Prediction of base and subbase resilient modulus (Mr) using regression methodology

作     者：Yilmaz, Altan 

作者机构：Mehmet Akif Ersoy Univ Fac Engn & Architecture Dept Civil Engn TR-15030 Burdur Turkey 

出 版 物：《JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY》 (J. Fac. Eng. Archit. Gazi Uni.)

年 卷 期：2020年第35卷第1期

页      面：507-517页

核心收录：

主　　题：Granular materials highway pavements resilient modulus linear multiple regression 

摘      要：Resilient modulus is an important design parameter for highway pavement structures because it represents the structural strength of pavement layers. The resilient modulus depends on the factors such as applied stress, loading time, water content, dry density and gradation. This paper demonstrates the applicability of regression methodology for estimating the (M-r) Resilient modulus of pavement layers using the results of dynamic triaxial tests. Aggregate samples were collected from several different quarries of central Anatolia. All of the aggregate sources were igneous rock (basalt, and trachy-basalt). Initial work content of triaxial tests to obtain the resilient modulus of cylindrical aggregate samples. Than new mathematical model proposed by using the results of dynamic triaxial tests. In the regression model;aggregate physical properties, aggregate mixture properties and loading factors (a total of 8 variables) which are used as input parameters and the resilient modulus of the aggregate mixture obtained as output. In order to compare the effectiveness of the new method, coefficients for the Uzan constitutive model were also determined for laboratory testing and were compared with the approach described in this paper. Performance parameters of R-2 :0.98 and Standard error: 10.51 was obtained from model prediction. These results are quite sufficient, and the regression model assumed the resilient response to be like a function by using the stated material parameters. So, this approach makes it possible to estimate the resilient modulus of the different aggregates samples in real-time.