D. Beebe, Emerson, Baton Rouge, Louisiana; and A. KALAFATIS, Aspen Technology, Houston, Texas
There is a new adage in process manufacturing circles, and it resonates particularly deeply in the hydrocarbon industry: If you do not like market factors today, just wait a month. In the wake of the pandemic, it seems as though there is a new supply chain disruption, market need, cost increase or labor shortage around every corner.
As frustrating as these market swings are, savvy organizations also see their upside. The companies maintaining the flexibility necessary to navigate new trends gain an opportunity to secure competitive advantage by increasing market share and improving profitability. One of the key technologies helping organizations unlock and ensure flexibility is advanced process control (APC).
In the past, many considered APC only feasible for large organizations with a deep bench of expert personnel. Those well-resourced organizations could more easily commit countless hours to integrating, configuring and maintaining the software alongside the plant’s control system. APC seemed out of reach for plants that could not find, afford and retain one or more staff members with an advanced degree in chemical engineering.
Today, however, the landscape has changed. The best modern APC solutions have been designed to democratize advanced control. Industry experts have worked much of the complex front-end engineering design (FEED) into intuitive, highly integrated software, maximizing the benefits of APC while minimizing the complexity, normal operation disruptions and operational challenges associated with adding advanced control to process manufacturing. Teams should focus on implementing advanced control with a holistic mindset by choosing integrated solutions from end-to-end. This enables a smoother path to meet operational efficiency and sustainability goals to more quickly achieve corporate objectives (FIG. 1).
Modern APC. APC uses software logic to increase the speed and scope of modern process control, helping organizations unlock the full value of their control architecture. Leveraging advanced control systems—such as statistical modeling, analytics, model predictive and multiple loop control—helps teams operate their systems closer to constraints to drive tighter control and more predictable and stable operation. These technologies help lower the variability of process parameters and product quality, ultimately reducing human error while improving quality, increasing throughput and reducing energy use and emissions by working in tandem with effective operator oversight.
Complexity creates barriers. Traditional APC design, implementation and configuration took many steps. First, teams needed to perform pre-testing, step testing and model identification before building and commissioning the controllers. Typically, this process took 3 mos–4 mos, and the required step testing was intrusive, interrupting production and, in many cases, creating off-specification (off-spec) products.
However, even if a plant had the time, resources and open schedule to perform the design phases, they often lacked the required skilled personnel. Teams needed an expert with years of domain expertise to understand the underlying calculations necessary to successfully create the models and controller configuration. Even if they could find a qualified model predictive control expert to fully understand control configuration and model identification, that person would be needed for ongoing software maintenance if the organization wanted long-term value from its solution. Keeping such a person on staff across the lifecycle of the APC software was a costly proposition.
Faster time to value. Today’s best-in-class APC solutions have upended the complex system integration and configuration paradigm. Modern control systems are built from the ground up to seamlessly connect to the most powerful and intuitive solutions, eliminating the weeks of pre-work necessary to get the systems communicating effectively.
Just as importantly, much of the domain expertise necessary to build and configure advanced control solutions has been built into today’s advanced control software. The most effective APC software is designed to be configured and operated by process engineers, opening the technology to more organizations and sites that did not previously have access to skilled experts.
In today's state-of-the-art software, embedded industrial artificial intelligence (AI) enables engineers to mine existing historical process data, select variables and build a seed model. The seed model generation step that used to take weeks can now be completed in a day or two.
Once a seed model has been developed, it is refined through calibration software. The calibration software performs step testing in the background to tune the model to current operating conditions. However, unlike manual calibration—which took months to close the loop, resulting in reduced production and off-spec product—modern calibration software runs in the background in much smaller increments while the controller stays online. Not only does the plant continue to run just as it normally would, but the entire process saves time, with many organizations seeing upwards of 75% faster time to value.
Today’s APC software is also more intuitive, with built-in tools to democratize configuration and operation. In the past, operators and engineers often had little-to-no insight into what the APC was doing at any given stage of operation. Today, the best solutions are built with virtual advisors to tell an engineer or operator why the advanced control performs a specific action and how to achieve a specific control objective. The end user is empowered with a real-time understanding of why the system is doing what it is doing and how to get the APC to do precisely what they desire.
Modern advanced control solutions are built with a much wider range of users in mind. With AI capability, teams can widen the operating region while using machine-learning (ML) algorithms to help build and refine seed models (guided by ML).
Increasingly sustainable operation. One of the key areas where APC delivers value is improved sustainability. Nearly every organization is under pressure to meet new sustainability goals. Some may simply need to improve energy use, while others are overhauling operations to meet ambitious new net-zero goals.
Regardless of the reasons for change, improving sustainability with traditional manual operations is challenging. Operators typically dial a lot of conservatism into how they run facilities, operating significantly away from constraints in each scope of control. They do this for good reasons.
