Measurement Of Turbulence In A Particle-laden Channel Flow Using PIV Technique

Previous investigations of particle-laden flows have demonstrated that loading of small, dense particles in gas flow can greatly influence the fluid turbulence level. However, the underlying physical mechanisms responsible for the turbulence modification are not well understood. The objective of this research was to investigate experimentally the turbulence modulation of dilute particle-laden flow in a horizontal channel.Experiments were conducted in a horizontal, fully-developed channel flow at a Reynolds number of 590(based on friction velocity). As a new field measurement technique of flow, particle image velocimetry (PIV) technique has been applied in a lot of aspect of research. So there have been much more concern on the robust and accuracy of the PIV algorithm. Firstly, the gaseous turbulence statistics of single phase flow, derived from PIV data, were compared with documented DNS results, and it demonstrated the feasibility of turbulence measurement using PIV technology. To obtain high resolution and simultaneous gas/particle flow fields measurement using PIV, dramatic improvements to the traditional PIV algorithm were made: using the method of image processing, the recognition and discrimination of two-phase image were realized based on statistics difference of intensity and pixel numbers between seeder and particle images; Considering the confinement of traditional PIV algorithm, a new PTV algorithm based on correlation was developed and realized, which can improve the spatial resolution and accuracy, especially in measurement of flow with high velocity gradient. Based the self-developed two-phase PTV algorithm and phase discrimination method, experimental measurements of channel flow laden with nominal monodisperse, spherical polystyrene particles (with density of 1030kg/m3 and diameter of 67μm) were conducted. It revealed that, the turbulence modification effects by particle was obvious even with particle mass loading as low as 1%, and its behavior near the bottom wall was different to those in the middle of channel.