Research On The Velocity Field Test Method Of Gas-liquid Two-phase Flow Based On Visual Detectio

Gas-liquid two-phase flow is a complex flow phenomenon.Analyzing and mastering the motion characteristics and flow principle of bubbles in gas-liquid twophase flow is of great significance to improve production efficiency and life quality.Although many scholars at home and abroad have done a lot of research in the field of two-phase flow,at present,the flow characteristics of gas-liquid two-phase flow are not comprehensive,and the research on the detection method of three-dimensional velocity field is not perfect.As a full scene and non-contact testing method,visual detection technology has been applied in the research of multiphase flow.In particular,particle image velocimetry(PIV)technology,as a new measurement technology,has unique advantages in flow field velocity measurement.Taking the bubble flow in gas-liquid two-phase flow as the research object,this work studies the reconstruction method of three-dimensional velocity field of gas-liquid two-phase flow based on binocular particle image.In this work,a binocular particle image acquisition system for bubble flow field was designed and built independently.The flow field image of gas-liquid two-phase flow containing tracer particles was collected by a high-speed camera,and the calibration of binocular camera was completed,providing reliable experimental data for reconstruction and analysis of fluid velocity field.For the reconstruction of two-dimensional velocity field,the cross-correlation algorithm based on window deformation iteration is used to match the particle images at adjacent times.The velocity vector is calculated by matching the displacement information of the region,and then the reconstruction of the two-dimensional velocity field of gas-liquid two-phase flow is realized.PIV image sub-pixel method and standardized median detection algorithm are used to improve the resolution of particle image velocity field,correct the velocity vector with large error,and improve the field reconstruction accuracy of two-dimensional velocity field.The simulation results show that the cross-correlation algorithm based on the window deformation iteration technique has the best reconstruction accuracy,and the PIV image post-processing method can further reduce the reconstruction error and improve the reconstruction accuracy.Aiming at the problem of three-dimensional reconstruction of fluid velocity field,this work combines binocular stereo vision with cross-correlation algorithm based on window iterative technology,and realizes the calculation of three-dimensional velocity vector of particle field by fusing the spatial depth information and time displacement information of tracer particles.Based on the parallax principle,the parallax image of binocular particle image is obtained by census stereo matching algorithm,and then the depth information of particle image is obtained.Simulation results show that the census based stereo matching algorithm has high matching accuracy and can effectively extract the spatial depth information of particle image.Then,the displacement information obtained by the cross-correlation algorithm based on the window iterative technology is fused to obtain the three-dimensional velocity vector of the particle image,and finally the reconstruction of the three-dimensional velocity field of the fluid is realized.Finally,the two-dimensional velocity field reconstruction method and threedimensional velocity field reconstruction method proposed in this work are applied to the bubble flow velocity field reconstruction experiment.Based on the reconstruction results,the motion characteristics of the bubble flow field are analyzed,and the motion law of the bubble flow field is summarized.