Unlocking the Potential of Electrical Submersible Pumps: the Successful Testing and Deployment of a Real-Time Artificially Intelligent System, for Failure Prediction, Run Life Extension, and Production Optimization

Abstract

Abstract This paper is a summary of the collaborative work between ADNOC (Abu Dhabi National Oil Company) and nybl, a deep tech development company, and the results of applying nybl's proprietary "Science-Based Artificial Intelligence" to ADNOC Electrical Submersible Pump (ESP) wells in real-time applications. The paper demonstrates the potential benefits of the real-life application of Artificial Intelligence (AI) / Machine Learning (ML) in conjunction with traditional Petroleum Engineering concepts and algorithms to predict imminent and future failures, extend and monitor run life, and maximize the production of ESPs. This paper will highlight ADNOC's innovative approach to pilot new technology through successful deployment on 27 wells, spread onshore and offshore, in real-time, with prescriptive actions.