Experimental study demonstrates neural network prediction of two-phase flow rates from differential pressure signals, suggesting viable industrial monitoring.
Gas-liquid two-phase counter-current flow in vertical annulus is involved in multiple industrial fields, due to the influence of the inner pipe, there are difficulties in measuring gas-liquid flow rates. To solve this problem, this paper proposes a new method for two-phase flow measurement based on differential pressure signals and machine learning models. Experiments of gas-liquid two-phase flow were conducted on a vertical annulus pipe with adjustable eccentricity, and the relationships between the probability density function and power spectral density function of two types of differential pressure signals, gas and liquid superficial velocities, and pipe eccentricity were analyzed. An unsupervised classification algorithm based on local density was used to identify the flow patterns. A gas-liquid flow rate prediction model was constructed based on the artificial neural network model and hyper-parameter optimization was performed, achieving an average absolute percentage error of 28.85% for liquid and 8.56% for gas.
No takes yet. Share an insight, caveat, or question.
Cao et al. (2026) studied this question.
Synapse has enriched 2 closely related papers on similar clinical questions. Consider them for comparative context: