Final Report Summary - MULTISOLVE (Conceptually new, multi-scale solutions to industrial flow problems.)
This project explored the wide range of potential industrial applications of multiscale- and/or fractal-generated turbulent flows and in some cases provided proof of concept evidence for some of these applications.
A broad panoply of computational and experimental techniques was used as well as a number of computational and experimental facilities including ARCHER, the UK National Supercomputing Service and wind tunnels at Imperial College London (UK), University of Oldenburg (Germany), Ecole Centrale de Lille and Universite de Poitiers (France), University of Ottawa (Canada) and University of Nagoya (Japan). Secondments at BAE Systems (UK), CEA (France) and MEL Consultants (Australia) were instrumental in obtaining results and reaching conclusions.
The first potential application is multiscale trailing edges for wings. Tests with unsteady inflow conditions revealed that they damp undesirable fluctuations in wing loading. Such wing designs offer potential for improving the performance and durability of applications involving wings subject to highly unsteady inflows, such as wind turbine blades, multi-element wings or cascades, and turbo-machinery stages.
The second potential application is in dynamic mixers. Mixers are used in very many different industries including the pharmaceutical, chemical, food and beverage industries to name a few. It was found that fractal blades draw less power for a given shaft rotational speed than other regularly used blades and that fractal blades also achieve mixing faster than regular blades. There is therefore clear potential for energy-efficient accelerated mixing with fractal dynamic mixers which now needs to be investigated on a case-by-case basis in various industrial settings.
The third potential application is in wind tunnel simulations of environmental conditions for wind load tests of scaled-down structures such as bridges and skyscrapers. A new non-uniform multiscale grid was invented for easy use at wind tunnels inlets. These grids can be designed to generate bespoke turbulent flow conditions in the wind tunnel which mimick in a scaled-down way given large-scale turbulent flow conditions in the location where the bridge or skyscraper is to be built. Compared to existing methods in the wind engineering industry our new non-uniform multiscale grids are easy to design and manufacture, easy to use, and require much less trial and error iterations. However, the range of applicability is restricted for the moment to relatively weak mean shears. A patent application has been filed and further work will also be needed in the future to broaden the range of environmental conditions where this new method can be applied.
A fourth research direction concerned aero-acoustics. Aero-acoustic characterization of the flow revealed that the main source of noise from wall-mounted grids/fences was the interaction of the flow with the fences themselves and that the presence of the wall was unimportant. A parallel investigation within the same project showed that vortex shedding from, and therefore vibrations on, fractal/multiscale grids/fences can be greatly reduced compared to non-fractal fences even when the non-fractal fences have conventional vortex shedding reduction devices.
The breadth of computation and experimental methods and facilities and of turbulent flows studied led naturally to advances in our fundamental understanding of turbulent flows and in methods to study them. Such knowledge is of paramount importance as it can underpin unexpected future applications in any industrial setting. For example, (i) a novel and robust post-processing method was developed to estimate the mean skin friction from the mean velocity profile with an uncertainty below 1%. (ii) A brand new calibration technique was developed for planar laser induced fluorescence measurements to account for secondary fluorescence and non-linear response of the fluorescent dye to the incident light source. (iii) Theoretical advances were made concerning energy exchanges across scales and spatial position, both for single and for interacting wakes. (iv) The importance of the fluctuating pressure field in the generation of fluctuating velocity gradients was highlighted and the relation between the fluctuating wall shear stress and intermittent velocity fluctuations away from the wall was elucidated. This list is not exhaustive; many more fundamental and applied results and conclusions can be found in the project's publications, see http://www.multisolve.eu/.
A broad panoply of computational and experimental techniques was used as well as a number of computational and experimental facilities including ARCHER, the UK National Supercomputing Service and wind tunnels at Imperial College London (UK), University of Oldenburg (Germany), Ecole Centrale de Lille and Universite de Poitiers (France), University of Ottawa (Canada) and University of Nagoya (Japan). Secondments at BAE Systems (UK), CEA (France) and MEL Consultants (Australia) were instrumental in obtaining results and reaching conclusions.
The first potential application is multiscale trailing edges for wings. Tests with unsteady inflow conditions revealed that they damp undesirable fluctuations in wing loading. Such wing designs offer potential for improving the performance and durability of applications involving wings subject to highly unsteady inflows, such as wind turbine blades, multi-element wings or cascades, and turbo-machinery stages.
The second potential application is in dynamic mixers. Mixers are used in very many different industries including the pharmaceutical, chemical, food and beverage industries to name a few. It was found that fractal blades draw less power for a given shaft rotational speed than other regularly used blades and that fractal blades also achieve mixing faster than regular blades. There is therefore clear potential for energy-efficient accelerated mixing with fractal dynamic mixers which now needs to be investigated on a case-by-case basis in various industrial settings.
The third potential application is in wind tunnel simulations of environmental conditions for wind load tests of scaled-down structures such as bridges and skyscrapers. A new non-uniform multiscale grid was invented for easy use at wind tunnels inlets. These grids can be designed to generate bespoke turbulent flow conditions in the wind tunnel which mimick in a scaled-down way given large-scale turbulent flow conditions in the location where the bridge or skyscraper is to be built. Compared to existing methods in the wind engineering industry our new non-uniform multiscale grids are easy to design and manufacture, easy to use, and require much less trial and error iterations. However, the range of applicability is restricted for the moment to relatively weak mean shears. A patent application has been filed and further work will also be needed in the future to broaden the range of environmental conditions where this new method can be applied.
A fourth research direction concerned aero-acoustics. Aero-acoustic characterization of the flow revealed that the main source of noise from wall-mounted grids/fences was the interaction of the flow with the fences themselves and that the presence of the wall was unimportant. A parallel investigation within the same project showed that vortex shedding from, and therefore vibrations on, fractal/multiscale grids/fences can be greatly reduced compared to non-fractal fences even when the non-fractal fences have conventional vortex shedding reduction devices.
The breadth of computation and experimental methods and facilities and of turbulent flows studied led naturally to advances in our fundamental understanding of turbulent flows and in methods to study them. Such knowledge is of paramount importance as it can underpin unexpected future applications in any industrial setting. For example, (i) a novel and robust post-processing method was developed to estimate the mean skin friction from the mean velocity profile with an uncertainty below 1%. (ii) A brand new calibration technique was developed for planar laser induced fluorescence measurements to account for secondary fluorescence and non-linear response of the fluorescent dye to the incident light source. (iii) Theoretical advances were made concerning energy exchanges across scales and spatial position, both for single and for interacting wakes. (iv) The importance of the fluctuating pressure field in the generation of fluctuating velocity gradients was highlighted and the relation between the fluctuating wall shear stress and intermittent velocity fluctuations away from the wall was elucidated. This list is not exhaustive; many more fundamental and applied results and conclusions can be found in the project's publications, see http://www.multisolve.eu/.