Waveform-Prediction Augmentation and Deep Manifold Learning Enable Imbalanced Fault Diagnosis in Rotating Machinery
Yu Tian, Shunsheng Guo, Yibing Li, Xin Zhang, Li Jiang · 2026
Read Waveform-Prediction Augmentation and Deep Manifold Learning Enable Imbalanced Fault Diagnosis in Rotating Machinery by Yu Tian, Shunsheng Guo, Yibing Li, Xin Zhang, Li Jiang (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-13568344.txt.
Description
Yu Tian and four co-authors fix a common failure of machine fault diagnosis: the faults that matter most are the rarest in the training data, in a 2026 paper. Models trained on lopsided vibration data learn the healthy and common faults and miss the rare ones. Raw signals are also high-dimensional and redundant.
Their pipeline attacks both problems. A forecasting neural network extends minority fault signals, rebalancing the dataset with synthetic waveforms. A stacked autoencoder then compresses signals into deep features. A manifold learning step, guided by a time-warping distance that respects waveform shape, reduces those features while keeping neighborhood structure intact. A simple nearest-neighbor classifier makes the final call. On bearing and gearbox datasets with ten-to-one imbalance, the method improved diagnosis markedly over the baselines tested.
Catalog record
| CREATOR | Yu Tian, Shunsheng Guo, Yibing Li, Xin Zhang, Li Jiang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 57 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |