A passive islanding detection method using deep neural bidirectional LSTM-CNN
Hakan Ozturk, Secil Varbak Nese · 2026
Read A passive islanding detection method using deep neural bidirectional LSTM-CNN by Hakan Ozturk, Secil Varbak Nese (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13473564.txt.
Description
Ozturk and Varbak Nese trained a deep neural network to catch islanding, the dangerous state where a local microgrid keeps energizing lines the main grid has abandoned. International rules give detectors two seconds to react; their 2026 design, a convolutional network feeding a bidirectional memory network, answers far faster on the IEC standard test system.
The detector flags an island in just 10 milliseconds, two hundred times inside the allowed window, and its non-detection zone, the blind spot of power mismatches it cannot see, shrank to 0.02% of cases in testing. The training set held 1,761 scenarios spanning power mismatches of plus or minus 40% and five kinds of innocent disturbances, from lightning-type faults to capacitor switching. No earlier passive method, the authors note, broke 20 milliseconds with so small a blind spot.
Catalog record
| CREATOR | Hakan Ozturk, Secil Varbak Nese |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 77 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |