Abstract:
Power system stability analysis is in transition towards a data-driven paradigm. However, in the context of digital transformation, the traditional centralized approach t...Show MoreMetadata
Abstract:
Power system stability analysis is in transition towards a data-driven paradigm. However, in the context of digital transformation, the traditional centralized approach to measurement data processing faces significant challenges, including heavy computational and communication burdens, as well as data privacy concerns. Driven by this motivation, this paper proposes a power system topology-based edge computing framework that leverages the computation and communication capabilities of Intelligent Electronic Devices (IEDs) to reduce the burden on the control center and facilitate the digital transition. In the proposed framework, the IEDs equipped on generators serve as edge nodes (ENs). Each EN collects the measurement data in its local zone to identify the dynamic model of the subsystem within the zone, and the identification results of each pair of adjacent subsystems are merged. After several rounds of merging, the dynamic model of the entire power system is obtained. The obtained model is stored distributedly in the ENs, enabling parallel processing. Eigen-analysis is then performed on the dynamic model using a divide-and-conquer strategy, recursively splitting the computational task into two subtasks executed by separate EN groups. Case studies demonstrate the effectiveness of the proposed framework and highlight its benefits.
Published in: IEEE Transactions on Power Systems ( Early Access )