Xianzhong Duan

Low-voltage distribution systems (LVDS) are prone to leakage current faults, which pose significant electrical safety risks. The wiring complexity, diverse potential causes, and scarce data make identifying and troubleshooting leakage faults in LVDS a persistent challenge in operation and maintenance. A common issue encountered onsite is the incorrect connection of the neutral wire and protective earth (PE) wires within the user’s panel box. Initially, the impact and hazards associated with such wiring faults are analyzed, taking into account the graded residual current protection device and the grounding protection configuration in LVDS. Subsequently, the differential impact on the residual current of the LVDS is examined when wiring faults occur in either a single user or multiple users. A ridge regression-based identification method for detecting users with wiring faults is then proposed, utilizing the time series data of the residual current of the LVDS and the load current of users. Further, considering the impact of complex fault scenarios and the number of users in LVDS, the sliding window technique combined with the Mean-shift cluster analysis algorithm is employed to enhance the accuracy of identifying users with wiring faults. Finally, laboratory data from an LVDS model is used to conduct numerical simulations of wiring fault leakage, verifying the effectiveness and superiority of the proposed method.

Extensive adoption of Information and Communication Technologies makes power systems and communication systems more tightly coupled to form cyber‐physical power systems. It causes power systems to be growingly susceptible to cyber contingencies. Cyber contingencies may sabotage measurement availability, further disrupt the observability analysis in state estimation, and thus threaten the stable operation of power systems. This paper investigates the impact of cyber contingencies on measurement availability. A workflow is presented to achieve accurate and meticulous impact identification under intricate communication architectures with various cyber contingencies. And a set of indicators are proposed to quantitatively evaluate measurement availability. Case studies demonstrate that the proposed method is effective in identifying and assessing the impact of cyber contingencies on measurement availability. Meanwhile, the proposed workflow and indicators can also be employed to dynamically evaluate the system's resilience in protecting measurement transmission against hypothetical cyber contingencies, which could benefit the offline planning and online operation of cyber‐physical power systems