。 |
| 关键词: 配电网 相量测量单元 多重故障定位 压缩感知 [ℓ1]正则化最小二乘法 |
| DOI：10.16081/j.epae.202112033 |
| 分类号:TM72 |
| 基金项目:国家重点研发计划项目(2019YFE0123600)；国家自然科学基金资助项目(52077146)；中国电机工程学会青年人才托举工程项目(CESS-YESS-2019027) |
|
| Distribution network fault location based on limited PMU configuration |
|
YANG Rui, GAO Hongjun, LIU Junyong
|
|
College of Electrical Engineering, Sichuan University, Chengdu 610065, China
|
| Abstract: |
| A distribution network fault location method is proposed, which only needs limited PMU(Phasor Measurement Unit) to locate single faults and multiple faults. Firstly, the voltage equation of fault network node is derived by superposition principle and equivalent transformation, based on which, the compressed sensing model is established. Then, the approximate fault current is solved by [?1] regularized least square method and then normalized to determine the suspected fault interval. Through the voltage residual formula and fault location formula, the corresponding voltage residual and fault distance under the number of possible faults are calculated, and the real fault interval and fault location are determined according to the calculation results. The simulative results of IEEE 33-bus distribution system show that under the limited PMU measurement information, the proposed method is slightly affected by the fault type and transition resistance, it is suitable for scenarios of single faults and multiple faults, and has certain anti-noise ability. |
| Key words: distribution network phasor measurement unit multiple fault location compressed sensing [ℓ1] regularized least square method |
