A new method for all-around identification of two-phase flow pattern based on SVM and ECT

Based on electrical capacitance tomography (ECT) technology and support vector machine (SVM) algorithm, a new method for all-around identification of two-phase flow pattern is proposed combining radial information with axial information of the gas-solid two-phase flow. Compared with the conventional sectional identification method, this method adds another axial one and thus providing more information for the industry. The SVM model of the radial identification is built on basis of the feature parameters extracted from the 28 capacitance values of the 8-electrode capacitance. Then the axial recognition is accomplished with the help of the radial identification results. Both static and dynamic experiments were carried out in lab and the identification accuracy separately reached 96.7% and 95.3%.