Page 139 - Intelligent Digital Oil And Gas Fields

Components of Artificial Intelligence and Data Analytics 103

Table 4.1 Classification of Data Types by Activity, Incumbent in Data-Driven DOF
Operations
Activity Data Types

Production optimization Production data, real-time data (pressure,
temperature, choke settings, gas injection flow,
pump parameters), well models
Well intervention Production data, well logs, geological maps, down-
hole surveys, well files and PVT (pressure,
volume, temperature) data
Field development planning Production data, time-lapse seismic and coring data,
down-hole measurements, geological maps, well
logs and tests, and PVT data
Artificial lift Pump parameters in real time, pump type, and
configuration
Processing of multi-format Seismic data (SEG-Y and SEG-Z formats) and well
data log data (LIS and LAS formats)
Acquisition of fiber optic Permanent reservoir monitoring, distributed
sensing (FOS) data acoustic sensing (DAS), distributed temperature
sensing (DTS)
Acquisition of conventional Seismic, reservoir, log, flow, completion, relational,
data intervention, lab, reservoir, and production
Acquisition of streaming data DTS/DAS and real-time data
Archiving of unstructured Well files, field development reports, drilling
data records, core Images, external studies,
completion, and workover reports

with time. This integration requirement means that the data infrastructure and
architecture must be defined and configured before rolling out the DOF pro-
gram, supported with strong information management and indexing of data.
Table 4.2 shows a decision-support framework for determining the busi-
ness value for data integration. With the assumption that we begin from
nonintegrated data, interpret the table as follows:
• No data integration is needed if the amount of data is <100TB, associ-
ated Health, Safety, and Environment (HSE) risk is zero, investment in
data management is less than $10 million, predicted uncertainty reduc-
tion via integration is less than two points, no associated professional ser-
vice automation (PSA), the project has reached the end of license and
portfolio size exceeds more than the 10 biggest assets.
• Otherwise the business decision should be taken to pursue data integra-
tion; however different combinations of decision attributes from
Table 4.2 are also viable. For example, one possibility is to track the value