- Title
- Aeroelastic measurements, simulations, and fatigue predictions for small wind turbines operating in highly turbulent flow
- Creator
- Evans, Samuel
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2017
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Small wind turbines fulfil a unique niche within the energy market, and are frequently utilised for decentralised generation or in domestic ‘off-grid’ applications. Small wind turbines are defined by IEC 61400.2, as having a rotor swept area less than 200 m2, which corresponds to a power output typically less than 50 kW. Wind turbines of this class often rely on a tail fin for yaw control, and operate at much higher rotor speeds than large turbines to ensure optimal aerodynamic efficiency. Additionally, they are subject to low Reynolds number aerofoil operation, and often use control systems with minimal inputs to regulate generator output. As a consequence, they are not simply ‘scaled-down’ large wind turbines. Due to commercial interests, significant research and development effort has gone into large wind turbine technology, with little flowing into small turbine technology. This work has focused on developing a detailed aeroelastic model of a small, 5 kW horizontal-axis wind turbine. Experimental validation has been undertaken using operational measurements, revealing a good level of accuracy. Site wind measurements have highlighted differences in open-terrain wind turbulence models specified in the IEC standard to those measured at built environment sites. It has been shown that increasing turbulence levels have a detrimental effect on structural loadings and fatigue life, whereby fatigue loads in the standard are highly conservative and not indicative of operational loading. Results and conclusions will facilitate the development of lightweight, cost-effective, and structurally sound blades for small wind turbines. Wind conditions, turbine performance, and structural loading at turbulent built environment locations are not well understood. IEC 61400.2 has proposed a simplified load model (SLM) for predicting load magnitudes, at the ‘cost’ of marked design conservatism. Few studies have undertaken detailed experimental measurements to record operational loading for predicting component fatigue life. To date, no studies have quantitatively compared fatigue life predicted by the SLM to aeroelastic simulations and measured operational loads. Work in this thesis focuses on bridging this knowledge gap by assessing the SLM with detailed aeroelastic simulations and experimental measurements of a 5 kW horizontal-axis Aerogenesis wind turbine installed at the University of Newcastle, Australia. A secondary aim is to measure and assess wind conditions and turbulence levels of built environment wind regimes. To measure the turbine performance, a simple, cost-effective data acquisition system was developed utilising Arduino-based microcontrollers. Turbine generator power, rotor speed, blade response, and tower accelerations were recorded using this system across a wide range of measured wind conditions. A detailed analysis of the site wind resource revealed that for all wind speeds, the ten-minute turbulence level was higher than that specified by the IEC standard normal turbulence model. The IEC turbulence model is therefore under-conservative for this complex terrain site, and may under-predict fluctuations in wind velocity, power generation, and structural loading. An aeroelastic model was developed within FAST (Fatigue, Aerodynamics, Structures, and Turbulence) software, and included all necessary structural, inertial, aerodynamic, and controller parameters. Very good agreement was found between experimental measurements and predicted azimuthal blade structural response and aerodynamic loading at design conditions, where the blade root flapwise moment was predicted to within 8%, and corresponding structural deflections within 10%. The predicted dynamic response of the blades and tower was reasonably well captured in the low-frequency range, with a reduction in accuracy for the higher frequency response. This may be due to FAST’s limited ability to capture both higher-order aerodynamic and structural behaviour. When comparing the simulated and measured control system, the simulated controller tended to act more aggressively during highly unsteady conditions. To date this is the most comprehensive aeroelastic model of a turbine of this class reported in the literature. Using both measured wind from the Newcastle site, and simulated wind as per the Kaimal model detailed in the IEC standard, FAST simulations were undertaken to assess the impact of turbulence on: turbine electrical power output, tip speed ratio, rotor thrust and torque, blade loads, drive shaft loads, and tower base bending moment. An increase in turbulence levels generally resulted in an increase in mean performance and loading. Across the turbulence range investigated, the maximum rotor thrust load increased from 846 N to 2,041 N, the maximum flapwise blade loading increased from 629 Nm to 1,103 Nm, and tower base load was found to increase from 17 kNm to 41 kNm. This is a significant increase in structural loading and is not predicted by the IEC wind model for design conditions. Damage equivalent loads (DEL’s) at the blade root were produced from the SLM and measured operational data at 603 Nm and 70 Nm respectively, representing a significant over-prediction. Current blade fatigue life was estimated via the Palmgren-Miner linear damage model, with a life of 0.09 years calculated when using the SLM. A fatigue life of 9.18 years was calculated for the measured fatigue spectra, and is equal to 102 times that predicted by the SLM. The constant blade fatigue damage amplitude and cycle ratio predicted by the standard was well in excess of measurements and simulations, indicating that the SLM is over-conservative and not representative of the variable physical damage spectra that is experienced by the blade particularly during unsteady operation.
- Subject
- small wind turbines; aeroelastic modelling; FAST; IEC 61400.2
- Identifier
- http://hdl.handle.net/1959.13/1349817
- Identifier
- uon:30448
- Rights
- Copyright 2017 Samuel Evans
- Language
- eng
- Full Text
- Hits: 2575
- Visitors: 3626
- Downloads: 1436
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details Download | ATTACHMENT01 | Thesis | 28 MB | Adobe Acrobat PDF | View Details Download | ||
View Details Download | ATTACHMENT02 | Abstract | 190 KB | Adobe Acrobat PDF | View Details Download | ||
View Details Download | ATTACHMENT03 | Data file | 25 MB | Zip Compressed File | View Details Download |