From Routes to Exceptions: Automating Plunger Lift Well Management

Large-scale plunger lift operations demand surveillance methods that can balance proactive optimization with targeted field intervention. To replace route-based monitoring, Occidental Petroleum developed an integrated closed loop control program as well as SCADA based exceptions for its 2,000+ plunger lift wells. Plunger Lift Artificial Intelligence (PLAI) delivers proactive plunger lift optimization by blending real time well data with machine learning and decision logic, enabling timely alerts and automated setpoint updates. By leveraging JSON-based logging, every data point and automated setting change is documented in a structured format, enabling personnel to clearly understand each system action. SCADA is utilized for manual setpoint changes, tracking plunger components and categorizes various alarm types to enable targeted responses. When actions from either remote system prove insufficient, the system allows company personnel to send field callouts for specific well maintenance issues. Implementation challenges include the continued redistribution of stakeholder responsibilities, keeping PLAI’s algorithms and capabilities current and managing automation equipment and reliability. This paper outlines the surveillance framework, discusses implementation challenges, and presents a case study showing efficiency gains from shifting to a combined automated and exception-driven strategy.