光流
流量(数学)
算法
两相流
明渠流量
稳健性(进化)
计算机科学
数学
人工智能
化学
几何学
图像(数学)
生物化学
基因
作者
Jinqiu Lv,Haifeng Ji,Yandan Jiang,Baoliang Wang
标识
DOI:10.1109/jsen.2023.3321632
摘要
A new flow pattern identification method based on the optical flow algorithm for gas–liquid two-phase flow in small channels is presented. In this method, an improved optical flow algorithm, the optical flow algorithm based on curvature domain descriptor, is proposed to overcome the influence of uneven illumination. Then, the characteristic of the optical flow field is analyzed, and two features of the optical flow field, the mean value and standard deviation of each column, are extracted. Further, principal component analysis (PCA) and K-means clustering algorithm are applied to implement flow pattern identification of the two-phase flow. The evaluation of the improved optical flow algorithm was performed on two optical flow datasets. The results indicate that the improved optical flow algorithm can increase the accuracy of optical flow estimation and improve the robustness of the algorithm to illumination changes. Flow pattern identification experiment of the gas–liquid two-phase flow was carried out in a small channel with an inner diameter of 4.23 mm. The experimental results show that the proposed flow pattern identification method is effective. With the proposed method, the accuracies of flow pattern identification for typical flow patterns are all above 92.5%.
科研通智能强力驱动
Strongly Powered by AbleSci AI