Steven McNeil and Jose Gerardo Villela
SSi Lift
Development of AI controllers has allowed abnormal pumping conditions to be accurately identified remotely. Due to the geometry of most rod lifting systems corrective actions need to be undertaken manually at the well location. This paper introduces an automated approach to implementing corrective actions with the use of a non-fixed stroke length rod pumping system. The focus includes case studies where manipulating the stroke length is used to rectify the abnormal pumping condition without the use of additional equipment or service rig intervention.
The method is derived from the experience of field operators who manually manipulate stroke lengths onsite with external lifting equipment or service rigs to rectify the abnormal pumping conditions. Procedures have now been implemented for operators to utilize the non fixed stroke length capabilities of the surface pumping unit to eliminate the need for a service rig or additional equipment. Case studies focus on spacing rods, tagging pumps, freeing stuck pumps and avoiding damaged sections of the downhole pump without the use of additional equipment. Preprogrammed solutions can now be implemented to automate the process and avoid the requirement for human intervention completely.
In each case study the abnormal pumping condition was rectified through manipulation of the stroke length. Wells report a significant reduction in downtime and lost production due to immediate correction of the abnormal pumping conditions. AI POCs coupled with variable stroke length allow automated sequence solutions to be implemented for identifiable abnormal conditions. The geometry of most surface rod pumping units significantly reduce the capability of AI POCs to implement corrective actions autonomously. This technology allows the PLC to bridge the gap from AI identification of abnormal pumping conditions and implementing the corrective actions needed to rectify the condition.
As industry moves further towards AI control it is important to emphasize the mechanical limitations of the surface equipment. Powerful AI programs without the ability to implement solutions erode the value of the AI technology. When designing wells or selecting artificial lift systems it is important to consider the overall efficiency of the system, specifically whether we are limiting the AI with the mechanical geometry of the surface unit.