Use of monitoring extreme data for the performance prediction of structures: General approach

Engineering structures are subjected to time-dependent loading and strength degradation processes. The main purpose of both designer and owner is to keep these processes under control. Several numerical approaches based on mechanical, physical, chemical or combined models have been recently proposed to describe time-dependent processes of engineering structures. Most of them require considerations of both aleatory and epistemic uncertainties. The inclusion of such uncertainties demands intensive studies in space and time of engineering structures under environmental and mechanical stressors. Existing mechanical models for structural performance assessment can be validated by using structural health monitoring. The use of monitored extreme data allows (a) the reduction of uncertainties associated with numerical models, and (b) the validation and updating of existing prediction models and, sometimes, the creation of novel models. This paper presents a general approach for the development of performance functions based on monitored extreme data and the estimation of possible monitoring interruption periods. An existing bridge in Wisconsin is used as an example for the application of the proposed approach.