Technology

DIAL units are the in-well component of the system and each one features up to 6 independent injection orifices with a different port size, offering a wide range of possible injection rates. The injection orifices can be individually opened and closed from the surface without requiring a drop in casing pressure, offering deeper injection and increased production. DIAL units are connected to surface via a tubing encapsulated conductor (TEC) that provides power and communications.

When combined with Silverwell’s Digital Integration Unit (DIU) and associated DIAL-It-In software suite gas lifted well production optimization can be fully automated. Aligning with the widening adoption of digital oilfield production management strategies DIAL systems automate several processes such as well unloading, production optimization, gas usage optimization, and troubleshooting can be automated under the supervision of the operator.

Silverwell understands what the new age of Machine Learning and Artificial Intelligence brings to the oil and gas industry. It provides an unprecedented level of optimization and understanding at a well and field level for both new drills and old. While Silverwell does provide the enabling AI algorithms and software, we understand that the algorithms and software are only a fraction of the solution. The most critical part of an AI System is the data used to calibrate the model.

Traditional gas lift and production systems can provide extensive data. However, it is most often the same data repeated with little variation. With no changes in well or production parameters, unchanging data adds little to a model’s accuracy. Data variations in conventional system occur only when there is an obvious problem or with the rare intervention.

The DIAL system provides new data by changing injection depths and rates at any time without intervention. Conventional systems only provide new data haphazardly as problem and workover information trickles in. With Silverwell, fresh data and unique datasets can be easily acquired through a scientifically designed experimental schedule. This accomplishes accurate and robust training of AI systems in weeks rather than years, all while eliminating the OPEX and environmental impacts associated with traditional gas lift operations.