Yihan Xing

A 5MW direct-drive generator for floating spar-buoy wind turbine: Development and analysis of a fully coupled Mechanical model

This work forms the first of a two-part investigation aimed at identifying the challenges and opportunities of implementing a direct-drive generator for a spar-buoy type floating wind turbine. Preliminary specifications are presented for a fully coupled aero-hydro-servo-elastic model of a floating wind turbine with a 5 MW direct-drive generator. The drive-train model uses a low-speed, high-torque radial flux permanent magnet generator supported by two main-shaft bearings. The mechanical properties of the drive-train, including the main dimensions, mass of major nacelle equipment and details for the hub/nacelle assembly are presented. The rationale behind the adjustments to the tower and platform properties and the motivation to selection of best arrangement that is appropriate for supporting the developed system is explained. A discussion on the development of the variable speed-variable pitch control system suitable for the direct-drive system including modifications to avoid negative damping and blade-pitch instability are presented. Fully coupled simulations for the developed aero-hydro-servo elastic model were carried out in HAWC2 for the normal operating conditions of the wind turbine. The aerodynamic response of the model was verified and compared with that of a geared floating wind turbine system. Some initial results comparing the main shaft loads of the land-based and floating versions of the direct-drive wind turbine suggest satisfactory dynamic behaviour of the drive-train. The results prompt further research using a detailed drive-train model to verify the internal response, loading and durability of the components to assess their compatibility with a floating wind turbine system.

Gear Train Internal Dynamics in Large Offshore Wind Turbines

Today in the wind turbine global analysis codes such as Hawc2 [1] or FAST [2], the entire gear train is modelled by one degree of freedom constant stiffness torsional spring. This is because the focus in the global analysis programs lies mainly on the aerodynamic loads and the dynamic behaviour of structural members. For the small size gear trains, since the internal natural frequencies are expected in a frequency range above the overall wind turbine harmonics, this approach can be justified. However, as the industry trend is toward the larger drivetrains in offshore developments, the internal dynamic of gear trains are required to be modelled more accurately. Moreover, the development in generator technology with low, medium and high speed options has brought a variety of gear train design options with specific dynamic behaviour.In this paper the natural modes and internal dynamic excitations of high ratio wind turbine gear trains is investigated. Case study gear trains of 0.6, 2, 5 and 10 MW are modelled by pure torsional elements in a Multi Body Simulation (MBS) program; Simpack [3], where the natural modes are obtained and possible excitation are evaluated. The results show the resonance trend in various size wind turbine gear trains.Copyright

A 5 MW direct-drive generator for floating spar-buoy wind turbine: Drive-train dynamics

This article proceeds with investigations on a 5 MW direct-drive floating wind turbine system (FWTDD) that was developed in a previous study. A fully integrated land-based direct-drive wind turbine system (WTDD) was created using SIMPACK, a multi-body simulation tool, to model the necessary response variables. The comparison of blade pitch control action and torque behaviour with a similar land-based direct-drive model in HAWC2 (an aero-elastic simulation tool) confirmed that the dynamic feedback effects can be ignored. The main shaft displacements, air-gap eccentricity, forces due to unbalanced magnetic pull (UMP) and the main bearing loads were identified as the main response variables. The investigations then proceed with a two-step de-coupled approach for the detailed drive-train analysis in WTDD and FWTDD systems. The global motion responses and drive-train loads were extracted from HAWC2 and fed to stand-alone direct-drive generator models in SIMPACK. The main response variables of WTDD and FWTDD system were compared. The FWTDD drive-train was observed to endure additional excitations at wave and platform pitch frequencies, thereby increasing the axial components of loads and displacements. If secondary deflections are not considered, the FWTDD system did not result in any exceptional increases to eccentricity and UMP with the generator design tolerances being fairly preserved. The bearing loading behaviour was comparable between both the systems, with the exception of axial loads and tilting moments attributed to additional excitations in the FWTDD system.