Research On Key Techniques Of 3D Measurement On Bubble Behavior In Gas-liquid Two-phase Flow

Measurement in gas-liquid two-phase flow is one of the difficult issues in hydromechanics research field. Bubble behavior in the gas-liquid flow has become a significant subject investigated because it can reflect the interactivity between the gas and liquid which has great sense in chemical reaction field. Most of the traditional measuring methods are belong to contact measurement which will bring interference into the result. Photographic method has been widely applied into the flow field measurement in two dimensions for many years. However, 3D measurement is initially starting both in home and abroad. This subject cooperating with the"interface mass transfer lab"in Chemical Engineering Department in Tianjin University is to explore a set of stereoscopic vision system and image processing system for the specified gas-liquid flow. The target system should capture a series of bubble growth images and calculate several parameters about the bubble. The dissertation has successfully designed a monocular stereoscopic vision system based on the theory of binocular vision system. It finished the camera calibration experiment using the non-coplanar method. Finally, it designed a set of image process software based on Visual C++ which integrated a series of feasible process algorithms. The main works included in the dissertation are shown as follows:1. Constituting the measurement scheme based on 3D-PTV technology by analyzing the simulated gas-liquid flow field, designing the construction of stereoscopic system and image process system.2. Designing a monocular stereoscopic system using one CCD camera. Optimally designed the structure by establishing the imaging model and error analysis model. Finished the mechanical design and done the related experiment.3. Designing the simulated gas-liquid system as a partner. It analyzed the effect of flow rate control, canal materials and the diameter of the injecting nozzle on system.4. Studying the non-coplanar camera calibration method and completing the calibration experiment.5. Designing a set of feasible image process software which implemented algorithms such as odd-even field separation, image filter, contrast stretching, threshold split, bubble extract, bubble filling, edge detect, parameter calculation and so on.