PSI - Issue 75

Jeroen Van Wittenberghe et al. / Procedia Structural Integrity 75 (2025) 111–119 Jeroen VAN WITTENBERGHE and Vitor ADRIANO / Structural Integrity Procedia (2025)

115

5

position (main trolley at 9m) and put down in the North position (26m). This causes a different load history along the length of the girders as becomes clear while looking at the measured strain evolutions during the lifting events. When the load is being picked up, the strain values become negative since strain gauges SG1, 3 and 5 are located at the top plate of the main girder which goes into compression when loaded. For the first lift, where the trolley is kept more in the North, the most ‘Southern’ gauge SG1 is going into compression up to -122 µ ε . D uring the second lift, a value of -156 µ ε is measured by this gauge. Hence for a lifting event with the same recorded load amplitude, the load stress cycle is 28% different. For SG5 both lifts result in compressive strain peaks of about -210 µ ε . However, for the first lift two such local peaks are measured due to the back-and-forth movement of the trolley. Hence, for this location the rain flow counting will result in more than one stress cycle for this lifting event. For SG3 a similar strain amplitude is recorded for both lifts. This illustrates that even for relatively simple differences between lifting events, there can be a complex impact in terms of local stress amplitudes as well as local number of stress cycles. For correct calculation of the actual fatigue damage, a detailed load history should be used. The difference of 28% on stress amplitude as described above results in an over- or underestimation of the fatigue life by a factor of 2.1 for fatigue categories with a slope of m = 3.

Figure 5: Example of two lifting events as captured by the SHM system.

3.2. Fatigue Damage The fatigue damage calculations are carried out based on the signals measured by the SHM system. Hence this way a very accurate load history is used, taking into account the considerations that are described above. Since the SHM system on the Crane-AM has been implemented on a crane with a certain unknown operational history, measurements were taken for a period of approximately one year. This dataset has been used to determine an accurate usage profile for each weld of the crane. An example of the daily fatigue damage of one of the crane’s welds is shown in Figure 6. This figure illustrates the daily fatigue damage D calculated by the SHM system based on the measurements made during crane service. The average daily fatigue damage for this specific weld is about D = 5 ∙ 10 -5 . Extrapolated over a year D becomes 0.018 which means that with an unchanged usage pattern the fatigue lifetime of this weld is approximately 54 years. Similarly, the same evaluation is used to calculate the accumulated damage of all the 1298 welds of the crane for the past years when the SHM system has not been installed yet. The current operation of the crane is now continuously monitored and accumulated damage as well as remaining fatigue lifetime are updated automatically based on the actual usage of the crane.