دانلود رایگان مقاله انگلیسی پاسخ دینامیکی تک ستون در شن و ماسه با استفاده از مدلسازی سانتریفیوژ - الزویر 2018

ABSTRACT

Monopiles are one of the most commonly used offshore foundation for wind turbines. Their static capacity, p-y curve and cyclic loading behaviour have been studied using 1 g tests and centrifuge tests, but there is little experimental data regarding their natural frequency, especially using centrifuge testing. The design of offshore wind turbine foundations is largely governed by natural frequency as resonance due to cyclic loading can cause damage and even failure. Understanding the dynamic response of the monopile under free vibration is thus critical to design. This paper presents the results of novel monopile (large diameter) and single pile (small diameter) tests in a centrifuge to for the first time directly determine the natural frequency (fn) of the pile-soil system. An experimental methodology was used to define the natural frequency via measured acceleration and force time histories and their fast Fourier transforms (FFT) under a force applied at a controlled frequency. The effects of pile diameter, embedded length, free length of the tower and soil density on fn were investigated in the centrifuge tests. The same models used in the centrifuge test at 50 g were also tested at 1 g in order to assess the relevance of earlier 1 g investigations into system behaviour. The measured natural frequency of wind turbine monopiles in centrifuge models during harmonic loading from a piezo-actuator, confirmed that soil structure interaction at an appropriate stress level must be taken into account to obtain the correct natural frequency. The experimental data was compared to theoretical solutions, giving important insights into the behaviour of these systems.

Conclusion

This paper presented a methodology to measure the natural frequency of monopiles in centrifuge tests for loose and dense sand in wellcontrolled conditions and compared the results with theoretical solutions. The experimental results show that the fixity provided by the pile has a large effect on the natural frequency of the system, although for many geometries tested here the length of the piles was sufficient that increasing pile length had only a modest effect on the natural frequency. The frequencies observed were, however, lower than those of the fixed base structure, a point of fixity existing beneath the soil surface leading to an increased cantilever length and thus a reduced natural frequency. The ratio of free length and embedded depth (LT/LP) was also investigated. Only two different free lengths were used, and it is clear that the natural frequency changes inversely to the free length. For the same free length (LP =10 m), the natural frequency and fn/fn-str increase when the embedded depth increases (or LT/LP decreases). Although the relative densities of the two sand samples are very different, the data show only small differences in the natural frequency obtained, possibly due to the conflicting effects of increased stiffness but also increased added mass from the participating soil. Small-scale tests under 1 g conditions were also performed using the same models, sand conditions, and other variables. When comparing the response, the low stress level of the 1 g tests reduced the soil stiffness while the pile stiffness was maintained. The experimentally observed dynamic response illustrates this difference with the fn measured in the centrifuge scale model being higher than that 1 g, the variation not being a constant or following a well-established correlation.