| The phenomenon of gas-liquid two-phase flow is widely present in power,petroleum,chemical and other industries.The study of gas-liquid two-phase flow in slits is of great significance to the design of "LBB,leak before break"pressure pipelines.This research carried out a visual experimental study of high-pressure argon-water leakage through a long straight slit.The research object of the experiment is a rectangular slit channel with equal cross-section.The length of the slit is 20.30mm,the length of the runner is designed to be 45.00mm,and the width is 0.08mm,0.10mm,0.12mm,0.15mm and 0.18mm.In the experiment,the inlet of the flow channel adopts the method of water inlet on both sides and air inlet in the middle to simulate the flow process of non-uniform fluid entering the slit channel.The experimental inlet pressure value range is 0-5000kPa,the water flow range is 15-450kg/h,and the air flow range is 0-1 kg/h,a total of 834 working conditions.In this study,we found the characteristics of gas-liquid two-phase flow in a slit channel with an equivalent diameter of sub-millimeter through visualization experiments:(1)Through the study of the pressure distribution in the experiment,the pressure drop of the two-phase fluid in the slit has a steep pressure gradient at the outlet position,resulting in an outlet acceleration effect,and the greater the inlet pressure,the more obvious the acceleration effect of the outlet pressure drop.It shows that the slit scale effect is obvious at this scale.(2)The calculation model of the two-phase frictional pressure drop was verified and evaluated using a large amount of experimental data of two-phase flow in this experiment,and it was found that the prediction results of the homogeneous model and the Sun-Mishima relationship are in the best agreement with the experimental data.(3)The radial basis function neural network is used to train 775 sets of experimental data.After training,the pressure drop is predicted to find that the error of the prediction result does not exceed 1%,which proves that with a large amount of basic experimental data,the neural network can be used to correct the rectangle The pressure drop of the slit channel is predicted.(4)With the help of image processing technology to obtain a visualized flow image,it is found that in the slit driven by high pressure,water and gas do not mix into the slit channel artificially,and there is still no common scale in the slit channel.Stratified flow,and mixed strongly at the entrance.(5)The two-phase flow pattern discrimination study and flow trajectory characteristics study were carried out using image processing technology.There are five two-phase flow patterns in the narrow channel in the scope of the experimental working conditions,and the flow pattern distribution map is divided according to a large amount of experimental data,and the movement characteristics of each flow pattern are analyzed,and it is compared with the conventional scale and similar scale flow channels.The flow pattern situation was compared and analyzed.In this visualization experiment,the pressure drop and flow pattern characteristics of the gas-liquid two-phase flow in the sub-millimeter-level slit channel are studied,which provides data support for the safe operation of high-pressure pipelines and LBB design,and further develops the understanding of two-phase flow in slit channels. |
PDF Full Text Request