Identification of Two-phase Flow Patterns in Pipeline Based on Zernike Moments and FCM
Two-phase flow patterns in transmission pipeline are identified using the image processing techniques.An experimental system of transmission system is made and a flow pattern identification method based on Zernike moments and fuzzy C-means clustering algorithm(FCM)is proposed.Gas-liquid two phase flow pattern images in the transmission pipeline are captured by a digital high speed camera,and the feature parameters of Zernike moments are extracted by using image processing techniques.Then these feature parameters are used as source data in which is analyzed by the method of FCM,and as a result,the flow pattern intelligent identification is realized.The results indicate that transmission fluid is a gas-liquid two-phase flow in hydraulic transmission lines,whose flow patterns are commonly divided into bubbly flow,plug flow,slug flow and annular flow.The main flow patterns are bubbly flow and slug flow during transmission.This method is a successful one to identify 4 typical flow patterns in the hydraulic-transmission system,and the whole identification accuracy is up to 97％.Due to its high speed and high accuracy,this method is an effective way to research the flow pattern of gas-liquid two phase flow in those pipelines.
万方数据
维普
