Hydrocarbon Processing - September 2021 - 41

Special Focus Refining Technology
J. CHELLANI, KBC, a Yokogawa Company, Abu Dhabi,
United Arab Emirates; and G. WINTER and A. ALSHAIKH,
Saudi Aramco, Dhahran, Saudi Arabia
Enhanced unit monitoring enabled by digital twin
Process engineers must continuously
monitor the health and performance of
process units. Unit monitoring ensures
reliable operation of the unit, identifies
disturbances early in the cycle and improves
the understanding of the process.
In addition, effective unit monitoring
results in the identification of shortand
long-term economic opportunities.
These opportunities typically reside in
yield optimization, cost reduction and
maximization of unit capacity.
Unit engineers use key performance
indicators (KPIs) to help define and
measure progress toward the plant's business
and strategic goals. They facilitate
the management process by focusing
employees on the critical measures of
the plant's performance and keeping the
achieved results in open view of the staff.
KPIs are quantifiable measurements-agreed
to in advance-that reflect
the critical success factors of a plant.
They must reflect the organization's
strategic goals and must be quantifiable.
The key aspect of developing KPIs is in
focusing them to ensure that only those
KPIs that are truly reflective of plant performance
are tracked and reported.
Pacesetters use a KPI tree that tracks
performance of the assets and the overall
site at different levels. Each level of the
KPI tree has certain objectives associated
with tracking and improving the performance
of the overall site. The targets in
this KPI tree usually follow a top-down
approach, whereas the flow of information
for reporting actual performance is
through a bottom-up approach.
Monitoring the performance data of
process units and key equipment is one of
the main inputs for the KPI tree. The rest
of the information at the different levels
is estimated using unit-level KPIs. In the
past, unit engineers have usually relied
on raw data taken out of trends from the
historian or on a laboratory information
management system (LIMS). Experience
in operating the unit, along with performing
manual reconciliation, has been used
by unit engineers to turn this into useful
data; however, it is a time-consuming activity.
In addition, the analysis is dependent
on the individual's expertise, which
is difficult to hand over to the next generation.
Visibility of the analysis performed
by individuals, and collaboration with
peers, are also some of the challenges associated
with unit monitoring.
Process digital twin. Increasing process
plant complexity requires more sophisticated
ways of approaching KPIs and targets.
This is where real-time connected technology
with deeper analytics-the process
digital twin-comes to the forefront, as it
accounts for the multidimensional factors
and nonlinear trade-offs that make effective
decision making a challenge.
These advanced analytics enable personnel
to see inside assets and processes,
and to perceive things that are not being
directly measured. They are wired so that
insights are instantly available without
data and model wrangling by end users
and are run in a consistent way that everyone
can understand and agree on. In this
manner, the process digital twin drives
agility and convergence in understanding
actions across the whole business. Key
inputs used for a process digital twin are
shown in FIG. 1.
Process digital twins are created using
applications that are linked to data historians,
allowing them to gather and process
data directly from the historian without
manual intervention. The process digital
twin uses a standard, automated methodology
to perform the following key tasks:
* Retrieve and screen process
and lab data
* Reconcile unit material balances
* Calculate key, unmeasured
process variables
* Calculate a set of KPI reporting
information and targets
* Assess the health of the tools,
including the linear programming
(LP) model
FIG. 1. Inputs used for a process digital twin.
Hydrocarbon Processing | SEPTEMBER 2021 41
Hydrocarbon Processing - September 2021

Table of Contents for the Digital Edition of Hydrocarbon Processing - September 2021
