SpectralTCN: A multi-resolution wavelet gating network for vibration-based structural damage detection

Quan Pham Hong, Trung Vu Manh, Bich Nguyen Thach, Le Nguyen Dan, Hoa Tran Ngoc · 2026

Read SpectralTCN: A multi-resolution wavelet gating network for vibration-based structural damage detection by Quan Pham Hong, Trung Vu Manh, Bich Nguyen Thach, Le Nguyen Dan, Hoa Tran Ngoc (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/plos-pone-0358224.txt.

Description

Quan Pham Hong leads a team of five designing a neural network that listens for damage in structures, in PLOS ONE in 2026. Bridges and towers betray cracks through subtle changes in how they shake. The warning signs are brief blips in narrow frequency bands. Models that read only the time signal miss the frequency story; models that transform the whole signal lose track of when each frequency occurred. Their network, SpectralTCN, is built to keep both.

Inside each layer of a temporal convolutional network sits a wavelet gating module. It splits the signal into sub bands that correspond to vibration modes. The network then decides how much of each band to keep. Then it rebuilds the signal before moving on. Because wavelets preserve timing and frequency together, a short damage signature keeps its moment and its pitch. The design targets inspection of structures that cannot afford missed hints.

Catalog record

CREATORQuan Pham Hong, Trung Vu Manh, Bich Nguyen Thach, Le Nguyen Dan, Hoa Tran Ngoc
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE81613
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0358224

Browse the full library · About the Tower of Babel