This thesis leads to a framework for micrositing, the process through which the specific location for mounting micro wind turbines in urban environments is determined. It can be used as a guidance on how to model an area of interest, find the optimum location for micro wind turbines installation and calculate the annual energy production, commenting on the accuracy that can be expected from the results. Essentially, it is composed of three parts, each one deals with different set of tasks associated with model development and simulation. The first part investigates the computational practices to the fields of turbulence in urban environments implemented in the open-source CFD library OpenFOAM. It examines the performance of a turbulence model, known as DES, which has not been previously used for external flows in complex urban environments and concludes that this approach offers improved robustness and accuracy over a range of wind conditions. It offers improved prediction of flows in wake regions compared to RANS methods and is less computationally demanding than full LES approaches. The validity of DES implementation is tested using data sets derived from both wind tunnel experiments and field measurements. In the second part, a procedure is developed to identify the optimum location for mounting wind turbines, based on the spatial variations in mean annual wind speed and the corresponding annual energy production (AEP). The procedure utilizes one year of measured wind data for one site to extrapolate (using the `Wind Atlas Methodology') the annual wind speed at the site of interest. Then combining the climate data with the CFD results and the power characteristics of the micro wind turbines, it estimates the mean wind speed and the annual energy yield. Essentially, this methodology leads to the formation of three dimensional fields of the average annual wind speed and the AEP (3d wind maps), which will enable identification of the effects of the complex urban topography on the wind flow, and the potential locations for micro wind turbines installation. The third part examines the accuracy that can be expected from the annual energy production estimation techniques and provides guidelines on the calculations. In particular, it investigates the validity of the standard power curves for the site-specific air density and evaluates their effect on the annual energy production estimations. Differences of the order of 10-3 between the default and the site specific mean air density (ρ), do not change substantially the energy production. However, for higher discrepancies of the order of 10-2 the power output can differ more than 10%. Turbulence affects the wind energy in two ways: through power performance impacts and through effects on turbine loads and fatigue. In the operational range of each turbine, TI increases the output at low wind speeds, while in the transition region to rated power it decreases the power output. In the context of this study, the DES approach was implemented to examine the flow at the De Montfort university campus in Leicester. The 3d wind maps for the mean wind speed and the annual energy production were developed and the optimum locations for micro wind turbines installation were identified. Although the rooftops of the higher buildings have mostly the potential for wind energy applications, the effect of the urban topography on the wind potential is not always apparent. Lower building can occasionally have higher potential for micro wind turbines installation than taller and roofs of the same height and close each other may differ substantially in their predicted energy output. Using the field measurements by two 3d ultrasonic anemometers placed in the campus, the site specific air density and turbulence intensity were considered to correct the energy yield estimations and evaluate their effect on the results.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:716181 |
| Date | January 2017 |
| Creators | Dadioti, Rallou |
| Publisher | De Montfort University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/2086/14219 |
Page generated in 0.0026 seconds