PSI - Issue 64

Julian Unglaub et al. / Procedia Structural Integrity 64 (2024) 918–924 Julian Unglaub / Structural Integrity Procedia 00 (2019) 000–000

919

2

1. Introduction The service life of steel bridges can be considerably reduced due to changes in traffic volume, execution deficiencies or insufficient maintenance. Especially in bridge construction, many existing bridges need to be retrofitted as a result. Maintenance and repair operations, usually in the form of welding work, are required in order to maintain the structure or until a new structure can be built as replacement. To carry out the welding work, bridges must be closed to vehicles in accordance with the current state of the art. This is to avoid the relative seam flank movement of the crack during welding. In addition, closure of bridges leads to detours or mobility restrictions in general and, in turn, to an increased load on the bridges on the detour routes. It is therefore desirable to repair the cracks under service conditions. Previous investigations have already shown that weld joints can be produced under service conditions. Fracture mechanics analyses have not revealed any differences in the behavior of such repaired components, Schiebel (1998). The investigations showed that hot cracks occur during welding as soon as the moving seam flank exceeds a critical value, Wichers (2006). Hot cracks form as solidification cracks in the center of the weld. They are caused by shrinkage of the molten metal. Tensile stresses arise during the solidification interval between the liquid and solid phases. In addition to the welding process parameters and the alloy composition, the clamping conditions are decisive in causing hot cracks, Cross and Böllinghaus (2005). Previous investigations have been limited to single and double-pass welds. In bridge construction, multi-layer welds are common in fatigue-relevant areas. In addition, the assessment and reproduction of realistic load conditions has not been the focus of research. As a result, welding work under service conditions has so far only been carried out in individual cases and at considerable expense. This is mainly due to insufficient knowledge about the achievable weld quality and fatigue strength of such welds. Accordingly, operators are reluctant to commission the method, although the following studies show that welding in service is technically possible subject to certain restrictions: compliance with limit values for gap opening amplitude and gap opening frequency Begemann et al. (2024). Bridges require regular inspection and maintenance during their service life, which is costly and time-consuming. In particular, the right time to use a retrofitting method is crucial. If an action is implemented too early, the infrastructure will not be used efficiently. If actions are carried out too late, unnecessary additional damage can occur, including total failure of the structure. Determining the optimal point in time is the field of predictive maintenance. Digital twins (DT) represent a potential solution for supporting the operation and maintenance process of existing bridges. DTs incorporate a geometric-semantic model of the bridge in question, offering a comprehensive and detailed representation of the bridge's physical and structural characteristics. The application of DTs in operation and maintenance is categorized by Jiang et al. (2021): monitoring, analysis and action. While defect detection and asset monitoring are sub-categories of monitoring, the subcategory retrofitting and demolition is part of the category of action. Jiang et al. (2021) reviews several case studies on retrofitting, but they are manly related to buildings environment. They can only be limited applied to the retrofitting of infrastructures due asset monitoring or defect detection nature. Although there is a need for concepts for integrating condition-based monitoring into DT for predictive maintenance, implementation in practical applications for retrofitting has not yet been demonstrated fully successful. 2. Monitoring of crack movement The effect of cracks on the load-bearing capacity of bridges depends on the location of the damaged detail in the structure. In order to assess the risk potential of a crack, Sedlacek et al. (2011) introduced four damage categories. While cracks between the deck plate and the cross girder (category 1) belong to the less severe damage, cracks on the main girder (category 4) are considered to be significantly more dangerous. From a measurement point of view, however, cracks close to the load application are much more challenging to measure, as high measurement frequencies of up to 200 Hz are required. These high frequencies result from the motion of the individual axes of passing traffic. A monitoring system inspired by fracture mechanics testing was used to determine the crack openings present at the connection of a cross member to the cover plate, see Fig. 1. Two clip-on extensometers (MTS, model 632-02F-20), which were attached to the cover plate and web on both sides using magnets with two adapters each. The magnetic