Study On Velocity Measurement Method Of Oil-water Two-phase Flow In Small Horizontal Pipe Based On PIV/PTV Technology

At present,most oil fields in China have entered the middle and late stage of development.In order to improve recovery efficiency,water injection,polymer injection,horizontal well and small wellbore are often used.The implementation of the new technology changes the flow state of oil and water,and the production fluid presents the characteristics of high water cut and low flow velocity.The flow pattern is the mixing state of the dispersed oil in the upper layer and the free water in the lower layer.The slippage between oil phase and water phase brings difficulties to the interpretation of logging data and the development of logging instruments.The interpretation of logging data and the accuracy of logging instruments are largely dependent on the flow characteristics.Therefore,it is necessary to study the velocity and velocity vector fields of oil-water two-phase flow in horizontal small-diameter pipelines.Firstly,the camera needs to calibrate during the acquisition process.The light passes through different media,and the image of oil-water two-phase flow in horizontal small diameter pipe is optical distorted.In this paper,according to Snell's law in optical physics and physical parameters such as refractive index of each medium,the pixel position of acquired image is modified to reduce the influence of refraction on fluid velocity.At the same time,in view of the limitations of the image acquisition equipment and the flow channel environment,there are a lot of noise and irrelevant information in the collected image,so the image is preprocessed to reduce the interference and operation of the subsequent processing process.Secondly,the measurement error is large in the process of measuring the velocity of oil-water two-phase flow with high water cut and low flow velocity with PIV algorithm.From the direction of displacement vector calculation,this paper proposes to apply the two-dimensional displacement sub-pixel fitting and the Kriging interpolation algorithm based on gaussian model to PIV algorithm to improve PIV algorithm.It not only overcomes the influence of the method of sub-pixel displacement fitting in X and Y directions independently at first and then combining them on the low accuracy of the position search of the maximum cross-correlation peak in the image,but also improves the accuracy of the blank vector calculation.Finally,PTV algorithm,ORB feature extraction and RANSAC algorithm are improved and combined to propose an oil phase velocity measurement method.The method is improved from the aspect of selecting points in the process of ORB feature description toreduce the computation in the process of feature description.Moreover,considering the model calculation data and observation data of RANSAC algorithm,the accuracy of mathematical model selection can be improved,and the number of correct matching point pairs can be increased while the accuracy of RANSAC algorithm is improved.At last,the improved algorithm is applied to the analysis of oil-water two-phase flow with different water cut and different flow rate.