Investigation Of Digital Image Measurement Methods And Its Applications For Gas-liquid Two-phase Flow

In virtue of the image analysis techniques and combined with the aerated water flows questions in hydraulic engineering, the corresponding questions in digital image measurement techniques for gas-liquid two-phase flow are studied. The main conclusions can be draw as follows:(1) A digital image measurement and analysis system fitted to the research of gas-liquid two-phase flow is set up. This system includes two parts of hardware and software. The image acquisition hardware is configured and the aerated water flows experiment setups are designed. The image measurement software presented in this paper has the functions such as image processing, flow visualization, two phase velocity simultaneous measurement, dispersed phase size and concentration information extraction as well as many assistant functions.(2) Based on the thoroughly analysis of two conventional PIV algorithm, a new velocity extracting methods called Pre-estimate Searching Algorithm (PSA) is presented which combines the superiorities of the two conventional PIV algorithms. The experimental results show that the new algorithm has the advantage of high precision and fast computing speed.(3) A new algorithm of correlation based 2-frame Hybrid Particle TackingVelocimetry (2-HPTV) for dispersed phase particles is proposed. The present algorithm has the advantage of improving the computational precision and lowering the requirement of frame rate for the camera, so it has high application values.(4) Wavelet analysis techniques are introduced into image velocimetry. A denoising method based on modulus maximum principle and a particle image inter-layer pairing method based on wavelet image pyramid is proposed. The practice proves that the two wavelet analysis techniques effectively increase the measurement precision and accelerate the computational course.(5) The use of image denoising, gray-scale transformation, histogram equalization, and uneven background correction techniques make the image more easily processed and analyzed. Bubble particles are labeled after exactly separated from the background using a dynamic threshold segmentation technique. The statistics function for bubble particle concentration and size distribution is realized.(6) Distinguish method between bubble particles and flow tracer particle is presented and phase discrimination is successfully achieved. The simultaneous measurements of gas-liquid two-phase velocity field are realized by PSA and 2-HPTV techniques.(7) Flow plunging into water cushion pool phenomenon is investigated under different cases by the method provided by the author. A lot of useful parameters are gain such as velocity fields, turbulent fluctuation kinetic energy fields and vorticity fields. Bubble size, concentration and the simultaneous two phases velocity fields are also gain.