基于高频纹波的DWT-RF故障定位方法  

DWT-RF Fault Location Method Based on High Frequency Ripple

作　　者：秦艺晗 张晋涛[2] 孙志锋[1] 马风力 

机构地区：[1]浙江大学电气工程学院,浙江杭州310027 [2]杭州华辰电力控制工程有限公司,浙江杭州311122

出　　处：《工业控制计算机》2022年第3期147-149,152,共4页Industrial Control Computer

摘　　要：为了解决光伏系统单相接地故障下的故障定位问题,提出一种基于电路中高频信号的DWT-RF故障定位方法。利用离散小波变换对含有高频纹波的信号进行分解,得到一系列特征数据。将故障定位问题视作单标签多分类问题,在建模时,在每种故障模式下训练0-1分类的RF模型,再将若干子模型结合起来。通过调优阈值法构建决策层,利用子模型的输出判定故障位置。使用PSCAD建立仿真模型构造故障模式,并获取不同故障模式下的数据,通过实验数据分析及模型对比,验证了该方法在集电线路故障定位问题上有较好效果。In order to solve the problem of fault location under the single-phase ground fault of photovoltaic system,a DWT-RF fault location method based on the high frequency signal in the circuit is proposed in this paper.DWT is used to decompose high-frequency ripple to obtain a series of characteristic data.The fault location problem is regarded as a single-label multi-classification problem.When modeling,a 0-1 classification RF model is trained under each fault mode,and then several sub-models are combined.The decision-making layer is constructed by the optimal threshold method,and the output of the sub-model is used to determine the location of the fault.Using PSCAD to establish a simulation model to construct a failure mode,through the analysis of experimental data,it is verified that the method has a good effect on the problem of fault location of the collection line.

关 键 词：故障定位 高频纹波 离散小波变换 集成学习 

分 类 号：TM615[电气工程—电力系统及自动化]