Modeling consolidation of wax deposition for progressive cavity pump using computational fluid dynamics

Abstract

This study aims to investigate wax deposition in progressive cavity pumps (PCP), which are widely used in the oil industry. Wax buildup on the walls during extraction can significantly impact the efficiency of the pump, with its thickness being determined by both oil properties and PCP operating conditions. Using computational fluid dynamics (CFD), a detailed analysis of the effect of different specifications, such as rod diameter, rotation speed, inlet velocity, dynamic viscosity, and rod height on wax deposition is conducted. The study also considers the impact of rotational momentum on swirl flow and includes a large-scale simulation of 3,000 cases across a wide range of conditions. Shear rate, a crucial parameter in determining wax thickness, is analyzed and modeled through regression using deep neural networks. This regression model can be used to predict wax thickness based on factors such as inlet velocities, rod diameters, viscosities, and rotation speeds. The results of these large-scale simulations and the proposed regression model will aid in understanding the relationship between operating conditions and oil properties in the context of PCP systems.

Keywords

• Computational fluid dynamics
• Deep neural network
• Progressive cavity pump
• Modeling
• Shear rate
• Wax