By Ben Simpson and Chris Humphris, Fluent Europe Ltd., UK
Countries such as the UK are aiming to increase generation from renewables to combat climate change. The resulting rise in generating costs makes the need for efficient energy production clear.
The UK government’s recent Energy White Paper set out ambitious targets to reduce greenhouse gas emissions by 60 per cent of 1990 levels by 2050. Taking advantage of Britain’s windy and wave-swept shores, a big investment in wind and tidal power was therefore proposed to increase renewables’ share of total national energy provision from 2.6 per cent in 2001 to 10 per cent by 2010.
If these targets are to be met, efficient energy production will be important. Computational Fluid Dynamics (CFD) is one way in which efficiency can be maximized, helping to ensure that the renewable energy bandwagon doesn’t fall flat.
Figure 1. Velocity contours behind one turbine show the wake effect on a second, smaller turbine
The Association for the Conservation of Energy (ACE) recently criticized the omittance of any targets on energy efficiency in the Energy White Paper. While the true value of targets in attaining results can be debated, it is certainly true that energy efficiency is going to play an important part in realizing the aim of significantly developing renewable energy, as a major contributor to UK energy needs.
It is estimated that relying more heavily on renewable energy will result in household bills rising by between five and 15 per cent, whilst industry bills could increase by up to 25 per cent on current prices. This makes the need for maximum efficiency clear, both for the profit potential of the supplier, and the cost to the consumer. Renewable energy needs to be made as cost effective as possible if it is to meet the ambitious targets that have been set.
Computational Fluid Dynamics (CFD) is one tool that has already been used to optimize renewable energy production successfully, and so its use in power generation is likely to grow rapidly with any increase in the number of wind farms and tidal power plants. By allowing the user to virtually model air and fluid flow, CFD can help to identify the best site for a potential wind farm, or optimize the efficiency of a wind or tidal turbine rotor.
One such example involved a study of a wind farm at Coal Clough in Lancashire, UK run by Renewable Energy Systems. They were interested in finding out how closely CFD predictions would agree with experimental data taken at the site, with a view to then using the technology to analyze the suitability of other potential sites. An analysis was conducted, orientated in the prevailing wind direction, with sufficient upstream and downstream distance from the existing location of the turbines.
The local topography was taken into account, based on gridded topographic data with a 50 m horizontal resolution, and a 1 m vertical accuracy. A mesh of 1 million cells was generated in which the necessary equations were performed iteratively, to create an overall picture of the farm. The grid was progressively coarsened in the vertical direction with the first cell layer approximately 0.05 m off the ground and gradually increasing to 25 m at the top boundary of the domain. By calibrating the CFD predictions with wind speed measurements taken at the locations of the turbines, the margin of error in initial tests, caused by the variation in topography, was almost entirely eliminated, leading to the production of an accurate wind map, predicting prevailing wind speeds across all points of the site.
Figure 2. Runner modifications taking place in one of Hydro-Quebec’s hydropower plants
A series of further studies focusing on wind speed have been conducted by TUV Nord, one of Germany’s Technical Inspection Agencies. It has utilized CFD analysis methods to study the effects of spacing between individual turbines. The spacing is important because a lesser distance gives rise to wake effects on other turbines downwind, which can lead to decreased and changeable wind loads, reduced energy yield, and vibration-induced fatigue on the rotors and on other nearby structures such as power lines. In order to accurately model the wake effect, blade pitch, wind speed and direction, turbulence intensity and length scale, and rotor speed are input for each simulation. Downstream distances of between six and ten times the rotor diameter have been modelled so far, making it possible to identify and examine variations in wind speed, both within and at the boundary of wake effects resulting from differing turbine spacing arrangements.
The type of information resulting from the two examples given above can be extremely important in developing new wind farms as well as optimizing existing ones. Wind turbines start to generate electricity at speeds of 16 km/h, with maximum rated power output achieved at 53 km/h, meaning electricity is produced for about 80-85 per cent of the time. As wind speed varies over the terrain, and at differing heights, as well as in the wake of individual turbines, such an analysis can be critical in identifying the optimum locations for the turbines to be situated. A further consideration is that the visual impact of a wind farm can be an important factor in a successful wind farm planning application. By locating the turbines at the most efficient point, the number required could be minimized, increasing the likelihood of an outcome in favour of the developer.
CFD was also used in wind power generation in a study of the efficiency of the actual turbine rotor blades. A CFD analysis was conducted on a test carried out by the National Renewable Energy Laboratory in Colorado, USA, looking at wind and angle, calculations of turbine power production, and aerodynamic and structural modes of rotors. An unstructured CFD mesh was used, consisting of 478 664 cells. The computed pressure distribution on the blades was used to determine the shaft power, from which the generation power was derived using available powertrain efficiency data. The computed generator power and operating efficiency that the CFD analysis predicted was found to be within one per cent of the test data from the reported power curve the NREL discovered, proving the accuracy of CFD applications in this area.
The UK is also fortunate enough to have several potentially viable sites for tidal power generators to be installed, as noted by the World Energy Council. One such site is the Severn Estuary where it is estimated that a tidal electricity plant bridging the Bristol Channel could provide seven per cent of the electricity needs for England and Wales. Across the globe, the number of sites available for exploitation is limited by the environmental requirements necessary to make tidal plants possible.
Figure 3. Airflow path lines through the turbine coloured by velocity magnitude
Tides have to vary in level by at least 5 m for the production of electricity to be practical, and a relatively narrow channel is required to build a barrage across, necessary to install the generators through which the tide can flow. At present there is only one such plant operating worldwide at Rance in France. The French government had intended to build several such plants in the 1960s, but changed its policy and focused on nuclear power instead – something the UK appears to have rejected.
The technology for the tidal production of electricity is based on the same principle as that used for hydropower plants on rivers, where the river is dammed and the flow directed through turbines positioned across the dam. While CFD has not been used on a tidal scheme, it has proved successful on such river-based projects. One example is when its use was commissioned by Hydro-Quebec, an electricity producer serving 3.5 million customers in Quebec, Canada. The company used CFD to investigate the reasons why a hydropower plant built in the early 1980s was not working as efficiently as had been expected. At the time of construction, technology enabling an investigation of the problem did not exist but the development of CFD has meant that in more recent years this has become possible.
Using CFD they were able to accurately model the water flow through the 12 identical turbines in operation, examining the water flow across each individual section of every turbine during varying water flow levels. The analysis identified the existence of a large recirculation zone in the elbow of each turbine, as well as a much smaller one in the runner, near the leading edge and crown on the pressure side of the turbine blade.
After various further tests on a series of proposals, a new final design of the runner at the blades’ trailing edge that eliminated the problem was identified and tested virtually using CFD. The modifications suggested were implemented on one of the turbines, and resulted in output being increased by 7.8 MW, raising the weighted efficiency of the turbine by 1.6 per cent. This translated into revenue gains of C$200 000-C$500 000 per year ($142 300-$355 900) for the turbine, making the conversion of the turbines hugely beneficial. This prompted the implementation of a programme to modify the remaining 11 turbines, with three being completed each year. Hydro-Quebec hopes to complete the project before next winter.
It remains to be seen whether the kind of increase in the use of renewable sources of energy put forward in the UK’s White Paper actually materializes. What is more certain however, is that, should the investment be forthcoming, CFD technology is likely to play a major role in making such renewable processes of energy production as efficient as possible, bringing the goal of significant reductions in CO2 emissions ever closer to reality.