MARCH 2013

The “Better Business” Publication Serving the Exploration / Drilling / Production Industry

Dynamic Simulation Solves Problems

By Glenn R. Dissinger of intermediate hold-up tanks can have a come more difficult to control and operate.
significant impact on the operability of a Dynamic simulation has been used to
HOUSTON–Dynamic simulation is a process by dampening disturbances as evaluate the performance and operation
process engineering technology that pre- they move from upstream equipment to of heat-integrated processes for potential
dicts how a process and its associated downstream units. While larger tanks transient operational problems.
control system will respond to various give better control and operability per- Then there is the design and analysis
disturbances as a function of time. formance, they are more costly, tie up in- of batch and cyclic processes. Batch and
Companies have applied dynamic sim- termediate inventory of chemicals, and cyclic processes, including pressure and
ulation successfully for critical applications often are a source of environmental, emis- temperature swing adsorption processes,
for more than 30 years. However, as sion, and safety problems. have been common in the specialty chem-
process, control and operating engineers For applications such as slug catchers icals and air separation industries. How-
face the more stringent demands on on offshore platforms (where size and ever, their use in the petrochemical and
today’s process plants, dynamic simulation weight are critical considerations), dy- petroleum industries has grown signifi-
technology to address design, control, namic simulation can be used to properly cantly for such applications as capturing/re-
operability and training issues is being minimize surge capacity while providing mediating carbon dioxide and greenhouse
increasingly applied. sufficient attenuation of process distur- gases, and for separating higher-value
Dynamic simulation can improve both bances. compounds (e.g., xylenes isomers) as pe-
the design and operations of a plant in a Another application is in heat-integrated troleum companies have embraced “mol-
variety of areas, including improved and processes. With crude oil prices hovering ecule management” to improve profits
more robust designs and process control, near $100 a barrel, the number of energy and value extracted from a barrel of
safety and reliability applications, de- efficiency projects and retrofits utilizing crude. Dynamic simulation is a critical
ploying models into on-line applications, pinch analysis and heat integration has technology for properly designing and
and operator training. increased. However, as plants become analyzing these inherently transient
Like other modeling techniques, the more heat-integrated, they sometimes be- processes.
key benefit of any dynamic model comes
from the improved process understanding
it can give a user. By understanding the
process (with or without the control sys-
tem) and its possible transient responses
to both planned and unplanned distur-
bances, improvements in the design, con-
trol, operation and safety of the process
can be achieved.
Process Design
While process design traditionally has
used only steady state simulators, many
process design decisions require knowl-
edge of the transient response and inter-
actions of the process. There are several
uses of dynamic simulation within process
design. Properly designing the flare header and stack is critical for offshore platforms.
One is surge capacity sizing. The size

Reproduced for Aspen Technology with permission from The American Oil & Gas Reporter www.aogr.com
SpecialReport: Production and Processing

cation, and can quantify the potential

economic and operational improvements.
Properly designing and implementing
advance control strategies, such as mul-
tivariable and model predictive control
algorithms, requires knowledge of the
transient responses of the process to
known disturbances. Known as the process
identification step, this knowledge usually
has been obtained through extensive plant
testing. Such testing, however, is expensive
and time-consuming.
A new best practice of using rigorous,
validated dynamic models in conjunction
with limited plant testing has emerged,
reducing the time and cost of the process
identification step. Likewise, once the
advance control algorithm has been de-
veloped, it can be tested using the dynamic
model to evaluate its performance prior
Process models built using dynamic simulation software can evaluate the performance to being implementing in the plant.
of control systems quickly and cost-effectively. Dynamic simulation also is used for
operability studies. While it is obvious
Finally, dynamic simulation is used production, safety, and environmental re- to anyone who has worked in a plant, it
in process improvements. Once a plant quirements. needs to be stated that processes may ap-
is in operation, engineers continually Using dynamic simulation, a process proximate, but never actually operate, at
look for ways to improve product quality, control engineer can quickly develop, the steady-state design operating condition.
minimize waste and emissions, maximize evaluate, test, and tune novel control Unplanned disturbances such as feed
yield, reduce energy and utility costs, strategies for both new and existing process- composition fluctuations, changes in
and increase capacity while responding es. In addition, the dynamic model often utility loads and availability, changes in
to both planned and unplanned distur- is a critical technology used to demonstrate environmental ambient conditions, and
bances. Dynamic simulation often is used the performance of the new strategy to catalyst activity all impact how the process
to analyze the impact and quantify the plant engineers and operators, in order to operates minute by minute.
value of these potential improvements gain their acceptance for the change. Other process “disturbances” are
on plant operations and control systems. Using composition analyzers in process planned, such as product campaign switch-
control loops has become an accepted overs, changes in production load targets,
Process Control, Operability technology. Dynamic simulation can eval- and startup and shutdown of the process.
Traditionally, process control has been uate the benefit of adding an on-line an- All these disturbances cause the plant to
designed at the end of process design alyzer, including identifying its best lo- be constantly in a transient state, and can
cycle, after the process flow diagram
(PFD) has been completed. This practice
sometimes leads to a process that requires
an elaborate control strategy or one that
is inherently hard to operate. Using dy-
namic simulation earlier in the design
process while the PFD is still under de-
velopment can identify important oper-
ability and control issues, leading to a
better process design and a smoother op-
erating plant. Again, there are several
uses for dynamic simulation in process
control and operability.
One example is developing and ana-
lyzing control strategies. Many control
strategies are developed using rules of
thumb and experience. However, such
methods often fall short as plants today
This rigorous dynamic model of a pipeline pigging operation was deployed to operations
have become more integrated and are personnel using a simplified Excel-based interface.
operated more aggressively to meet quality,
SpecialReport: Production and Processing

