Cyclic behaviour of laterally loaded (mono)piles in sand: With emphasis on pile driving effects

At the end of 2019, the European Union (EU) put forward the European Green Deal to facilitate the technological progress necessary to achieve CO2-neutrality by 2050. Such a monumental achievement would require massive investments in infrastructure for the harvesting, storage and the transnational transportation of green energy. To date, the more mature of the scalable (cf. to hydroelectric) green-energy resources is offshore wind, with joint academic and industry efforts allocated to reduce its capital expenditure. Approximately 13-37% of the required investment for offshore wind farms is currently expended on the design, manufacturing, and installation of the substructure. Further reduction in the cost of offshore wind can be achieved by addressing the main technical challenges associated with the predominant offshore wind foundation, i.e., the monopile. The main challenges typically relate to its lifetime operations, namely, (i) the identification of the wind turbine's fundamental frequencies, which are strongly dependent on the monopile-soil interaction, (ii) and the prediction of the lifetime foundation tilt, but also the current installation technology (impact driving); the current norm in the offshore industry. In particular, impact driving is associated with (i) long installation times, especially in the presence of competent soils, (ii) excessive use of construction material (steel) to avoid pile damage under many hammer blows, and (iii) costly underwater noise mitigation measures to reduce noise the levels of installation-borne noise emissions harmful to marine life.

In an attempt to accelerate the growth of offshore wind, the Netherlands, country of origin of this study, has supported several research initiatives to reduce the engineering and manufacturing costs for the prevalent offshore wind foundation in the country (the monopile). This study elaborates upon the experimental findings of two major research projects, namely the DISSTINCT (2014-2018) and the Gentle Driving of Piles (2018-2022) projects, each designed to address specific technical uncertainties associated with the foundation concept. The DISSTINCT project (launched in 2014) aimed to improve the understanding of the natural frequency of installed monopiles as well as the engineering procedures used in the identification thereof. By conducting experiments at full scale on a monopile installed in the IJsselmeer lake in the Netherlands, the experimental campaign produced invaluable data on the dynamic response of monopiles during small amplitude lateral vibrations. Later, the GDP project (launched in 2018) was designed to propose, engineer, and demonstrate a novel monopile installation procedure, foreseen to alleviate most of the aforementioned installation-related challenges; the Gentle Driving of Piles (GDP) method. Moreover, the project would provide answers to questions concerning the long-term response of (mono)piles in sandy soils, relative to the installation method. For these reasons, an extensive experimental campaign was conducted in the port of Rotterdam (Maasvlakte II), where a total of 9 piles were driven into the sandy Maasvlakte soil via different driving procedures, namely with the established impact hammering, the traditional axial vibro-driving, and the new GDP method. Subsequently, the cyclic lateral performance for four of these piles (which were heavily instrumented), was evaluated via an elaborate 82.000 load cycle (≈42 hours) loading programme of slow (0.1 Hz) high amplitude, and fast (0.1 - 4 Hz) low amplitude cyclic force applied to the (mono)piles' head.

This study elaborates and builds upon experimental findings from the above-mentioned test campaigns. These measurements were first carefully examined, and later interpreted using a variety of modelling tools (both 1D and 3D FE modelling) formulated and adapted to meet the particular geotechnical and loading challenges of the examined fieldwork. Enabled by the diversity of the field and numerical work performed, this study addresses a number of engineering challenges and knowledge gaps related to the design of monopiles, namely i) their post-installation resonance frequency, ii) the long-term response to environmental loading, and iii) the impact of the installation method on the long-term operations. In particular, 3D FE modelling was adopted to successfully simulate the dynamic response of the examined monopile in the DISSTINCT project. The modelling efforts enabled the interpretation of the field test measurements, and in turn, inspired confidence in the suitability of available simulation tools to identify the resonance frequencies of monopile foundations, and accurately calculate dynamic soil-monopile interactions. For the interpretation of the GDP field test data, 1D FE modelling was employed. In the field, the elaborate lateral loading programme returned a fairly complex cyclic pile response, with pronounced differences in the performance of piles installed by different installation methods. The particular geotechnical conditions at the GDP site, i.e., site inhomogeneity and the 4 m deep unsaturated topsoil, prevented the direct comparison of the installation methods. This was later achieved through the formulation of a cyclic soil reaction p-y model able to simulate soil ratcheting and gapping effects. The results provided rich insights into the impact of relevant installation effects on the cyclic pile response on many loading cycles and indicated that the GDP-installed piles performed excellent overall in lateral cyclic loading.