Flow Pattern Identification of Porous Media Based on Signal Feature Extraction and SVM

• In this paper, the visualization experiment of gas-liquid two-phase flow in porous media is carried out. The typical flow patterns of bubbly flow, slug flow and annular flow are photographed by high-speed camera, and the corresponding differential pressure fluctuation signals are measured and recorded, Using probability density function (PDF) and power spectral density (PSD) curves, the time-domain and frequency-domain characteristics of differential pressure signals corresponding to each flow pattern are analyzed, and the quantitative characteristic parameters are introduced to construct the characteristic vector reflecting the time-frequency characteristics of differential pressure signals. A two-phase flow pattern identification method in porous media based on support vector machine (SVM) is proposed. The results show that the overall recognition rate of the three flow patterns measured by the method is 98.18%, which can provide a new technical support for the on-line recognition of gas-liquid two-phase flow patterns in porous media. 

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Maskinteknik -- Strömningsmekanik och akustik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Mechanical Engineering -- Fluid Mechanics and Acoustics (hsv//eng)

Nyckelord

Differential pressure characteristics

Feature extraction

Flow pattern identification

Porous media

Support vector machine

Publikations- och innehållstyp

ref (ämneskategori)

art (ämneskategori)