基于多源信息融合的配电网故障区段定位方法
Distribution Network Fault Section Location Method Based on Multi-Source Information Fusion

针对单一信息源易受故障类型、测点配置及开关量异常影响，导致配电网故障区段定位不稳定的问题，提出一种融合电气量故障路径识别与状态量可信诊断的定位方法。对于大电流故障，依据测点过流状态和拓扑关系确定故障电流路径；对于小电流接地故障，利用零序电流首波方向识别故障路径，并通过Teager能量算子提取初始暂态能量分布，计算候选区段的电气量支持度。基于网络拓扑和保护逻辑构建状态量期望动作集合，综合保护启动、故障指示、开关跳闸及隔离动作等SOE信息，引入时序可信度，通过贝叶斯诊断计算候选区段故障概率。将两类诊断结果转换为基本概率分配，并采用D-S证据理论进行融合。PSCAD/EMTDC仿真结果表明，该方法能够准确定位不同区段故障，且在分支点量测不足以及状态量通信丢失、误动作等情况下仍具有较高的定位可靠性。
To address the instability of fault-section localization in distribution networks caused by the susceptibility of single-source information to fault types, measurement-point configurations, and abnormal switching-status data, a localization method integrating electrical-quantity-based fault-path identification with credibility-aware status-data diagnosis is proposed. For high-current faults, the fault-current path is determined according to the overcurrent states of measurement points and the network topology. For small-current grounding faults, the fault path is identified using the initial-wave directions of zero-sequence currents, while the Teager energy operator is employed to extract the initial transient-energy distribution and calculate the electrical-quantity support degree of each candidate section. Expected status-action sets are constructed based on the network topology and protection logic. Sequence-of-events (SOE) information, including protection pickup, fault indication, circuit-breaker tripping, and isolation actions, is comprehensively considered. Temporal credibility is introduced into Bayesian diagnosis to calculate the fault probability of each candidate section. The two diagnostic results are then converted into basic probability assignments and fused using Dempster-Shafer evidence theory. PSCAD/EMTDC simulation results demonstrate that the proposed method can accurately locate faults in different sections and maintain high localization reliability under insufficient measurements at branching points and status-data anomalies such as communication loss and maloperation.