This paper describes a variable threshold technique that can be applied to any particle image velocimetry (PIV) post-analysis outlier identification algorithm which uses a threshold such as the local median or the cellular neural network techniques. Although these techniques have been shown to work quite well with constant thresholds, the selection of the threshold is not always clear when working with real data. Moreover, if a small threshold is selected, a very large number of valid vectors can be mistakenly rejected. Although careful monitoring may alleviate this danger in many cases, that is not always practical when large data sets are being analysed and there is significant variability in the properties of the vector fields. The method described in this paper adjusts the threshold by calculating a mean variation between a candidate vector and its eight neighbours. The main benefit is that much smaller thresholds can be used without suffering catastrophic loss of valid vectors. The main challenge in obtaining this threshold field is that it must be based on a filtered field to be representative of the underlying velocity field. In this work, a simple median filter which requires no threshold was used for preliminary rejection. A local threshold was then calculated from the mean difference between each vector and its neighbours. The threshold field was also filtered with a Gaussian kernel before use. The algorithm was tested and compared to the base techniques by generating artificial velocity fields with known numbers of spurious vectors. For these tests, the ability of the algorithms to identify bad vectors and preserve good vectors was monitored. In addition, the technique was tested on real PIV data from the developing region of an axisymmetric jet. The variable threshold versions of these algorithms were found to be much less susceptible to erroneously rejecting good vectors. This is because the variable threshold techniques extract information about the local velocity gradient from the data themselves. The user-adjustable parameters for the variable threshold methods were found to be more universal than the constant threshold methods.

No abstract

The effects of vertical confinement on a neutrally-buoyant turbulent round jet discharging from a circular nozzle into quiescent shallow water were investigated. The focus was on identifying changes in the mean flow, turbulence characteristics, and large vortical structures of a horizontal water jet at different degrees of vertical confinement.The confinement resulted from the proximity of a lower solid wall and an upper free surface. The jet exit Reynolds number for all cases was 22,500. The depth of the water layer was the principal parameter. The axial and lateral confinements were negligible.Three different degrees of vertical confinement were investigated in addition to the free jet case. For the confined cases, the water layer depth was 15, 10 and 5 times the jet exit diameter. The centreline of the jet was located midway between the solid wall and the free surface. Particle image velocimetry (PIV) was used to investigate the flow behaviour. Measurements were taken on two orthogonal planes along the jet axis; one parallel and one perpendicular to the free surface. For each case, measurements were taken at three locations downstream of the jet exit where the effects of vertical confinement were expected to be significant. All image pairs were acquired at a frequency of 1 Hz using a 2048 × 2048 pixel camera. This rate was slow enough that the velocity fields were uncorrelated. At each location, two thousand image pairs were acquired in order to extract statistical information about the behaviour of the flow.After completing the cross-correlation analysis of the PIV images and filtering outliers using a cellular neural network with a variable threshold, the statistical quantities such as mean velocities, turbulence intensities, Reynolds shear stress, centreline velocity decay, centreline turbulence intensities, and spread rate were obtained. The proper orthogonal decomposition (POD) technique was applied to the PIV data using the method of snapshots to expose vortical structures. The number of modes used for the POD reconstruction was selected to recover ~40% of the turbulent kinetic energy. An automated method was employed to identify the position, size, and strength of the vortices by searching for closed streamlines in the POD reconstructed velocity fields.iii This step was followed by a statistical study to understand the effect of vertical confinement on the frequency of vortex occurrence, size, strength, rotational sense, and preferred locations.The results showed that the structure of the flow underwent significant changes because of the vertical confinement. The axial velocity profiles in the vertical plane become almost uniform over the entire depth with a mild peak below the centreline of the jet for the shallowest case, while the axial velocity profiles in the horizontal plane are Gaussian but narrower than the free jet profile. The mean vertical and horizontal velocity profiles show that fluid is drawn from the sides of the jet to its centreline and then diverted upward and downward from the jet ax...