ENZIN LIMITED LIABILITY COMPANY — Department of Energy SBIR Phase I: 21b

Description

Today, the US Land Unconventional Resource market increasingly deploys advanced geosteering methods with the primary objective of maximizing length of wellbore lateral section in target zone to optimize production. Real time geosteering work flows are best driven by relatively expensive near-bit measurements but instead are generally supported with less effective, low cost MWD solutions that are in excess of 50 feet from the drill bit. This results in less than optimal realized recovery of reservoir reserves. eNZin LLC will employ machine learning methods to build complex models for advanced subterranean sensing techniques that will enable greater adoption of geosteering for well bore placement. Proposed solution will develop machine learning algorithms using independent data sets from near-bit sensors and from MWD tool to effectively replace high end near-bit solutions. We anticipate by providing a markedly lower service cost for equivalent near-bit sensing, the market adoption rate will eclipse the use of competing expensive technologies. This will enable E&P's to significantly improve production recovery rates by placing greater percentage of laterals in target zone to optimize outcome of engineered completions' work flows. The primary challenges to overcome with Phase I development effort are to build a high resolution data set from two or more independent data sets and to establish relationship between the data sets. We will collaborate with Foundation3.ai LLC to complete the following tasks: (1) Develop models to support event synchronization between distinct data sets; (2) Develop models for correlation of structural and dynamic responses of data sets; and (3) Machine learning techniques will be applied on an ongoing basis to adapt developed models from one well to the next adjusting for changes in formation, well bore geometry, drilling parameters, drilling assemblies, etc. Completion of Phase I and II will lead to performance improvements in workflows currently deployed and demonstrate the value proposition of applying advance analytical methods using HPC. Improvement in work flows will also serve to reduce potential environmental risks.