Gas-liquid two-phase flow state monitoring method based on multi-modal dynamic nuclear analysis

Abstract

The invention relates to a gas-liquid two-phase flow state monitoring method based on multi-modal dynamic nuclear analysis. The method comprises the following steps: acquiring flow process data X through multi-sensor signal acquisition and processing; establishing a multi-modal DKPCA model for different flow states, and extracting dynamic self-correlation, cross-correlation and nonlinear characteristics from flow state process data; calculating T2 and SPE statistic control limits of each mode; carrying out state monitoring by using a sliding window technology, enabling data at a current sampling moment and sampling data at previous (1'-1) moments to form a new data sample with a window width of l', wherein the condition of l'=l+1 is met; and monitoring the flow state of the gas-liquid two-phase flow based on the established multi-mode DKPCA model and the calculated monitoring statistic control limit.

Description

technical field [0001] The invention belongs to the technical field of multiphase flow process testing, and relates to the realization of flow state monitoring of gas-liquid two-phase flow by using a multiple dynamic kernel principle component analysis method (Multiple dynamic kernel principle component analysis) and a sliding window technology (Moving window). Background technique [0002] Multiphase flow refers to a fluid in which two or more "phases" of substances flow at the same time and have a clear interface, where "phase" is defined as the form of existence of substances, that is, gaseous, liquid or solid. Gas-liquid two-phase flow refers to the flow of gas and liquid at the same time and has a clear interface. It exists widely in nature and many industrial fields such as energy, power, petroleum, chemical industry, metallurgy, and pharmaceuticals. It has complex flow structures, process state fluctuations, The process has the characteristics of many variables, nonli...

AI Technical Summary

Problems solved by technology

[0004] Although dynamic kernel principal component analysis and its related methods are widely used in industrial process monitoring, the value of time lag l does not fully consider the characteristics and operating mechanism of the actual industrial process, such as the evolution of the process state, the duration of alternation, etc. etc.; In addition, there are no literature reports on the research and application of gas-liquid two-phase flow flow state monitoring

Images