引用本文: | 李 春,赵玉勇,吴方舟,等.基于高频谐振识别的光伏直流汇集支路故障定位[J].电力系统保护与控制,2021,49(15):150-158.[点击复制] |
LI Chun,ZHAO Yuyong,WU Fangzhou,et al.Fault location for photovoltaic DC collecting branch based on high frequency resonance identification[J].Power System Protection and Control,2021,49(15):150-158[点击复制] |
|
摘要: |
光伏柔性直流汇集支路发生直流故障时,呈现快速变化的暂态过程,传统基于固定特征频率阻抗量测的故障定位方法难以适用。因此,亟待研究快速可靠的故障定位方法,实现光伏直流汇集并网系统故障快速恢复。提出一种基于高频谐振识别的双极短路故障定位方法。利用直流故障极间电压的阶跃特征和直流变压器(DC transformer, DCT)出口并联电容与限流电抗器构成的串联谐振,得出高频谐振电流与故障距离的关系。通过连续小波变换(Continuous Wavelet Transform, CWT)提取故障谐振电流,计算出故障距离,最终完成故障定位。与传统故障定位方法相比,该方法无需额外高频注入源,不受换流器状态和控制系统影响。PSCAD仿真结果表明所提方法只需要故障前后2.5 ms的故障数据就可以进行可靠的故障测距,所需数据窗较短,满足光伏直流汇集系统精确定位故障与快速恢复运行的要求,并且具有较高的耐受过渡电阻、噪声和分布电容能力。该方法为现有直流系统保护设计提供新的可选方法。 |
关键词: 光伏直流汇集系统 高频谐振 谐振电流 小波变换 故障定位 |
DOI:DOI: 10.19783/j.cnki.pspc.201201 |
投稿时间:2020-10-04修订日期:2020-11-27 |
基金项目:国家电网公司科技项目资助“交直流混合配电网的故障特性与保护研究”(2019-LHKJ-017) |
|
Fault location for photovoltaic DC collecting branch based on high frequency resonance identification |
LI Chun1,ZHAO Yuyong2,WU Fangzhou2,JIA Ke3,SHI Zhiming3,TANG Huaguo4 |
(1. Jiaxing Power Supply Company, State Grid Zhejiang Electric Power Co., Ltd., Jiaxing 314000, China; 2. Haining Power
Supply Company, State Grid Zhejiang Electric Power Co., Ltd., Haining 314400, China; 3. State Key Laboratory of
Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University,
Beijing 102206, China; 4. Haining Jinneng Industrial Co., Ltd., Haining 314400, China) |
Abstract: |
When a DC fault occurs in a flexible photovoltaic DC collection branch, the transient process is rapidly changing. Thus it is difficult to apply traditional fault location methods based on fixed characteristic frequency impedance measurement. Therefore, it is urgent to study fast and reliable fault location methods to quickly achieve photovoltaic DC collection system recovery. This paper proposes a bipolar short-circuit fault location method based on high-frequency resonance identification. It uses the step characteristics of the DC fault pole-to-pole voltage and the series resonance of a DC Transformer (DCT) outlet parallel capacitor and a current-limiting reactor. The relationship between the resonance current and the fault distance is extracted by Continuous Wavelet Transform (CWT), the fault distance is calculated, and the fault location is finally completed. Compared with the traditional fault location method, this method does not require additional high-frequency injection sources and is not affected by the state of the converter and the control system. PSCAD simulation results show that the proposed method only needs 2.5 ms of fault data before and after the fault to perform reliable fault ranging. The required data window is short, which meets the requirements of photovoltaic DC assembly to accurately locate faults and quickly resume operation, and the method is capable of withstanding transient resistance, noise, and distributed capacitance. This method provides a new alternative to existing DC system protection designs.
This work is supported by the Science and Technology Project of State Grid Corporation of China “Research on Fault Characteristics and Protection of Hybrid AC/DC Distribution Network” (No. 2019-LHKJ-017). |
Key words: photovoltaic DC pooling system high-frequency resonance resonant current wavelet transform fault location |