基于改进LMD和能量相对熵的主动配电网故障定位方法

针对配电网发生单相接地故障且分布式电源（distributed generations, DGs）大量接入后，配电网结构和运行方式复杂多变、故障后电气量不明显、故障特征弱等特点，提出一种基于改进局部均值分解（improved local mean decomposition, ILMD）和能量相对熵的主动配电网故障定位方法。首先，利用镜像延拓将信号在两端延拓，消除LMD存在的端点效应，同时，在信号中加入自适应噪声，克服LMD存在的模态混叠问题，对各区段暂态零序电流进行ILMD分解；然后对分解后所有乘积函数（product function, PF）的能量相对熵进行计算，所有PF分量的相对能量熵之和即为区段间的相对能量熵，通过所设置的定位判据，对故障区段进行判断。仿真结果表明，所提算法在10 kV小电流接地系统和改进的IEEE33节点系统中，在不同的仿真条件下故障定位结果准确率高，验证了所提方法的准确性。

Fault Location for Active Distribution Network Based on Improved LMD and Energy Relative Entropy

In view of the characteristics of distribution network with large-scale integration of distributed generation when single-phase grounding fault occurs, such as complex structure and operation mode, obscure electrical quantity after fault, weak fault characteristics and so on, this paper proposes a new fault location method for active distribution network based on improved local mean decomposition (ILMD) and energy relative entropy. Firstly, the mirror extension is used to extend the signals at both ends to eliminate the endpoint effect of LMD. At the same time, the adaptive noise is added to the signal to overcome the mode aliasing problem of LMD, and the ILMD is used to decompose the transient zero sequence current of each section. Then the energy relative entropy of all product functions (PF) after decomposition is calculated. The sum of relative energy entropy of all PF components is the relative energy entropy between sections. The fault section is judged by the positioning criteria set in this paper. Simulation results show that the proposed algorithm has high accuracy in fault location for 10 kV small current grounding system and improved IEEE 33 node system under different simulation conditions, which verifies the accuracy of the proposed method.