Pipeline Two-Phase Flow Pressure Drop Algorithm for Multiple Inclinations

A Generalized Additive Model (GAM) is proposed to predict the pressure drop in a gas–liquid two-phase flow at horizontal, vertical, and inclined pipes based on 21 different dimensionless numbers. It fitted from 4605 points, considering fluid pattern classification as Annular, Bubbly, Intermittent, Segregated. The GAM non-parametric method reached high prediction capacity allowed great degree of interpretability (i.e., it helped visualize test statistical inference), that each predictor’s marginal effects could be described, unlike other Machine Learning (ML) methods. model for gradient obtained an adjusted R2 mean relative error 99.1% 12.93%, respectively. This maintained even when ignoring Bubbly training sample. regularization technique filter some variables was used, but most predictors must maintain model’s predictive ability. For example, numbers such Reynolds, Froude, Weber show p-values less than 0.01% explain patterns. performs adequately 500 randomly sampled data points not used fit with lower 15%. variable importance relationship evaluated splines p-values.