Study On Gas/Liquid Two-phase Flow Patterns Based On Conductance Sensor And Image Processing Techniques

Multi-Electrode Array Conductance measurement technique is one of online real-time measurement methods that has been widely researched in recent years, whose main measurement object is two-phase flow or multi-phase flow. We can obtain the characteristic parameters phase volume fraction and flow velocity of two-phase flow with the multi-electrode array conductance sensor, and also calculate other characteristic parameters of two-phase flow through further study. On the principle of conductance sensor measuring, constant current source exciting module is adopted in the construction of Vertical Multi-Electrode Array (VMEA) conductance sensor measuring system, so as to acquire more impedance information of two-phase fluid about different flow patterns. Two cases are designed, namely one based on special waveform generating IC MAX038 and another one based on DDS(Direct Digital Synthesis) technology. Through circuit hardware implementation and performance test comparative analysis, the second solution is adopted and the advantage of constant current mode has been verified through dynamic experiment.Digital image processing as a kind of modern information processing technique has been applied extensively in many different measurements of parameters. The complex fluid structure of two-phase flow will be a certain extent reflected on the dynamic image made by high-speed dynamic camera system. The dynamic experiments of gas/liquid and oil/water two-phase flow are carried out by two-phase flow monitoring system composed of conductance sensor, high-precision differential micro pressure sensor, laser sensor, and high-speed dynamic camera system. Evolvement processes of all the flow patterns are shot completely. Then author describes these 10 flow patterns of gas/liquid two-phase flow combining dynamic image and voltage wavy signals by conductance sensor. With regard to analysis of image characteristics, gray level co-occurrence matrix is a quantitative token for 2D information in local neighborhood of image, and is capable of expressing the image texture fully. 6 characteristic parameters based on gray level co-occurrence matrix are extracted to analyze the evolvement process and complexity of different flow patterns. Also these parameters which are characteristic of temporal and spatial variation are capable of identifying gas/liquid two-phase flow patterns effectively.