Modeling flow pattern transitions in electrical submersible pump under gassy flow conditions

• Published in: Journal of petroleum science & engineering 2019-09, Vol.180, p.471-484
• Main Authors: Zhu, Jianjun, Zhang, Jiecheng, Cao, Guangqiang, Zhao, Qingqi, Peng, Jianlin, Zhu, Haiwen, Zhang, Hong-Quan
• Format: Article
• Language: English
• Subjects: Bubble behavior; Electrical submersible pump; Flow pattern transition; Gas-liquid flow; Mechanistic modeling
• ISSN: 0920-4105
• DOI: 10.1016/j.petrol.2019.05.059

Gas entrainment is frequently encountered in electrical submersible pumps (ESP). When occurring, ESP suffers from moderate to severe performance degradation accompanied by unstable operations, depending on the inlet gas volumetric fraction (GVF) and pump rotational speed. The resulted pressure surging and instabilities may cause pump vibrations and short run-life. For a better design of ESP-based production system, the mechanistic model is needed to predict its performance under gas-liquid flow conditions. Similar to multiphase pipe flow, the flow pattern identification and classification inside a rotating ESP is important. In this paper, a new mechanistic model is proposed, which can map different flow patterns inside ESP with gas presence. Experimental results reveal that the boosting pressure of ESP degrades as GVF increases. With ESP's H-Q performance curves obtained, the flow patterns including dispersed bubble flow, bubbly flow, intermittent flow and segregated flow can be identified by deflection points on H-Q curves (Gamboa and Prado, 2008). The transition boundaries between different flow patterns are mapped. Starting from the free body diagram on a stable bubble in the rotating flow field, the transition boundaries of dispersed bubble flow to bubbly flow and bubbly flow to intermittent flow are formulated. Based on the combined momentum equation, the transition criterion of intermittent flow to segregated flow is derived. The flow pattern map predicted by mechanistic model agrees well with that detected from the experimental H-Q curves of ESP under gassy flow conditions. •A mechanistic model to predict transition boundaries of flow patterns inside ESP is proposed.•The flow patterns inside a rotating ESP impeller are recognized and characterized.•The mechanistic model predicts transition boundaries in good agreement with experimental detections.•Mechanistic model can identify severe flow regions for ESP to improve its lifespan.