IoT aims to reduce unplanned
downtime, maximize production
The Lower Tertiary in the Gulf of Mexico is estimated to hold as much as 40 MMboe, and is considered to be one of the world’s most promising deepwater frontiers. However, a key challenge of the Tertiary is that the reservoirs are located at great depths, beneath miles of water and layers of salt and
dense rock. Further, these great reservoir depths are characterized
by extremely high-pressure/high-temperature conditions.
These extreme conditions are pushing the technical equipment
limitations and technological boundaries for offshore exploration,
drilling, formation evaluation, well testing and completions. In addition
to these technical and technological limitations, there are also financial limitations. There is an enormous cost and risk associated with
drilling and completing wells in deepwater. In recent years, operator
returns have been further challenged as rig rates, the cost of subsea
equipment, pipelines, steel, and other materials have escalated while
oil and gas commodity prices have remained low.
Innovative technical, operating, and technology solutions are required to improve well completions, boost recovery rates and reduce
costs for offshore oil and gas investments. Operators can seek to leverage innovations in digital technology to optimize asset functionality
and maximize equipment uptime. Specifically, innovative technology
solutions are available that can provide data-driven insights to help
predict potential mechanical failures before they occur and automati-
cally diagnose and remediate equipment issues when they do arise.
This reduces exposure to unplanned downtime from underperforming
assets and supports efforts to inspect, maintain, and repair critical
equipment in compliance with regulatory requirements.
How it works
A key aspect of the technology solution is to leverage the rich streams
of data already being collected from equipment sensors and connected
devices in field production and processing facilities. Typically, a fully
digitalized production site will have hundreds (if not thousands) of sensors capable of generating data at discrete intervals. This data is already
critical to facility operating and control systems in order to provide operators with the information they need to maximize production in real time.
Importantly, this same stream of data can also be pulled into a
cloud-based architecture using Io T (Internet of Things) technology
that can seamlessly integrate with existing facility data collection and
control systems. The Io T architecture allows for massive volumes of
machine-generated data to be collected and aggregated from any con-
nected device, including from dispersed production sites, where it can
then be analyzed to identify issues, trends, and patterns. This process
is commonly referred to as “predictive analytics.”
This powerful data-driven analytics technology considers both
historical data (i.e., what happened before) and transactional data (i.e.,
what is happening now) to make predictions about unknown events
(i.e., what might happen in the future). In this way, the technology
is able to provide customized insights that include: ( 1) the ability
to predict critical equipment failure before it happens - such as an
expected time to failure for a specific piece of equipment or operating component; and ( 2) the ability to streamline the diagnostics and
remediation process for equipment maintenance and repair.
How IoT makes data smart
The Io T technology solution works by applying sophisticated rules
and intelligence to the massive streams of raw data so that it can be
refined into “smart data” that is meaningful and actionable. These
rules are essential. They harness the intellectual capital of production, maintenance, and engineering teams while also leveraging the
advanced machine learning capabilities of the Io T software technology.
Human intelligence reflects the gathering of prescriptive operating
criteria from in-house subject matter experts regarding the unique
equipment, operating, and control parameters of the facility. Generally, this information gathering process includes regular equipment
maintenance schedules, repair history, manufacturer’s specifications,
and expected equipment life. It also identifies the range of normal safe
operating parameters so that alerts can be automatically triggered
when equipment is operating outside of this range.
Further, any unique equipment issues that have been experienced
– coupled with the technical solutions to resolve them – can be
documented and incorporated into the rules. For instance, technical
experts will be able to identify unique equipment issues that have occurred in the past and that can be expected to re-occur in the future.
This expert advice provides guidance as to root cause determination,
the fix/repair that was applied and the outcome achieved (i.e., if this
happens, then that happens, and this is how to fix it). In effect, this
human intelligence “primes the pump” on the front-end to guide the
self-learning capabilities of the software.
Predictive failure defined
The combined application of human intelligence and machine
learning is used to shape the massive streams of historical and transactional data into a precise digital model (a sort of digital twin) for each
The ability to aggregate and analyze information
from dispersed production sites can help to refine
these predictive insights across an entire portfolio
of connected assets. (All images courtesy Bsquare)