For example, consider a situation where an operator was running right at a limit, and excess oxygen could mean the process does not have complete combustion. Running regularly in such a state would require operators to check the process status minute by minute—a stressful and potentially dangerous strategy.
Moreover, lean staffing is common these days, and an operator might have five or six scopes of control. An experienced operator can likely run one constraint closer to optimal, but each of their five or six controllers might have multiple settings, each of which must be optimized to drive peak performance in tandem with sustainability. No human operator can handle 25 or 30 variables minute to minute.
However, multivariate APC software can handle all those variables and more. In addition, the software can manage the constant calculations necessary to compare operations against the operating environment to ensure peak performance (FIG. 2).
APC predicts a system state ahead of time to reduce variability. The software constantly checks the process and performs optimization calculations, never getting tired or distracted. With this capability, the system can run every process element very close to constraints to maximize efficiency.
When the process is optimized, teams see less reprocessing, leading to less waste, lower energy consumption and lower overall emissions while simultaneously maximizing production. In addition, operators are freed from low-value monitoring tasks to focus on improvement and optimization across the plant to further improve performance and sustainability (FIG. 3).
More easily achieved corporate objectives. Advanced control software installed and properly configured at the unit level can do wonders for plant productivity, but that value is only the beginning. When an organization builds a strong advanced control foundation at each plant, unit by unit, implementing APC-ready control systems across each unit and equipping them with modern, user-friendly advanced control software can enable coordination across multiple process units.
One of the benefits of a holistic, real-time, multi-unit dynamic optimization solution is that teams no longer need to worry that they are creating excess quality in one area but giving it away in another.
For example, perhaps a facility is over-cracking middle distillates. The team could solve the problem on the front end at the crude column, putting those molecules in one bucket and not running it through the process. Ultimately, they can convert them late in the process, but in doing so, the team has used excess energy to process those molecules into one grade and then, at the last minute, blend them in or crack them into another grade.
With dynamic optimization, the operations team coordinates multiple process units and their respective APC controllers to achieve peak performance end-to-end across the entire production process. This type of closed-loop, real-time optimization of multiple units creates a communication chain between each process unit, running the entire process as a cohesive, dynamic whole. A change upstream in the process will be communicated downstream so other units can compensate, ensuring they are ready for the results of the change in operation (FIG. 4).
In fact, the most powerful APC solutions offer smart tuning capabilities to capitalize on that inter-unit cohesiveness, making it fast and easy to take advantage of market changes. As costs update and profitability changes, operators can use smart tuning solutions to adjust those variables in the software. Upon receiving the change—or the direction to focus on a different core strategy, such as energy use instead of production—the software immediately updates operations to reflect the new goals. Operations teams can be more proactive, capturing value for the organization.
However, operational agility is more than simply pivoting to meet a new need in the marketplace, as staying agile means meeting other market challenges, such as staffing issues. Again, this is an area where APC shines. Because the simplicity of modern advanced control democratizes the technology, even when a group of experts leaves the facility, new, less-experienced personnel can still maintain the system. Those new engineers can much more easily sustain the system’s performance over time, ensuring that the organization’s ability to maximize the value of its operations is not lost.
Adapting to secure competitive advantage. For plant personnel that have tried APC in the past but were unable to achieve success—and for those that have never given advanced control a try due to cost, time or staff limitations—this is the perfect time to reevaluate. New technology has changed the game for advanced control, making it easy for a plant to implement and maintain a holistic infrastructure.
Moreover, by putting an APC-ready control system in place and pairing it with best-in-class technologies built to democratize its design and use, teams can unlock the flexibility necessary to drive more sustainable operations and secure competitive advantage. Those teams will no longer need to wait for market factors to change; they will simply identify them, and operations will instantly adapt. HP
ALEX KALAFATIS is the Vice President of Product Management at AspenTech. He is responsible for the manufacturing and supply chain management product portfolio. Dr. Kalafatis has spent most of his 25-yr career in the control and optimization of processing industries across the refining, upstream, chemicals and polymers industries. He strongly believes in digital transformation's value in industrial facilities as a critical factor in achieving new levels of operational excellence, including emissions reduction, energy efficiency and sustainability with profitability—advanced process control, optimization and embedded AI are key enablers for this transformation. Dr. Kalafatis earned an MS degree and PhD in chemical engineering from the University of Toronto.
DUSTIN BEEBE is the Vice President of Performance Software for Emerson. He is responsible for aligning the global control performance, operator performance and simulation businesses and the strategy synergy between Emerson and AspenTech. Before joining Emerson, Beebe served as the President of ProSys until Emerson acquired it in 2018. He has been in the industrial automation business since 1996. Beebe earned a BS degree in chemical engineering from the University of Arkansas in Fayetteville, Arkansas.