Show simple item record

dc.contributor.authorAtkinson, Judith Ann
dc.date.accessioned2018-06-13T21:00:08Z
dc.date.available2018-06-13T21:00:08Z
dc.date.issued2002
dc.identifier.urihttp://hdl.handle.net/11122/8598
dc.descriptionDissertation (Ph.D.) University of Alaska Fairbanks, 2002
dc.description.abstractThe two main objectives of this research were first to investigate existing backcalculation techniques and the theories behind them for pluripotence (mathematical uniqueness of solution). Second to develop a backcalculation technique that addresses the pluripotence issue and attempts to overcome it. The first objective was accomplished by demonstrating that most existing backcalculation programs do have pluripotence issues of some sort. Each program was analyzed and an attempt was made to determine how pluripotence is evaded. To accomplish the second objective the computer program KISS was developed based on Composite Plate theory. The program was verified by analyzing the deflection equations used for Composite Plate theory and comparing them to some popular deflection equations used in backcalculation today. The Composite Plate theory equations were also validated using finite element analysis. Two types of sensitivity analysis were conducted. First the Composite Plate theory deflection equation was analyzed for sensitivity to various parameters. It was found that the deflections are most sensitive to layer thickness values and least sensitive to Poisson's ratio values. Secondly, the KISS program was analyzed for sensitivity. Plate size was investigated and it was determined that the plate size has a significant effect on backcalculation results. Variations in applied stress, seed values for D? and k and AC assumed strength (strong, average or weak) were also included in the sensitivity analysis. The KISS program was then compared with other popular backcalculation programs using data from the LTPP database. Three types of comparisons were made. First a Program-to-Program comparison was made in which the KISS program was compared to each of the other programs on a one-to-one basis. Secondly a Case-by-Case comparison was made in which the results from all programs were compared for each pavement system from the LTPP database. Thirdly, theoretical deflections from three different sources (Composite Plate theory, Elastic Layer theory, and finite element method) were generated for eight different theoretical pavement systems. In this method, the original moduli are known and were used as a basis for comparison.
dc.subjectCivil engineering
dc.titleThe Pluripotence Of Backcalculated Moduli
dc.typeDissertation
dc.type.degreephd
dc.contributor.chairRaad, Lutfi
refterms.dateFOA2020-03-05T15:51:41Z


Files in this item

Thumbnail
Name:
Atkinson_J_2002.pdf
Size:
6.489Mb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record