A One-Dimensional Depthwise Separable Convolutional Neural Network for Bearing Fault Diagnosis Implemented on FPGA
Yu-Pei Liang, Hao Chen, Ching-Che Chung, Pavlos Lazaridis · 2024
Read A One-Dimensional Depthwise Separable Convolutional Neural Network for Bearing Fault Diagnosis Implemented on FPGA by Yu-Pei Liang, Hao Chen, Ching-Che Chung, Pavlos Lazaridis (2024) 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-11644920.txt.
Description
A neural network small enough to live on a chip can still hear a failing bearing, report Yu-Pei Liang and co-authors in a 2024 hardware paper. Their design squeezes a one dimensional convolutional network with separable convolutions and aggressive number rounding, then runs it on a standard programmable chip board.
Fed raw current signals from a public bearing dataset, the system reached 96 percent accuracy in telling fault types apart. The whole model fits in about 22 kilobytes and sips half a watt. The authors aim the work at factory floors, where diagnosis must happen on the machine, not in a data center.
Catalog record
| CREATOR | Yu-Pei Liang, Hao Chen, Ching-Che Chung, Pavlos Lazaridis |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |