气-液两相流图像灰度脉动信号的多尺度双分形特性研究

李洪伟; 周云龙

气-液两相流图像灰度脉动信号的多尺度双分形特性研究

李洪伟¹,
周云龙²

1. 华北电力大学能源与动力工程学院, 北京, 102206;
2. 东北电力大学能源与机械工程学院, 吉林, 132012

基金项目:

国家自然科学基金项目(50976018)

详细信息

作者简介:
李洪伟(1982—),男,博士研究生。2005年毕业于东北大学自动化专业。现从事气-液两相流流型识别研究。周云龙(1960—),男,教授,博士研究生导师。1991年毕业于西安交通大学热能工程专业,获博士学位。现从事锅炉水动力特性,油-气-水多相流动特性和气-液两相流流型识别研究

中图分类号: O359.1
计量
- 文章访问数: 27
- HTML全文浏览量: 8
- PDF下载量: 0
- 被引次数: 0
出版历程
- 收稿日期: 2010-05-05
- 网络出版日期: 2025-03-08

Study on Multi-Scale and Double Fractal Characteristics of Gray Pulse Signals in Gas-Liquid Two-Phase Flow Images

LI Hongwei¹,
ZHOU Yunlong²

1. College of Energy Resource and Mechanical Engineering, North China Electric Power University, 102206, Beijing, China;
2. College of Energy Resource and Mechanical Engineering, Northeast Dianli University, Jilin, 132012, China

摘要

摘要: 应用高速摄影技术拍取气-液两相流水平管中3种典型流型的动态图像视频,对每一帧图像的平均灰度脉动信号进行提取;将提取的信号进行多尺度固有模态函数分解,然后与极差/标准偏差(R/S)分析方法相结合,提取各尺度的HURST指数和双分形特征。对气-液两相流的3种典型流型进行了气泡群和单个气泡2种形式的动力学行为分析,应用峭度系数提取对分析结果进行验证,并论述了HURST指数值随气相表观速度的变化情况。结果表明:固有模态函数分解(EMD)结合R/S分析能够很好地揭示气-液两相流的非线性动力学特征。
- 气-液两相流 /
- 图像灰度 /
- 固有模态函数分解 /
- HURST指数 /
- 峭度 /
- 功率谱
Abstract: Dynamic image videos of three typical flow patterns have been taken by a high speed camera in the horizontal tube of the gas-liquid two-phase flow.The average gray scale of each frame of the pulse signals is extracted,which decomposed by the multi-scale intrinsic mode function.Combined with R/S analysis method,the HURST exponent of each scale and double fractal characteristics is extracted.Two forms of the dynamic behavior of the bubbles and the single bubble are analyzed,which belong to the three typical gas-liquid two-phase flow patterns.The results are checked by extracting the kurtosis coefficient,and the variation of HURST exponent with the increase of the gas apparent velocity is discussed.The results show that the EMD decomposition combined with the R/S analysis can commendably reveal the nonlinear kinetic characteristics of gas-liquid two-phase flow.
- Gas-liquid two-phase flow /
- Gray image /
- EMD /
- HURST exponent /
- Kurtosis /
- PSD