A Novel Hybrid Statistical Method for Power Swing Detection in Transmission Lines With Grid-Following Inverter

This paper presents a new statistical approach for reliably detecting power swings and faults in transmission lines that include grid-following (GFL) inverters. The method combines the strengths of the cumulative sum control chart (CUSUM) and entropy analysis (EA), using CUSUM's ability to quickly pick up on signal changes and EA's insight into system complexity. To test the proposed method, simulations were carried out in PSCAD/MATLAB on a modified IEEE 14-bus system with a GFL-type inverter-based resource. The results show that the method can effectively detect both faults and power swings, even in the presence of additive white Gaussian noise (signal-to-noise ratio = 10 dB). It also accurately distinguishes between different transient events, such as load switching, generator trips, and capacitor bank switching, with a 100% success rate in identifying non-fault conditions. Furthermore, it consistently detects faults across a range of fault resistances (0 to 10 Ω) with perfect accuracy. Compared to existing techniques, this approach performs better in systems that integrate GFL inverters and offers a more efficient solution with lower computational requirements for power system protection.