Degree Type

Dissertation

Date of Award

2015

Degree Name

Doctor of Philosophy

Department

Civil, Construction, and Environmental Engineering

First Advisor

Sri Sritharan

Second Advisor

Jon M. Rouse

Abstract

In a prestressed concrete bridge, quantitates characterizing structural behavior such as deformations and stresses are prone to change during and after the construction due to thermal effects, varying time-dependent material properties and variations in loading and/or support locations. In order to ensure the satisfactory performance of bridges as a function of time, prestressed bridges should be designed to fulfill short-term and long-term functional requirements such as those defined by serviceability criteria, strength requirements, and durability conditions. Depending on the method of prestressing, the long-term response of the bridge will vary and the design of certain structural components may be governed during the construction. To obtain a more realistic assessment of the functional requirements throughout bridge’s service life, this study was systematically undertaken to improve the prediction of the strain and stress build-up during and after construction in both pretensioned and posttensioned concrete bridges.

With respect to the use of pretensioning as a means of prestressing, this study investigates the long-term camber of precast pretensioned concrete beams (PPCBs). Construction schedule delays and additional costs are common problems when the actual cambers of PPCBs are different from those expected during bridge design. To reduce the discrepancy between the predicted and actual camber, a systematic study was undertaken to identify the key parameters affecting camber, needed improvements to construction practices, and potential refinements to the predictive analytical models. Using the finite-element analysis (FEA), the long-term camber was predicted with a mean error of and 24.1±29.5% and 8.6%±14.5% for the small- and large-camber PPCBs, respectively, when the thermal effects were ignored. By incorporating a linear temperature gradient with a mean temperature difference of 15°F in the FEA, the corresponding errors were reduced to -14.7±22.5% and -1.2±10.7% for the small- and large-camber PPCBs, respectively. In consideration of the design practice, suitable long-term camber multipliers, which account for the support locations and the thermal effects were proposed.

While using a linear temperature gradient satisfactorily modeled the thermal effects on camber, a more detailed investigation was carried out to quantify thermal effects with due consideration to the weather conditions and meteorological seasons. Thermal effects were quantified and combined with the effects due to dead load and prestress for PPCBs to improve the accuracy of camber and corresponding stresses. The Monte Carlo simulation was adopted to probabilistically model the thermal deflections and stresses to account for the characteristic variability of the temperature gradients. To utilize the outcomes of this study in design practice, suitable thermal multipliers are proposed which effectively reduced the mean absolute error between the measured and predicted camber from 13.3±10.0% (i.e., when ignoring the thermal effects) to 5.0±4.6%.

For posttensioned concrete bridges, this study investigates the time-dependent effects on cast-in-place posttensioned concrete box-girder bridges (CIP/PCBB). It was found that the displacement-induced column forces in CIP/PCBB caused by the time-dependent shortening of the superstructure are not systematically addressed in the current design methods. Due to unrealistic estimate for shortening strain rate of the superstructure and neglect of the beneficial effects of concrete relaxation in the current design methods, the displacement-induced forces are overestimated. When these forces are combined with the effects of live loads and seismic loads, the end result is inefficient designs of columns and foundation, and increased costs.

Given consideration to the shortcomings of the current design guidelines, a systematic investigation was undertaken to more accurately determine the displacement-induced forces. In the first part of this investigation, a combination of an experimental study and the FEA was used to characterize the concrete relaxation and subsequently demonstrate its beneficial effects on displacement-induced forces in the columns of a prototype CIP/PCBB. Three different specimens were used to characterize the relaxation of the normal strength concrete over short durations (i.e., less than five days) after the concrete had fully matured (i.e., after 28 days). Then, using the FEA, the time-dependent deformations and stresses of the demonstrative CIP/PCBB were evaluated from the time of construction to completion of the CIP/PCBB with due consideration to the relaxation. Both the experimental study and the FEA verified the beneficial effects of concrete relaxation in significantly reducing the displacement-induced forces. In the second part of the investigation, the effects of creep and shrinkage on eight CIP/PCBBs of various lengths and configurations, were examined using the FEA. The beneficial effects of concrete relaxation were integrated in the FEA. For the eight CIP/PCBBs, the shortening strain rate of the superstructure together with the variation of lateral column displacement and the corresponding base shear force with time were calculated and then compared to the corresponding values estimated by the current design practice. It is shown that the current design practice underestimated the design strain rates by a mean value of -77.2% for eight PCBBs, underestimated the design column top lateral displacements by a mean value of -67% for 37 columns, and overestimated the design base shear forces by a mean value of 20% for 37 columns compared to the corresponding results from the FEA. Based on the findings of the FEA, modifications to the current design guidelines are proposed to more accurately determine the displacement-induced column forces.

Copyright Owner

Ebadollah Honarvar Gheitanbaf

Language

en

File Format

application/pdf

File Size

262 pages

Share

COinS