Analysis and prediction of surface crack growth under fatigue loading
This research uses several fatigue crack growth models to examine the cyclic evolution of fatigue cracks in a shaft. Three fatigue crack growth models are used to forecast crack growth: Walker, Paris Law, and others. Experimental data support these models. The main problem is accurately estimating...
| Main Authors: | , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
IOP Publishing
2025
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/45014/ http://umpir.ump.edu.my/id/eprint/45014/1/Published-Analysis-and-Prediction-of-Surface-Crack-Growth-under-Fatigue-LoadingJournal-of-Physics-Conference-Series.pdf |
| Summary: | This research uses several fatigue crack growth models to examine the cyclic evolution of fatigue cracks in a shaft. Three fatigue crack growth models are used to forecast crack growth: Walker, Paris Law, and others. Experimental data
support these models. The main problem is accurately estimating the propagation of fractures in shafts under cyclic loads because the existing models frequently exhibit variations in real-world applications that could lead to failures. This study compares the experimental results with model predictions to assess the accuracy of several models and improve our understanding of fatigue crack behaviour in
practical settings. The experimental approach for 4 point-bending is compared with the simulation result, including boundary conditions and material properties. Paris's and Walker's fatigue crack growth models are employed in the S-version Finite Element Model (S-FEM) to simulate the 4 point-bending models' analysis. The surface fatigue crack growth prediction is simulated and compared with the experimental results. The prediction beach marks of crack depth are slightly similar to the experimental results. Moreover, the prediction beach marks of crack length differ from the experimental results. The crack closure effect influences the
difference between the experimental results. In summary, no single model is perfect in general; the selection is based on the particular circumstances and characteristics of the material. This work seeks to help engineers select the best
model by improving prediction tools for maintaining mechanical components and increasing safety and performance in engineering applications. |
|---|