Artificial Lift Techbook 2019

Issue link:

Contents of this Issue


Page 47 of 59

46 | April 2019 | ARTIFICIAL LIFT: AUTOMATION AND WELLSITE MONITORING help of these two systems, we are now able to do autonomous control on rod lift as well as for gas lift." Case study Equinor recently deployed Ambyint's IoT-enabled monitor on 50 horizontal wells in the Bakken. The devices replaced programmable logic control (PLC) systems, which Equinor stated were limited in their computational capabilities and were unable to accommodate sophisticated mathematics, which were required to accurately calculate downhole parameters and enable well autonomy. The pilot test included seven months of run time. A case study on the process and its results was recently compiled by Equinor and Ambyint and pre- sented at the 2018 Society of Petroleum Engineers Artificial Life Conference and Exhibition in The Woodlands, Texas. The study reported that the monitoring system offered high-performance computational capabili- ties and direct communication with a cloud-based analytics software platform. The platform was devel- oped to execute higher-order mathematics, artificial intelligence operations and machine learning on high-resolution data, sampled in real time from the rod pump system. "Legacy technologies have helped to keep fields optimized to some extent," the authors reported. "But in the modern, unconventional age these tools are heavily manual, less accurate and, most importantly, unable to fully automate well operations, which was a goal of Equinor." Ambyint's IoT device was connected to the oper- ator's legacy rod pump controller via Modbus connection over two phases. Phase 1 included 20 wells and Phase 2 added 30 wells. "Immediate differences in key downhole parameters were observed when comparing the results from the traditional rod pump controller to the IoT device," the study reported. The data from the rod pump units was fed into Ambyint's machine learning algorithms and identified under-pumping, over-pumping and dialed-in wells. "Using improved downhole infor- mation, Equinor was able to automate well optimization setpoint decisions, resulting in reduced well volatility, bet- ter pump efficiency and increased pump fillage," the authors stated. Equinor and Ambyint identified wells that were either over-pumping or under-pumping in order to optimize stroke-per-minute (SPM) setpoints, which resulted in the operator achieving higher efficiency results with the same or increased production. Equinor found that by implementing the IoT con- trollers, it was able to increase production on the under-pumping wells by 605 bbl, or 33%. For the under-pumping wells, Equinor decreased the number of strokes by 11% and increased pump efficiency by 14%, according to the study. In addition, the authors noted that the SPM reduction on the under-pumping wells resulted in equivalent electricity cost savings from 59 potential fleetwide workovers annually. "This pilot shows that the vision of autono- mous well operations is possible to implement," the authors stated, "and the operator investment in modern optimization technology over and above that which has already been deployed to enable autonomy provides lasting, repeatable value through a multitude of operational parameters." Value proposition The ultimate goal of any new technology in the upstream oil and gas industry is twofold: to increase production and to lower costs while doing it. The same is true for automation systems for artificial lift operations, whose proponents say can help com- panies better meet decline curve expectations and improve recovery rates. "Of the couple of thousand wells that we're on, only about 10% to 15% of those wells are actually optimized," Robart said. "Which means a lot of these Weatherford's ForeSite software integrates physics-based models with advanced data analytics to maximize production. (Photo courtesy of Weatherford)

Articles in this issue

Links on this page

Archives of this issue

view archives of Playbooks - Artificial Lift Techbook 2019