The objective of the current work is to present the development of a Particle Tracking Velocimetry (PTV) algorithm for the analysis of oil drops behaviour in two-phase oil–water dispersions within a centrifugal pump impeller. The drop tracking was realized through high-speed camera images in a transparent pump prototype, which enabled the visualization of oil drops dispersed in water in all the impeller channels. The PTV algorithm is based on deep-learning techniques for image processing. The drops are detected by a combined U-Net and Convolutional Neural Network (CNN) method, with the former generating a binary image and the latter detecting valid oil drop contours. After detection, the Labelled Object Velocimetry (LOV) is adopted to calculate the instantaneous oil drop velocity. A synthetic image generator based on a Generative Adversarial Network (GAN) is then developed to assess the results from the U-Net, CNN, and LOV models. Additional validation studies are performed using the results from Perissinotto et al. (2019a). The results reveal that the presented deep-learning PTV algorithm is robust and provides consistent and reliable data for the dispersed oil phase in two-phase oil–water flows.
Tag: Particle tracking velocimetry
Flow visualization in centrifugal pumps: A review of methods and experimental studies
Methods for flow visualization have been decisive for the historical development of fluid mechanics. In recent years, technological advances in cameras, lasers, and other devices improved the accuracy and reliability of methods such as High-Speed Imaging (HSI) and Particle Image Velocimetry (PIV), which have become more efficient in visualizing complex transient flows. Thus, the study of centrifugal pumps now relies on experimental techniques that enable a quantitative characterization of single- and two-phase flows within impellers and diffusers. This is particularly important for oil production, which massively employs the so-called Electrical Submersible Pump (ESP), whose performance depends on the behavior of bubbles and drops inside its impellers. Visualization methods are frequently used to study gas-liquid flows in pumps; however, the visualization of liquid-liquid dispersions is complex and less common, with few publications available. Methods to characterize the motion of gas bubbles are often unsuitable for liquid drops, especially when these drops are arranged as emulsions. In this context, there is room to expand the use of visualization techniques to study liquid-liquid mixtures in pumps, in order to improve the comprehension of phenomena such as effective viscosity and phase inversion with focus on the proposition of mathematical models, for example. This is a main motivation for this paper, which presents a review of researches available in the literature on flow visualization in centrifugal pumps. A broad set of studies are reported to provide the reader with a complete summary of the main practices adopted and results achieved by scientists worldwide. The paper compares the methods, investigates their advantages and limitations, and suggests future studies that may complement the knowledge and fill the current gaps on the visualization of single-phase flows, gas-liquid, and liquid-liquid mixtures.