With the development of offshore wind power, the reliability analysis of offshore wind turbines is increasingly significant due to the system complexity and negative impacts in harsh operating conditions. In this study, the Fault Tree Analysis method is adopted for both qualitative and quantitative evaluation of semi-submersible floating offshore wind turbine failure characteristics. The floating offshore wind turbine is divided into several assemblies, including support structures, pitch and hydraulic system, gearbox, generator and the other systems. Failure rates of relevant offshore structures are collected from previous studies, reports and reliability databases. On this basis, the quantitative assessment of Minimum Cut Sets and Importance Measures are achieved. The calculated results are generally in conformity with statistical data, indicating that most of the failures are caused by several basic factors. Marine conditions, especially the salt-spray and high wind speed, show the most significant impact on floating offshore wind turbine performance.

INTRODUCTION

The economic efficiency of fixed offshore wind turbines decreases with the increase of water depth. The wind turbines, installed on the floating structures, offer a feasible solution to deal with this problem (Global Wind Energy Council, 2014). The advantages of floating offshore wind turbines (FOWTs) in comparison to fixed turbines can be listed as follows:

 More flexible construction and installation procedures

 More insensitive to water depth

 Higher wind speed

 Less noise pollution

 Lower demolition cost

The major constraints for offshore wind development is the costs. FOWTs are installed far from shore, leading to higher installation and construction expenses because of the complex marine conditions (Castro Santos et al., 2016). Besides, the difficulty of maintenance procedure calls for vast expenditure (Santos et al., 2015a). One way to provide effective maintenance is through reliability analysis, by predicting the weak points in the system at the design stage (Santos et al., 2016). Blanco (2009) showed that the Operation and Maintenance (O&M) costs can be 20%-30% of the total Level Cost of Electricity (LCOE) over the project’s lifetime. Early detection of incipient faults prevents major component failures and allows for the implementation of predictive repair strategies (Yang et al, 2012). Therefore, appropriate actions can be planned in time to prevent major failures which would result in significant O&M costs and downtimes.