Modelling and control for structural load mitigation of wind turbines
Girsang, Irving Paul
Date of Issue2016-05-10
School of Mechanical and Aerospace Engineering
The capacity of installed wind turbines have been steadily growing. In 2015 alone, 63 GW has been installed worldwide, which is 21% more capacity than what was installed in the previous year. This has been possible by larger and taller wind turbines that allow for more energy capture. To keep up with the trend of increasing turbine sizes and hence structural flexibility, it is important to ensure mitigation of its structural loads so that the cost of wind energy can be kept low by lessening the maintenance requirements and improving the overall turbine reliability. Compared with upgrading the mechanical system to preserve the components’ lifetime, advanced controllers have been identified as more attractive and cheaper cost reducing strategies. Implementing sophisticated control systems can assure safe and optimal operation in terms of load mitigation and power enhancement. In this thesis, several novel controller designs have been developed to mitigate fatigue loads on wind turbines’ blades and drivetrain. The controllers are designed to attenuate loads in ways that have not been paid much attention before and verified to yield superior load attenuation as compared with the ones have been achieved so far. A novel individual pitch controller (IPC) has been designed based on the new knowledge of mitigated blade loads at a yaw misaligned condition. Compared with the industrial standard pitch controller, the proposed controller is shown to contribute at least a 31.54% reduction in the blade out-of-plane fatigue load at various turbulent wind conditions. A new integrated wind turbine model that couples high-fidelity aerodynamic, structural, drivetrain and electrical models is presented. This new integration allows for consideration of the grid conditions as well as assessing the mitigation responses, in term of wind turbine loads. It can save the design costs by allowing dynamic interactions (including the controller) to be taken into account prior to physical assembly. A new controller has been designed and verified based on this integrated model to avoids drivetrain resonance through addition of virtual inertia.
DRNTU::Engineering::Mechanical engineering::Alternative, renewable energy sources