Temporal-Aware Chain-of-Thought Reasoning for Vibration-Based Pump Fault Diagnosis
Shranjeno v:
| izdano v: | Processes vol. 13, no. 8 (2025), p. 2624-2647 |
|---|---|
| Glavni avtor: | |
| Drugi avtorji: | , , |
| Izdano: |
MDPI AG
|
| Teme: | |
| Online dostop: | Citation/Abstract Full Text + Graphics Full Text - PDF |
| Oznake: |
Brez oznak, prvi označite!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3244057786 | ||
| 003 | UK-CbPIL | ||
| 022 | |a 2227-9717 | ||
| 024 | 7 | |a 10.3390/pr13082624 |2 doi | |
| 035 | |a 3244057786 | ||
| 045 | 2 | |b d20250101 |b d20251231 | |
| 084 | |a 231553 |2 nlm | ||
| 100 | 1 | |a Zeng Jinchao |u College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou 325035, China; zeng18779777724@163.com (J.Z.); li13045765073@163.com (Z.L.) | |
| 245 | 1 | |a Temporal-Aware Chain-of-Thought Reasoning for Vibration-Based Pump Fault Diagnosis | |
| 260 | |b MDPI AG |c 2025 | ||
| 513 | |a Journal Article | ||
| 520 | 3 | |a Industrial pump systems require real-time fault diagnosis for predictive maintenance, but conventional Chain-of-Thought (COT) reasoning faces computational bottlenecks when processing high-frequency vibration data. This paper proposes Vibration-Aware COT (VA-COT), a novel framework that integrates multi-domain feature fusion (time, frequency, time–frequency) with adaptive reasoning depth control. Key innovations involve expert prior-guided dynamic feature selection to optimize edge-device inputs, complexity-aware reasoning chains reducing computational steps by 40–65% through confidence-based early termination, and lightweight deployment on industrial ARM-based single-board computers (SBCs). Evaluated on a 12-class pump fault dataset (5400 samples from centrifugal/gear pumps), VA-COT achieves 93.2% accuracy surpassing standard COT (89.3%) and CNN–LSTM (Convolutional Neural Network-Long Short-Term Memory network) (91.2%), while cutting latency to <1.1 s and memory usage by 65%. Six-month validation at pump manufacturing facilities demonstrated 35% maintenance cost reduction and 98% faster diagnostics versus manual methods, proving its viability for IIoT (Industrial Internet of Things) deployment. | |
| 653 | |a Accuracy | ||
| 653 | |a Deep learning | ||
| 653 | |a Wavelet transforms | ||
| 653 | |a Artificial neural networks | ||
| 653 | |a Real time | ||
| 653 | |a Signal processing | ||
| 653 | |a Computer applications | ||
| 653 | |a Gear pumps | ||
| 653 | |a Long short-term memory | ||
| 653 | |a Vibration analysis | ||
| 653 | |a Machine learning | ||
| 653 | |a Fault diagnosis | ||
| 653 | |a Maintenance costs | ||
| 653 | |a Fourier transforms | ||
| 653 | |a Neural networks | ||
| 653 | |a Process controls | ||
| 653 | |a Reasoning | ||
| 653 | |a Support vector machines | ||
| 653 | |a Network latency | ||
| 653 | |a Industrial applications | ||
| 653 | |a Computers | ||
| 653 | |a Vibration | ||
| 653 | |a Latency | ||
| 653 | |a Predictive maintenance | ||
| 653 | |a Industrial Internet of Things | ||
| 700 | 1 | |a Li, Zicheng |u College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou 325035, China; zeng18779777724@163.com (J.Z.); li13045765073@163.com (Z.L.) | |
| 700 | 1 | |a Zheng Zuopeng |u Zhejiang TONGLI Transmission Technology Co., Ltd., Ruian 325205, China | |
| 700 | 1 | |a Lin Qizhe |u College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou 325035, China; zeng18779777724@163.com (J.Z.); li13045765073@163.com (Z.L.) | |
| 773 | 0 | |t Processes |g vol. 13, no. 8 (2025), p. 2624-2647 | |
| 786 | 0 | |d ProQuest |t Materials Science Database | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3244057786/abstract/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text + Graphics |u https://www.proquest.com/docview/3244057786/fulltextwithgraphics/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text - PDF |u https://www.proquest.com/docview/3244057786/fulltextPDF/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |