Reliability model for lifetime multi-objective optimization of a structural health monitoring system embedded in a deteriorating reinforced concrete bridge deck

The objective of the research conducted for this thesis is to improve the accuracy of lifetime reliability estimates calculated using a predictive model for a RC bridge deck subject to one-way flexure. Model accuracy can be improved by incorporating spatially and temporally variable corrosion data collected by an optimized, embedded SHM corrosion sensor network. Multi-objective optimization techniques are applied to find the best combinations of sensor cost and spacing in the design of a SHM corrosion rate sensor network. The optimal combinations of sensor cost and spacing yield the best tradeoff between total system cost and performance. This optimization yields corrosion rate sensor data for multiple critical sections throughout a RC bridge deck over time based on empirical spatial and temporal relationships. Applying this data with Monte Carlo simulation, an existing spatially invariant computational reliability model is improved. The improved reliability model incorporates several sub-models to determine the changes in load effects on and resistance of a RC bridge deck slab over time, as well as spatial correlation of corrosion and a system approach to account for spatial variability. The improved reliability model provides a better estimate of the service life of a RC bridge deck. An improved reliability model incorporating SHM sensor data will allow infrastructure managers to make more informed decisions on when and how to maintain and repair a RC bridge deck in order to maximize its service life at minimal cost. (Abstract shortened by UMI.)...