In many extrusion processes, the metering section is the rate-controlling part of the screw. In this functional zone, the polymer melt is pressurized and readied to be pumped through the die. In Part A, we proposed a new hybrid modeling approach for developing symbolic regression models to predict the flow of shear-thinning polymer melts in multidimensional metering channels. These novel theories remove the need for numerical procedures and can be implemented in both traditional processes – such as screw design and troubleshooting – and in emerging applications – such as digital twins, soft sensors, and predictive control systems. The research presented here (Part B) was carried out in an effort to validate our new melt-conveying models against experimental extrusion data. Extensive experiments were carried out on a well-instrumented laboratory single-screw extruder 19/33D using various materials, screw designs, and processing conditions. In addition, we applied our novel pumping models to network theory which allowed us to replicate the flows in the metering section in silico. Network theory originates from the field of electrical engineering, but has also proven useful in modeling the flows in melt-conveying zones. Our main idea was to reduce the complexity of a multidimensional extruder flow by subdividing the geometry into small passages for which our analytical flow equations are valid. Similar as in electrical circuits, these geometrically simpler sections are connected forming an equivalent flow network, which is then solved in a manner analogous to nodal analysis. The predictions for the axial pressure profile along the screw applying our three-dimensional symbolic regression model are in excellent agreement with the experimental extrusion data, confirming the validity of the approach for a variety of processing conditions.