Recognition method of gas-liquid two-phase flow pattern based on ica and svm

Abstract

The invention discloses a gas-liquid two-phase flow pattern recognition device and method based on ICA and SVM. The flow pattern recognition device includes a standard Venturi tube, a differential pressure transmitter 1, a differential pressure transmitter 2, and an AD converter and computer. Differential pressure transmitter 1 is installed at 45° upward in the vertical section of the Venturi tube, and differential pressure transmitter 2 is installed at 45° downward in the vertical section of the Venturi tube. Differential pressure transmitter 1 and differential pressure transmitter 2 The measured two-way differential pressure signals are input to the computer through the AD converter. In this flow pattern recognition method, firstly, the differential pressure transmitter 1 and the differential pressure transmitter 2 are used to measure two differential pressure signals, and then the ICA method is used to separate the measured two differential pressure signals to obtain two separate signals, and according to The correlation coefficient determines the order and phase of the separation signal, and then extracts the characteristic parameters of the two-phase flow flow pattern recognition. Finally, the extracted characteristic parameters are normalized and used as the training sample set to be input into the support vector machine for training to obtain the flow pattern recognition model. The flow pattern recognition method is simple, easy to realize and good in real time.

Description

technical field [0001] The invention belongs to the technical field of fluid measurement, and in particular relates to a gas-liquid two-phase flow pattern recognition method based on ICA and SVM. Background technique [0002] The phenomenon of gas-liquid two-phase flow is very common in daily life and production, and has an important impact on the development of the whole society. Various gas-liquid two-phase flow systems are included in power generation, chemical industry, petroleum and other industries, such as fractionation process in chemical production, oil and gas mixed transportation process, etc. Different flow patterns of two-phase flow have different characteristics, such as the heat transfer properties of the fluid, the flow parameters of the fluid, and so on. If the flow pattern of the two-phase flow can be effectively identified, it will help to improve the quality of the product, improve the safety of the production process, avoid personal and property losses,...

AI Technical Summary

Problems solved by technology

Although scientific researchers at home and abroad have carried out a lot of theoretical research and experimental tests on flow pattern identification, there are still many problems, such as the selection of characteristic parameters, the coupling effect between gas-liquid two phases, etc.

If the feature parameters are improperly selected and redundant feature parameters are mixed in, it will cause a large amount of calculation and a low recognition rate.

The interaction between the gas phase and the liquid phase leads to the fact that the extracted characteristic parameters cannot effectively characterize the characteristics of the gas phase and liquid phase

Images