Published: Jan 1994
| ||Format||Pages||Price|| |
|PDF (704K)||35||$25||  ADD TO CART|
|Complete Source PDF (8.0M)||35||$79||  ADD TO CART|
The backcalculation procedures are separated into five categories; (a) static linear elastic, (b) static nonlinear elastic, (c) dynamic linear using frequency domain fitting, (d) dynamic linear using time domain fitting and (e) dynamic nonlinear analysis. In this paper each category is described and case studies are presented comparing their results. Advanced techniques require more complete material characterization models. In the nonlinear elastic procedure a universal k1 - k5 model (an extension of the bulk modulus model) is used. In the dynamic analysis technique a generalized power law relationship is used for the asphaltic layer. Two case studies are presented. In the first the nonlinear elastic scheme was found to give excellent results at matching deflection bowls at four different load levels for each of two test sites analyzed. In these analyses only the k1, of the k1 - k5 model was backcalculated. It was found that the backcalculated k1 in all layers including the asphalt concrete are larger than those measured in the laboratory. In the second case study dynamic analysis techniques are applied to full wave shape data obtained from the SHRP data base. Both the frequency and time domain procedures are shown to yield reasonable results.
linear, nonlinear, dynamic backcalculation, elastic and viscoelastic material, flexible pavements
Associate Professor of Civil Engineering, TECHNION, Israel Institute of Technology, Haifa,