Two phase fluid phase concentration measuring method based on main component analysis and neuron network

Abstract

In the capacitance tomography system composed of the capacitance sensor array, the electronic measurement circuit and the computer for image formation, the method of principal component analysis is introduced to analyze the normalized value of capacitance measured. These components with contribution rate less than mu are removed from the principal components. The data of principal components withcontribution rate larger than mu are as the input for the forwarding neuron network. The output of the neuron network is the phase concentration. The substance of the principal component analysis is that the normalized sample matrix of the capacitance values measured is transformed to the correlation matrix, from which each eigenvalue and related principal component is obtained.

Description

technical field [0001] The invention relates to a two-phase flow phase concentration measurement method based on principal component analysis and neuron network, which belongs to the field of two-phase flow measurement technology and relates to information mining technology in the field of process flow imaging research. Background technique [0002] Two-phase flow has more complex flow characteristics than single-phase flow. Due to the interface effect and relative velocity between the phases of two-phase flow, the phase interface is randomly variable in time and space, resulting in a variety of flow structures, and these changes are random, resulting in extremely complex flow characteristics. In view of the complexity and randomness of two-phase flow, it is quite difficult to detect the parameters of two-phase flow. To recognize the complex phenomena of the two-phase flow system, reveal the mechanism of the two-phase flow, establish a two-phase flow model, and predict, des...

AI Technical Summary

Problems solved by technology

[0010] In view of the shortcomings of the above methods, we use data mining technology on the basis of the hardware of the electrical capacitance tomography system, and use a two-phase flow concentration measurement method based on principal component analysis and neural network technology, which can be obtained from the tomography system. The phase concentration value is directly calculated from the raw measurement data provided by the array sensor, which solves the problem that the current measurement results rely on the image reconstruction algorithm when using tomographic imaging technology to measure the phase concentration of two-phase flow

Images