A real-time serial fuzzy-based model for the simulation of the mechanical behaviours of gamma irradiated 3D printed specimens
Gamma irradiation has been employed in various areas. The effects of ionizing radiation on thermal, mechanical and chemical properties need to be carefully investigated and modelled, particularly, for 3D printed specimens which have been extensively utilized nowadays. However, these properties need...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Results in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590123025025587 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Gamma irradiation has been employed in various areas. The effects of ionizing radiation on thermal, mechanical and chemical properties need to be carefully investigated and modelled, particularly, for 3D printed specimens which have been extensively utilized nowadays. However, these properties need to be examined in real-time, this being due to the post-irradiation impacts resulted from free radicals. Therefore, this paper proposes a novel serial fuzzy-based framework. The idea of this framework stems from (i) the need to monitor the properties on real-time bases; (ii) the dynamic behaviour of the 3D printing processes; (iii) that data may not be sufficient at early stages; and (iv) the need to investigate the model behaviours in the spaces. To develop such a model, a serial framework that consists of type-1 fuzzy logic systems is designed to map the irradiation and 3D printing parameters to the mechanical behaviours of 3D printed specimens. The predictive performance is then evaluated in the spaces. Then, the serial model is expanded in the spaces examined in a way that allows the online prediction and, thus, monitoring of the changes in the mechanical behaviours. Validated on two experimental data sets for six polymers, the proposed dynamic framework has significantly improved the predictive performance by an overall percentage of 13.7 % in the coefficient of determination. Moreover, the proposed framework outperformed artificial neural networks and regression models developed to map the irradiation and 3D printing parameters to the mechanical behaviours of 3D printed specimens with an average overall improvement of 29.6 %. |
|---|---|
| ISSN: | 2590-1230 |