POWER February 2020 - 35

CONNECTED PLANT
and optimization of industrial processes.
They are typically involved in analyzing,
upgrading, modifying, and optimizing
equipment and production processes.
If the production process is stagnating
or under-performing, they need to
figure out why, but also in the shortest
time possible to avoid production losses,
maintain product quality, and elude high
maintenance and repair costs.
Implementing self-service industrial
analytics can enable engineers to get
more robust and faster insights into their
operational production data. It enables
them to identify new areas for performance
optimization with advanced root
cause analysis capabilities, monitor production
to avoid abnormal situations, and
even predict future evolutions of batch
runs, transitions, or equipment startups
in minutes. It enables the business users,
such as process and asset engineers, to:
■ Solve previously
performance issues.
unsolved process
■ Verify hypotheses and prove them to
be either true or false, so they can be
addressed or ruled out for the future.
■ Find new ways to improve production
performance, because data with captured
events and early warnings provides
new insights.
■ Use contextual information from thirdparty
business applications to get new
insights in operational performance.
■ Use actionable dashboards to monitor
operational performance in real-time
(Figure 2).
By working on all these situations, the
result is a higher overall equipment efficiency,
where the asset performance is
optimized within its operational context,
leading to waste reduction, higher and
cleaner uptime, and reduction of the carbon
footprint of the plant.
Analyze, Monitor, and Predict WAGES
Consumption
The major process-related energy consumers
can be easily remembered using
the acronym WAGES, which stands for
hot water, air, gas, electricity, and steam.
WAGES can be directly or indirectly analyzed
through all the sensor data. The
data can be analyzed descriptively to see
what has happened, which provides the
best understanding if a long period of performance
can be assessed. Sometimes,
certain issues happen only a couple of
times per year, but can have a big impact
on energy consumption, for example, a
trip causing a shutdown. Discovery anaFebruary
2020 | POWER
Due to external factors, such as unplanned
downtime and market-driven load
reductions, the team needed to dig deeper
to measure the actual performance of
the unit. To do this, researchers compared
the periods of good performance with the
periods of bad performance.
The engineers found that the gas
2. Analytics-driven production cockpit, helping
engineers to monitor and continuously
improve overall equipment effectiveness.
Courtesy: TrendMiner
lytics helps to understand what has happened
and through diagnostic analytics
the organization can start monitoring the
performance of the site.
Because asset performance is contextualized
by the process it functions
in, best operating zones or best performance
windows need to be subtracted
from actual process behavior rather than
theoretical data. Historical data fingerprints
can be created to monitor good
and bad behavior, optionally with an energy
consumption focus. Next, live operational
performance data can be used
for predictive analytics; for example, performance
downstream might be caused
by behavior an hour or more upstream.
Practical Use Cases
Advanced analytics have been applied
to analyze, monitor, and predict process
and asset performance for many energy
management-related use cases. Two
instances are described below.
Increased Efficiency of a Power Generation
Plant. A typical combined cycle
gas power plant for producing electricity
is equipped with two turbines that power
a generator. The primary turbine is powered
by hot air, which is generated by
burning the gas. The secondary turbine is
powered by steam that is generated by
heating up water with the exhaust gases
coming from the primary turbine.
Over time, the performance of a unit
at one particular power station began
to worsen, leading to capacity loss and
lost revenue. The loss was gradual, so
the change went unnoticed at first. A
decrease in performance is usually a
gradual phenomenon that only comes to
surface after a long period of time. With
the self-service advanced analytics tool,
the engineers could clearly visualize operations
over time. Using a value-based
search, the decline in performance was
quickly confirmed and quantified. This
formed the foundation for the rest of the
troubleshooting analysis.
www.powermag.com
fuel flow, the compressor discharge
temperature, and the inlet guide vanes
reference angle differed consistently
for both groups of layers. With the selfservice
analytics tool it could be proven
that the root cause of the problem was
the non-calibrated inlet guide vanes,
which impeded air and fuel supply, and
ultimately power generation in the gas
turbine. In only four hours, the engineers
efficiently identified and resolved
the 2-MW capacity loss without the
need of data scientists.
Control Energy Consumption within
the Cooling Water Network. A number
of reactors were consuming cooling
capacity from the utility network for
cooling water. Sufficient cooling capacity
is critical for these reactors, as thermal
runaway could occur when the available
capacity is insufficient.
To avoid this undesirable situation, advanced
analytics were set up to monitor
the cooling capacity in real time. Early
warnings were created and designed to
trigger only on actual problem situations,
avoiding false positive alarms that could
be triggered by measurement noise or
spikes in the data. Now, upon a warning,
there is ample time for the process engineer
and operators to re-balance the reactors
and deprioritize other equipment,
so that the critical components can consume
the maximum cooling capacity
while keeping overall energy consumption
within target boundaries.
Energy management is not new; many
companies have a structured energy management
program in place. However, new
self-service analytics tools allow subject
matter experts to analyze, monitor, and
predict process and asset performance
more quickly and accurately, which can
result in a huge contribution to meet the
organizational carbon-footprint goals. This
is especially important after the low-hanging
fruit for energy savings have been
picked and more knowledge is needed to
improve operational performance. Companies
benefit by improving overall profitability
and increasing safety. ■
-Edwin van Dijk is vice president of
marketing with TrendMiner, a company
that develops advanced analytics
solutions for the process industry.
35
http://www.powermag.com
POWER February 2020

Table of Contents for the Digital Edition of POWER February 2020