be only truly evaluated rigorously by operation of a process significantly. dynamic model, which can accept opera-
using dynamic simulation. tional data and run faster than real time,
On-Line Applications can predict the possibility of slugs in the
Safety Studies Rigorous dynamic models have been pipeline, allowing operators to anticipate
Safety is (or should be) job number used as part of an operator training system the arrival of the slug flow and make any
one within any plant or facility. Accidents for many years. More recently, interest necessary adjustments to level and pressure
caused by poor design or improper oper- in moving dynamic models into on-line set points at the onshore facility.
ator responses can lead to damaged equip- applications has grown considerably. An example of such an application is
ment, loss of production, and even injuries Among the current and potential on- the production surveillance done by
or death. Dynamic models have played a line applications is distributed control Process Ecology, a Calgary-based engi-
key role in identifying potentially haz- system (DCS) checkout. DCS configura- neering consulting company, for a large
ardous situations and the changes in tion and checking the process control independent North American oil and nat-
design and operation policies needed to strategy is a time intensive exercise during ural gas company.
mitigate or avoid them. any new DCS plant implementation or Process Ecology developed a dynamic
Among the examples of how dynamic extensive process or control modification. model of the production asset, and cali-
simulation models are used for safety A dynamic model can be used to assist brated the model based on historical op-
studies are designing and analyzing emer- in testing the control strategy prior to erational data. The dynamic model was
gency and relief systems. Emergency and putting it on line for real operation. used by operations personnel as a pro-
pressure relief systems are critical parts The dynamic model is linked to the duction surveillance application, helping
of a process. Properly designing these DCS control system, and emulates the them anticipate changes in production
systems is required to protect both the actual operating process to test the control and calculating real-time well flow rates
safety of personnel and the capital invested algorithms’ responses to various distur- where physical instrumentation was not
in process equipment during severe op- bances. Configuration errors in DCS con- available. Overall response times and op-
erational transients. trol strategies can be identified quickly, erational flexibility was improved using
Dynamic simulation is the only mod- and control loops can be tuned prior to the on-line dynamic model. ❒
eling technology that can evaluate, test, actual plant startup, leading to much
and quantify the performance of these smoother and quicker plant startups and
emergency and relief systems. Evaluating substantial economic benefits.
reactor and distillation relief systems, and Modern control algorithms need timely
compressor surge control are typical ex- input of key process operating data, in-
amples of dynamic simulation in this area. cluding many critical compositions and
One of the largest uses of dynamic process performance parameters that cannot
models has been in the area of hazard easily be measured directly. Dynamic
and safety “what if” studies. With a dy- models running on line, accepting process-
namic model, a process or control engineer operating data and “shadowing” the actual
can explore such questions as “What if I operation of the plant in real time, can act
lose cooling water to my reactor?” or as “soft sensors” for those critical process
“What if this valve fails to open?” data not easily measured. The performance GLENN R.
The dynamic model provides quanti- of advanced control algorithms can be DISSINGER
tative answers to the questions on how improved significantly on such applications Dr. Glenn R. Dissinger is a director
the process will respond, how much time as polymer reactors and complex refinery of product management for Aspen
is available to respond to the situation, fractionation towers. Technology, specializing in dynamic
what can be done about it, and/or how it Many operational problems could be simulation and the oil and gas products
can be prevented. These situations are avoided if it was possible to detect the within the Aspen HYSYS product family.
time critical. The dynamic simulation disturbance long before its impact was Dissinger has 30 years of experience
model allows one to examine many more seen in measurable parameters or within in managing, marketing, and applying
scenarios than would be possible without critical equipment. For example, slugging mathematical models and computer
it, thus making it possible to quickly val- can occur in long distance pipelines that simulation technology within both op-
idate the danger of a number of scenarios transport oil and gas from offshore fields erating and process simulation software
and focus corrective actions on the most to onshore separation facilities. companies, including Mobil, Air Prod-
critical ones. In this manner, using dynamic An on-line dynamic model of the ucts, KBR and Simulation Sciences.
simulation can improve the safety and pipeline can provide an early warning. The