Abstract
Electric Submersible Pump (ESP) is very popular artificial lift method to produce at high production rate. However, ESP power consumption is costly and mandate proper optimization to operate at the optimal condition. ESP energy optimization directly improve operational efficiency and reduces lifting cost and greenhouse gas emission.
The ESP power optimization is challenging exercise and requires involvement from multiple organizations. The process has to go through excess of possible combinations of surface choke valve positions and pump speed until reaching an optimized setting. Optimization of ESP operations helps operators to meet production target while maximizing the efficiency of the pump operation. However, it cannot be achieved easily, especially where time and resources are scarce. An innovative approach capitalizing the digital-twin model and artificial intelligence technologies to optimize the ESP performance and adhering to the environmental challenges of reducing CO2 emission has been introduced to expedite the optimization process.
The new approach automatically utilizes ESP real time data, ESP installation data along with the well model to determine the ESP's optimal WHP and ESP frequency. The modelling will be automatically updated based on real time data and provide predictions of recommended operation frequency and optimum wellhead pressure, which is translated to choke size, to operate the ESP at BEP (Best Efficiency Point) and to maximize production rate and minimize power consumption. The optimization process has been implemented in the field resulting in a significant power consumption saving and CO2 emission reduction. Further gains are also expected from ESP performance, resulting in longer ESP run life and reduce number of workover cost.
This paper will share the ESP optimization journey using the advisory tool from obtaining the data, performing the calculations, providing the recommendations, and field implementation of these recommendations.
The best practices and lessons learnt were captured as a reference for future ESP power optimization advisory tool enhancement.
