Paper: Field Study Verification of Advances In Beam Pumping Diagnostic Software

Today's cutting edge diagnostic software for beam pumping surveillance, analysis, and optimization includes improved methodology based on time-tested techniques as well as practical new functionality. Specifically, this paper will reference dynamometer card pattern matching to aid the well analyst, lease operator, or other interested parties in understanding well operating conditions. This is technology available from the 1980's, but refined in the pattern matching algorithms used and the presentation of results to the user. Available new functionality includes diagnostic reporting that produces a collection of outputs or warnings which are the result of a statistical analysis of surface and downhole card information, calibration or predicted dynamometer card information, and trended data for each beam well addressed by the diagnostic software. The required data is gathered and the resulting calculations are performed each day by the diagnostic software. The purpose is to apply logic that an experienced well analyst would use to determine whether each well needs any corrective action. The diagnostic logic can be customized, allowing users to specify statistical limits for creating warnings. Those diagnostic warnings deemed unnecessary can be de-activated by the user. Major areas of interest for this paper include: 1) recognition of RPC load calibration problems, 2) gearbox torque and pumping unit counterbalance, 3) correct prime mover size, and 4) verification of pattern matching usability. The field test will include beam pumped wells located in conventional primary recovery areas and wells pumping under the influence of injected CO2 for secondary recovery. The data presented in this paper was randomly selected from wells in a 101 well system located in western United States. Average pumping depths ranged from 5400' to 6500'. All wells were equipped with RPCs (Rod Pumped Controllers), calibrated load cells (all with five years or less service), and some combination of magnetic proximity switches and inclinometers for position input. RPC and end device maintenance and software well configuration history should be consider as average.