PSI - Issue 64

Serdar Soyoz et al. / Procedia Structural Integrity 64 (2024) 484–491 Soyoz et al. / Structural Integrity Procedia 00 (2024) 000 – 000

485

2

Recent studies on vibration-based monitoring of wind turbine towers are summarized in the following lines. Hansen et al. (2006), estimated modal values of a 2.75 MW wind turbine under operational conditions. Similarly, a long-term monitoring of an offshore 5 MW wind turbine to obtain modal parameters of the support structure under parked conditions was presented by Devriendt et al. (2014). In a different study, in addition to monitoring campaign of a 2 MW wind was explained and acceleration to SCADA data were correlated to the non-operating and operating conditions by Oliveira et al. (2016). Continuous monitoring of a 2 MW wind turbine under operational conditions was also presented by Oliveira et al. (2018a). The variability of the dynamic properties of the turbine was shown and modal damping values were obtained using identified shut-down events. Effects of operational and environmental conditions as well as resonance phenomenon were investigated in studies by Hu et al. (2015). Oliveira et al. (2018b) developed a continuous monitoring system to determine the modal properties of a 2 MW wind turbine and to detect changes in these properties after the reduction of the effects of environmental and operational conditions on the modal properties through regression models. Also, three damage scenarios were created based on offshore and onshore foundation and blade damage, and it was shown that the system was able to detect damage. Seismic analyses of wind turbines have also been reported in the literature. A review paper by Katsanos et al. (2016) pointed out that unlike the popular opinion of wind turbines being slender and, therefore, "earthquake repellent", in many cases the earthquake load is driving the design in seismic zones. Some severe damage types were found out to be overturning moments, high steel stresses that can exceed yield strength, soil compliance and resonance risk. Diaz and Suarez (2014) performed seismic analyses of a wind turbine considering operational conditions. Critical stresses occurred on the tower when the extreme wind was combined with chosen earthquake motions. In a study by Patil et al. (2016), seismic fragility analysis of a 1.65 MW wind turbine with 80 m hub height was presented. Analysis of a 450 kW wind turbine tower under gravitational static, seismic and aerodynamic loading was reported by Bazeos et al. (2002). Buckling was considered as the limit state and placing soil springs under the structure and their effect on overall analysis was discussed. A methodology for seismic risk assessment of steel tubular tower wind turbines is discussed by Nuta et al. (2011). Incremental dynamic analysis of a 1.65 MW wind turbine was performed. Fragility curves were developed by defining the damage limit states as residual tilt, first buckling and the first yield of the tower. Mo et al. (2017) assessed the vulnerability of an off-shore 5 MW wind turbine with a monopile foundation using fragility curves. In this study, long-term vibration-based monitoring of a 900-kW onshore wind turbine in Istanbul, Turkey was presented and the effects of environmental and operational conditions on the modal parameters were discussed. Then, the FEM of the structure was updated based on identified modal parameters. After all, local buckling of the tower was considered as damage state, and the seismic fragility curves considering different levels of wind speed were developed. Therefore, this study presents an overall framework for seismic fragility analysis of wind turbines with realistic and calibrated finite element models. 2. Modal identification 2.1. Wind turbine and sensor layout A 900 kW onshore wind turbine (Fig. 1) at Saritepe Campus of Bogazici University, Kilyos, Istanbul has been monitored since 2016; in this paper results between April 2018 to April 2019 are presented. Technical specifications of the wind turbine are given in Table 1.