Dynamic Behavior Analysis Based Process State Monitoring for Gas-liquid Two Phase Flow in Horizontal Pipe

The process of gas-liquid two phase flow is a dynamic multivariate system affected by multiple complex factors. Focusing on the dynamic behavior in the flow process, a novel strategy based on multiple dynamic principal component analysis (Multi-DPCA) and Dempster-Shafer (D-S) evidence theory is proposed for state monitoring of gas-liquid two-phase flow. Based on multi-sensor acquisition and processing, Multi-DPCA is utilized to simultaneously extract self-correlation, static cross-correlation and dynamic crosscorrelation characteristics embedded in the process data of different flow states. D-S evidence theory is applied to achieve the integration of multiple monitoring indicators. The experimental results demonstrate the feasibility and effectiveness of the strategy with 100% identification rate for three typical flow states and effective monitoring for transitions.