Influence of hybrid fibers on rheological properties of asphalt binders and its prediction with computing techniques
The research investigated the influence of hybrid fibers on rheological properties of asphalt binders and its prediction with computing techniques. First, the morphology and high temperature resistance of steel wool fibers and basalt fibers were evaluated. Then, the asphalt binders with different fibers-asphalt volume ratios and basalt fibers-steel wool fibers volume ratios were prepared, and the rheological properties of asphalt binders were characterized. The statistical analysis including multi-factors analysis of variance and simple-effects analysis were conducted. Finally, non-linear regression, back propagation (BP) neural network with illustration by SHapley Additive exPlanations (SHAP) were employed for the prediction. The results indicated that there was a great difference in morphology between basalt fibers and steel wool fibers, meanwhile steel wool fibers are more stable under high temperature environment. The hybrid fibers were proved contributed for increment of complex modulus and rutting resistance, but it showed a severe segregation between fibers and bitumen as the temperature was raised. The volume content of 6% might be a critical value for the formation of fiber network in statistics. Furthermore, the results of computational evaluation clearly showed that the BP neural network was more capable of accurately calculating the rheological properties of asphalt binders than non-linear regression. The sensitivity analysis by SHAP demonstrated temperature and steel wool fibers content are the more influential inputs for rheological properties. Therefore, the use of hybrid fibers improves in general the rheological properties of asphalt binders, but the proportion needs to be limited.
- Record URL:
- Record URL:
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/issn/09500618
-
Supplemental Notes:
- © 2023 Elsevier Ltd. All rights reserved. Abstract reprinted with permission of Elsevier.
-
Authors:
- Xu, Haiqin
-
0000-0001-8843-6252
- Wu, Shaopeng
- Chen, Anqi
- Li, Zhenpeng
- Zou, Yingxue
-
0000-0002-1213-5821
- Cui, Peide
- Publication Date: 2023-10-19
Language
- English
Media Info
- Media Type: Web
- Features: References;
- Pagination: 132904
-
Serial:
- Construction and Building Materials
- Volume: 401
- Issue Number: 0
- Publisher: Elsevier
- ISSN: 0950-0618
- Serial URL: http://www.sciencedirect.com/science/journal/09500618?sdc=1
Subject/Index Terms
- TRT Terms: Asphalt additives; Backpropagation; Fibers; Neural networks; Predictive models; Rheological properties
- Subject Areas: Data and Information Technology; Highways; Materials; Pavements;
Filing Info
- Accession Number: 01893776
- Record Type: Publication
- Files: TRIS
- Created Date: Sep 20 2023 9:08AM