Investigating the performance of four-component model in flow loop experiments with sand-gravel mixtures of varying grain size and gradation
Type:
Presented during:
CEDA Dredging Days 2026
Authors:
J.K. de Ridder, A. Visser, E. Meshkati, C. van den Berg and E. de Hoog
Abstract
Accurate estimation of pressure losses is a critical factor in the design and optimization of pump-pipeline systems. The 4-Component Model, widely recognized as the state-of-the-art framework for slurry pipeline transport, describes how individual solid fractions of a mixture contribute to overall transport behaviour and resulting pressure losses (Wilson&Sellgren, 2001; Visintainer et al., 2023). The Durand (1953) model, on the other hand, is a legacy model widely used across dredging and mining industry to estimate pressure losses in the presence of heterogeneous solid transport regime in pipelines. To assess the performance of these models, flow loop experiments were conducted at the Royal IHC test circuit using sand-gravel mixtures with varying median grain sizes and uniformity coefficients (Cu). Tests were performed across a range of mixture flow velocities and volumetric concentrations. Measured pressure gradients were compared with model predictions to evaluate the performance of these models. This study presents the overall predictive capability of 4-Component Model versus Durand model across different solid transport regimes and particle size distribution forms, providing insights into their applicability and limitations for practical slurry pipeline design. It was found that, both models perform satisfactory as overall prediction accuracy of these models fall within 10% accuracy for most cases. However, Durand model in general performed better for cases with uniformity coefficients (Cu) ranging between 1.5 to 6 and the 4 component model for cases with wider PSD grading, Cu around 25.