Building a better route to green energy

CFD modelling in 3D is essential to the development of ducted wind turbines built into buildings – which could be a better way of harnessing wind energy than simply placing turbines on roofs. As the airflows they harness are enhanced and greatly affected by the buildings in which they are mounted, it is only possible to optimise the turbine design, and its positioning on the buildings, if the airflow against and over building, and through the turbine, is modelled as a whole. Such is the view of Professor David Infield, formerly engaged in studying the problem at Loughborough University and now Professor of Renewable Energy Technologies at the University of Strathclyde. Ducted turbines within the upper parts of buildings are an attractive idea, because they cannot get blown off. And they can take advantage of the ‘blocking’ effect of the building, so that additional airflow over and above that available in free space is forced through them. “The design challenge is to identify a combination of duct profile and rotor that delivers good performance,” he observes. “This has been approached in the past through 1D modelling. In practice, however, such ducted turbines will be mounted on buildings and the influence of the building itself on the flow through the duct will be significant. 1D models cannot capture this complexity and, although a useful guide to outline duct design, more sophisticated flow computation is required.” The study he describes – and presented in a paper in London recently – was undertaken using Ansys CFX. It showed the results of modelling a cylindrical rotor, located on the top of a flat-roofed office building. This is, in effect, a vertical axis rotor laid on its side. The ‘free’ wind speed approaching the building from a distance was 5m/s and the results of the study showed a significant speed-up of flow through the duct. The rotor was located in the throat of the duct, and the challenge then was to come up with a design of rotor and duct that did not block flow to an extent that would deflect air from entering. It was found that the optimal value of K, a dimensionless measure of rotor blockage, depended on the contraction ratio – the ratio of the inlet area to throat area, with lower optimal values of K at higher contraction ratios. In practice, integral ducted turbines that have actually been built have an intake on the side of the building near the top – and then a curved section that turns the airflow upwards through a conventional bladed turbine, with its axis mounted vertically. A project to develop such turbines commercially in Glasgow was undertaken by W K McMillan, in conjunction with researchers at Strathclyde University, aided by a grant of £21,750 from the Carbon Trust. The energy yield of a 1m diameter ducted turbine was estimated at 340kWh/year. A recent paper from Strathclyde suggests that ducted wind turbines should achieve a substantially higher output per square metre of building frontage than photovoltaics in areas such as Scotland, where there is a reasonable wind resource, but “lower levels of sunshine”. Pointers * CFD is crucial to the modelling of ducted wind turbines to be incorporated into buildings * As they make use of the ‘blocking’ effect of buildings, they are an attractive alternative to free-standing, roof-mounted turbines – but have to be modelled in conjunction with the buildings in which they are to be mounted