Using heterogeneous measurement data to update the corrosion degree in bridge girders

Existing concrete bridges can show signs of deterioration, e.g. due to corrosion, in which case the question arises whether these bridges still have an acceptable safety level. To obtain additional information about the state of the structure, different types of information can be gathered. However, the measurements performed are often not used directly to update the estimate of the corrosion degree of the bridge and the resulting information from different measurement types is not combined. In this contribution, it is investigated how parameters of the service life models for concrete girders subjected to chloride-induced corrosion can be updated based on heterogeneous measurement data, with the focus on modal measurements and strain measurements. The posterior distributions of these parameters are generated based on Bayesian updating. These updated distributions can be used to get a more accurate estimate of the real deterioration state of the bridge. Based on this, the failure probability of the bridge and its remaining service life can be estimated more accurately.