(2024017) Successful ESP Optimization With Machine Learning Deployed At Scale In The Permian Basin – A Case Study

Many oil and gas companies rely on natural intelligence, resident knowledge, and rules-based logic to optimize production. This is especially true for fields where electric submersible pumps (ESPs) make up a considerable proportion of production on artificial lift. The nature of ESP artificial lift systems makes them well suited for greater remote monitoring, enhanced automation, and implementation of machine learning for autonomous optimization. Extensive use of electric surface controls integrated with downhole sensors provide an ideal operating environment to implement Artificial Intelligence (AI) to achieve autonomous full self-pumping (FSP) operation. However, most operating companies stop short of using automation and machine learning to its full potential. 

This paper will present a case study of an autonomous full self-pumping ESP artificial lift system operating multiple wells in the Permian Basin. The paper will discuss key learning points on how to effectively lead change ensuring field operations and continual innovation are set up to enable success. The overarching goal of the paper is to assist operators in their digital journey by avoiding mistakes in system design and field implementation.

The case study will provide a summary of,
• A field-tested autonomous ESP operating system outlining key components and capabilities. 
• Specialized automation and instrumentation technologies including control and regulation equipment, chemical pumps, and “edge” devices. 
• Developed digital solutions including remote monitoring and autonomous production optimization. 
• Deployment methods to gain acceptance of field personnel and support change management.
• Collaboration of the operating company, ESP supplier, third party partners. 
• Steps to address challenges pumping unconventional wells including rapid decline rates, limited number of field personnel, inconsistencies and biases in optimization tactics, prioritization of uplift opportunities, competing incentives, and uplift vs. ESP run life balancing.

The results of the case study will include,
• Operational benefits including enhanced optimization of ESPs setpoints, improved utilization of personnel, solution scalability, and operational adaptability which favorably impact production, up-time, and run life.
• Development of additional skillsets necessary to supervise autonomous operations.
• Key learnings for successful implementation and continual innovation. 
• Collaboration necessary to break down barriers that can exist between operators, equipment suppliers, and third-party partners. 
• Alignment needed to foster a culture of innovation and “fail forward” mindset; enhanced methods discovered through iteration and continuous improvement.
• Additional benefits including deeper insights into production operations, ESP system technology and software development.