Design of laterally loaded monopile foundations in sand for offshore wind turbines

Date

2017-05

Authors

Dasgupta, Udit Shankar

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Globally renewable energy is gaining popularity with the largest scope in offshore wind energy. Although USA is one of the leaders in the onshore wind industry, it is yet to tap into the abundance of coastline it has available for this offshore wind development. High costs of installation have been a deterrent, with the foundation being a major part of that cost. Among the various options for offshore foundations, the most popular and at the forefront of feasibility in construction are the monopiles. The aim of this research is to see how well the p-y method specified in the American Petroleum Institute (API) design standard predicts the response of these laterally loaded piles in sands. This thesis also investigates possible approaches to modify the codes to better predict the response and improve the efficiency of the design of the monopiles. The p-y approach was developed for analyzing laterally loaded long slender piles which are widely used for offshore oil and gas platforms. These piles tend to bend and fail through formation of flexural plastic hinges. The large diameter relatively short monopiles for wind turbine applications are relatively stiffer and behave differently. They tend to rotate and/or translate instead of bending, evidenced by sizeable movements of the tip of the pile. Additionally, there is a serviceability criteria of the wind turbines to operate at low displacements i.e. low strain levels. This research focuses on the small strain response of the piles. The main behavior investigated with the long- slender pile load tests historically has been the response at large strain levels i.e. failure. The empirically calibrated API method may not be valid for such short-rigid piles and need to be verified before they can be used with confidence. A database of laterally loaded piles in sands is compiled and presented. Specifically the impact of the initial stiffness modulus (k) on the response was investigated using the finite difference code LPILE. The Embedded Length to Diameter Ratio (L/D) is an important factor of design that influences the response. Based on the few field tests chosen for analysis from the database, it is observed that the API code overestimates the initial stiffness of the pile for piles with lower L/D ratios for monotonic loading. For cyclic loading, the LPILE analysis does not lead to much change in initial stiffness from the monotonic simulation, but the actual measurements suggest an increase in stiffness. There is an added effect of base shear which might be an important force that is taken into consideration for short rigid piles, that the current code does not specify. Higher stresses locked into the soil due to pile driving might lead to higher base shear. Finally, since the problem is one of small strain, feasibility of future in-situ seismic testing is looked at by collection of sands from Mustang Island, TX, where the original field load tests were conducted that led to the specifications in the API code.

Description

LCSH Subject Headings

Citation