Methodology for error propagation analysis of the complex stiffness modulus of asphalt mixes. (5th July 2021)

• Record Type: Journal Article
• Authors: Mikowski, Alexandre; Casali, Rafael Machado; Soares, Paulo; da Silva, Wyllian Bezerra; Barra, Breno Salgado
• Is Part Of: Construction & building materials. Volume 290(2021)
• Publication Date: 2021-07-05
• Subjects: Error propagation -- Propagation of uncertainty -- Huet-Sayegh rheological model -- Complex stiffness modulus -- Asphalt mix; Building materials -- Periodicals; 624.18
• DOI: 10.1016/j.conbuildmat.2021.123156
• Languages: English
• ISSNs: 0950-0618

Abstract

Highlights: A mathematical methodology for analyzing the error propagation is developed. Experimental complex stiffness modulus of an asphalt mix is taken as data base. Huet-Sayegh rheological model is parameterized by computerized tool. Partial derivatives discretize the influence of each complex modulus variable. The methodology developed fits the rheological model and the experimental data. Abstract: This paper reports on the development of a mathematical methodology for analyzing the error propagation, also called uncertainty, with regards to the determination of the complex stiffness modulus for asphalt mixes. Experimental data were taken as reference, considering two-point bending tests carried out upon French methodology standardized criteria. The Huet-Sayegh rheological viscoelastic model was parameterized by using the computerized tool Viscoanalyse Beta. The model developed was analytically deducted, based on partial derivatives, aiming at to discretize the influence of each variable from complex modulus rheological behavior, represented graphically by the Cole–Cole plan. Error levels of 2.5%, 5.0%, 7.5% and 10.0% were statistically simulated to verify the adherence between the experimental data and the rheological model mentioned. The results obtained indicate the efficiency of the methodology developed for establishing error propagation analysis between theoretical and experimental basis, reaching close adherence for maximum error levels of 10.0% and 5.0%, respectively.