Computing low-frequency vibration energy with Hölder singularities as durability predictive criterion of random road excitation
| Autor: | Shahrum Abdullah, Dieter Schramm, C. H. Chin, Ahmad Kamal Ariffin, Salvinder Singh Karam Singh |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
0209 industrial biotechnology
business.industry 02 engineering and technology Structural engineering Durability Signal Coil spring Theoretical Computer Science Power (physics) Vibration 020901 industrial engineering & automation Maschinenbau 0202 electrical engineering electronic engineering information engineering Range (statistics) 020201 artificial intelligence & image processing Geometry and Topology business Suspension (vehicle) Software Energy (signal processing) Mathematics |
| Popis: | This study aims to compute low-frequency energy with Holder singularities in vibration signals of suspension system to predict the durability of coil spring. High frequencies in vibrations often had minimal contribution towards fatigue damage due to low amplitude range and thus induce errors in energy analysis of vibration signals. Since traditional low-pass method had not only been ineffective in reducing high frequencies, it also resulted in the loss of signal information. This study had therefore proposed characterising low-frequency energy for road excitations using Holder singularities and power spectral analyses. Singularities and low-frequency energy of road vibration signals would first be identified through Holder local regularity analysis. This was then followed by fatigue life prediction using the strain-life approaches (i.e. Coffin-Manson, Morrow and Smith–Watson–Topper models). The energy-based fatigue life prediction models had not only shown good fit with R2 values higher than 0.8, but had also demonstrated an accurate prediction of fatigue life with more than 95% of the data being within the acceptance boundary. The Morrow-based model provided the highest accuracy in fatigue life prediction because of its highest R2 value of 0.8625 and 100% data survival in the fatigue life correlation study. This showed that energy-based fatigue life prediction models provide an accurate and effective prediction of the durability performance. This study proposed a more precise energy characterisation method for energy-based durability prediction of suspension coil spring under random loading conditions. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|