Abstract

     An increasing number of offshore wind farms are being constructed in seismic regions over liquefaction susceptible soils. This paper presents a methodology for the analysis and design of monopiles in seismically liquefiable soils by extending the established “10-step methodology” with an additional 7 steps. These additional steps include assimilation of seismic data, site response analysis, stability check of the structure (ULS check through the concept of load-utilization ratio), input motion selection, prediction of permanent tilt/rotation, and ground settlement post liquefaction. A flow chart, which shows the interdependence of the different disciplines, is presented and can be extended to routine design. This proposed method is validated using the observed performance of an offshore and nearshore turbine from the Kamisu wind farm during the 2011 Great East Japan earthquake. Predicted results based on the proposed methodology compare well with the field observation and demarcate the (i) good overall performance of the offshore turbines and (ii) limit state exceedance of the nearshore turbine. It is envisaged that the proposed method will be useful towards the design of monopiles-supported wind turbines in seismic areas.

Introduction

     Towards the end of 2020, the global wind energy generation capacity amounted to 733 GW, with new installations accounting for 111 GW, almost doubling that of 2019 [1]. Due to higher efficiencies and more stable wind conditions offshore than onshore sites, an increasing proportion of wind power is produced through large offshore wind farms. This demand for cost-efficient wind energy production has facilitated the industry to develop larger turbines with higher capacities. Fig. 1 presents the evolution of turbine capacity over the last 16 years, color-coded with rotor diameter. However, due to the high capital investments involved, profitability of these wind farms requires continuous operation and immediate functionality post-natural hazards. Therefore, the seismic resilience of OWTs is an important consideration in their engineering design and is, therefore, an area of active research. Readers are referred to recent publications [[2], [3], [4], [5], [6], [7], [8], [9]] on various aspects of seismic design.

Discussion and conclusions

     An existing framework for the foundation design of monopile-supported wind turbines was extended to include seismic load considerations in liquefiable and non-liquefiable soil. The framework was validated using a case study from Wind Power Kamisu, a near-shore farm that performed well during the 2011 Tohoku earthquake and subsequent tsunami.

The discussion and conclusions are presented succinctly below:

     ۱٫ A quantitative appraisal of appropriate damping, as recommended by Lombardi and Bhattacharya [58] and Adhikari and Bhattacharya [57], could vary the structural demands and therefore needs proper consideration. Back analysis of a case study from the Kamisu wind farm further indicated that structural demands could vary if suitable damping models are not selected.

      ۲٫ Definition of mechanism-specific p-y spring models for liquefiable and non-liquefiable soil is necessary as their selection could underpredict or overpredict structural response (deformation, tilt, RNA acceleration). As near-level ground conditions prevailed in this study, the authors employed hyperelastic (strain hardening) p-y curves for liquefiable soil to account for the increased resistance attributed to shear strain-induced soil dilation. During shear strain excursions, such dilative tendencies can transfer high acceleration pulses to the RNA and therefore need to be accounted for in the design stage.