| RC bridge deck deterioration has been a problem in bridge engineering for a significant period of time. Until recently, more and more researchers began to recognize that truck traffic load is an important factor in deteriorating RC bridge decks. In order to study the stress distribution in RC bridge deck, a three-dimension strain transducer has been designed to measure the strain in three-dimension space, and field test is conducted on the RC deck of Harper Bridge embedded with transducers. A global finite element model with a submodel is used to simulate the deck system under truck wheel load. Three-dimension continuum solid element was selected to model the submodel in order to get more detailed strains in the deck thickness. The comparison with the field test shows that both of them are ready to be used in analyzing RC deck fatigue under truck wheel load. Based on continuum damage mechanics, micro mechanics, and fracture mechanical analysis, a nonlinear fatigue accumulation model is derived in the thesis. A fatigue damage threshold is selected to simplify the nonlinear damage accumulation so that a linear fatigue life prediction model, which considers the difference between field service performance and lab testing, can be achieve and will be easy to use by the engineer. On the other hand, a truck wheel weight histogram (WWH) generation model is developed to simulate real truck WWH based on FHWA VMT data. Wheel weight regression residual is used to reconstruct real truck WWH, which is modeled as skewed double exponential distribution derived by the author. After several important parameters for the fatigue model are investigated by means of field test, analytical and numerical analyses, two calibration procedures are developed for design and evaluation specifications for RC decks, which is based on a probabilistic concept to cover uncertainty involved in the fatigue problem of RC bridge decks. Finally, two sets of target beta and load safety factor are selected to achieve uniform failure probabilities for both design and evaluation. |
