Cargando…
A Phenomenological Primary–Secondary–Tertiary Creep Model for Polymer-Bonded Composite Materials
This study develops a unified phenomenological creep model for polymer-bonded composite materials, allowing for predicting the creep behavior in the three creep stages, namely the primary, the secondary, and the tertiary stages under sustained compressive stresses. Creep testing is performed using m...
Autores principales: | , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8309501/ https://www.ncbi.nlm.nih.gov/pubmed/34301110 http://dx.doi.org/10.3390/polym13142353 |
Sumario: | This study develops a unified phenomenological creep model for polymer-bonded composite materials, allowing for predicting the creep behavior in the three creep stages, namely the primary, the secondary, and the tertiary stages under sustained compressive stresses. Creep testing is performed using material specimens under several conditions with a temperature range of 20 °C–50 °C and a compressive stress range of 15 MPa–25 MPa. The testing data reveal that the strain rate–time response exhibits the transient, steady, and unstable stages under each of the testing conditions. A rational function-based creep rate equation is proposed to describe the full creep behavior under each of the testing conditions. By further correlating the resulting model parameters with temperature and stress and developing a Larson–Miller parameter-based rupture time prediction model, a unified phenomenological model is established. An independent validation dataset and third-party testing data are used to verify the effectiveness and accuracy of the proposed model. The performance of the proposed model is compared with that of an existing reference model. The verification and comparison results show that the model can describe all the three stages of the creep process, and the proposed model outperforms the reference model by yielding 28.5% smaller root mean squared errors on average. |
---|