The fatigue life estimation of the porous AlSi9Cu3 alloy based on the classification of pore geometry using multivariate probability distribution density
A method for assessing the fatigue life of porous structures using the strain-life approach (εa-N) is presented. Cellular voids and other forms of porosity in products introduce nonlinear geometric anomalies that result in greater scatter in fatigue life. With the aim of predicting the fatigue life...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-05-01
|
| Series: | Journal of Materials Research and Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2238785425012311 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | A method for assessing the fatigue life of porous structures using the strain-life approach (εa-N) is presented. Cellular voids and other forms of porosity in products introduce nonlinear geometric anomalies that result in greater scatter in fatigue life. With the aim of predicting the fatigue life of such products, specimens with different porosity topologies were first classified by considering the typical shape of the probability density distribution of porosity parameters. Each pore was first described with five geometric parameters. Then, for each test specimen, a mixture of multivariate Gaussian functions was estimated, whereby the interdependence of geometric porosity parameters was statistically modeled. With such statistical processing of data, samples with certain geometric characteristics of pores can be placed in a specific fatigue life group. This is an important step in estimating the fatigue life, as it enables easier predicting of fatigue life for a certain porosity topology. The presented method was successfully tested on experimental samples made of AlSi9Cu3 material. The proposed method offers a significant improvement in the accuracy of modeling fatigue life curves with respect to the spatial topology of porosity because it is leveraging μ-CT imaging for quantitative macrostructural characterization with statistical analysis. This approach enables robust data extraction and is suited for integration into industrial-scale manufacturing and quality control workflows. |
|---|---|
| ISSN: | 2238-7854 |