Computational method for aerodynamic and aeromechanical analysis of offshore wind turbines

Abstract

Innovative wind power technologies have led to wind turbines with significantly longer and more flexible blade designs in order to meet the rise in green energy demands. A trend towards larger wind turbine sizes could potentially result in the blades experiencing aeroelastic instability problems. Furthermore, a typical wind farm is composed of multiple large-scale wind turbines, and therefore, the aerodynamics and aeroelasticity of a wind turbine can be influenced by various sources of flow unsteadiness generated by neighbouring wind turbines. The overall aim of this project is, therefore, to analyse the aerodynamics and aeroelasticity of wind turbines by taking various sources of flow unsteadiness into account using a high-fidelity computational method at an affordable computational cost. The computational resources and costs required for the aerodynamic and aeromechanical simulations of wind turbines using high-fidelity numerical methods are very high, which is the main challenge for the wind energy research community. Frequency domain methods, which are widely used in turbomachinery analysis but relatively new for wind turbines, are not only accurate but also computationally efficient for predictions of aerodynamic and aeroelasticity parameters. In this study, a nonlinear frequency domain solution method is proposed for the in-depth aerodynamic and aeromechanical analysis of wind turbines including multiple wind turbine models, taking numerous sources of flow unsteadiness into account. Various sources of flow unsteadiness, such as the harmonic inflow wakes, the oscillation of a blade structure, and the wake and turbulence from a neighbouring wind turbine are considered, and their effects on the aerodynamics and aeroelasticity of wind turbines are investigated. Different levels of modelling complexity are discussed in this thesis, including modelling and simulation of wind turbine blade aerofoils, rotor blades, complete wind turbine model including a tower, and multiple wind turbines in arrays. The frequency domain solution method makes it possible to model and simulate realistic flow conditions in consideration of the blade vibration as well as the effects of multiple wind turbines models without requiring significant computational resources. The present study reveals that the proposed nonlinear frequency domain solution method not only provides accurate predictions of aerodynamics and aeroelasticity of wind turbines but also reduces the computation time by one to two orders of magnitude compared to the conventional time domain methods.

Date of Award	11 Oct 2021
Original language	English
Awarding Institution	Northumbria University
Supervisor	Mohammad Rahmati (Supervisor) & Hamed Farokhi (Supervisor)