The control system is key to optimal loop tuning

Leveraging the proportional-integral-derivative (PID) tuning tools embedded in automation software helps identify the ideal tuning values and maintain them, regardless of process changes. Example: Startup and product formula change time fell 40%; production rates of some products increased 35%.

By James Beall April 1, 2022
Courtesy: Emerson

 

Learning Objectives

  • Explore how embedded loop-tuning tools deliver easier control.
  • See how process awareness drives smoother process changes.
  • Understand embedded tuning optimization, interactive loops.

Accounting for process industry changes to ensure consistency requires careful control loop tuning, and, in more complex scenarios, can necessitate advanced process control strategies. In the process industries, change is constant in product composition, production rates and even personnel. Processes are subject to a wide variety of factors impacting quality and performance. Many plants struggle with maintaining enough personnel to perform loop tuning and advanced control.

Organizations do not need to adjust staffing and schedules, potentially delaying other critical work, to monitor and tune control loops. Many organizations want the control system, the technology central to process operations, to be a strategic tool to assist in maintaining proper proportional-integral-derivative (PID) tuning.

Using PID tuning tools integrated into modern control systems let process control personnel tune processes on demand, adapt to process changes and automate the tuning process for better performance and quality. Plants using control-system-integrated PID often see significant benefits in throughput, energy costs, quality variation, and equipment availability (Figure 1).

Embedded loop-tuning tools deliver easier control

A wide variety of tools are available in the marketplace to help personnel better tune control loops for plant processes, which can make selection difficult. The best combination of performance, cost, and ease of use often comes from control loop tuning tools embedded in the control system itself because integrated loop tuning technologies can take advantage of all the configuration data already stored in the control system. This helps ensure critical data is correct from the moment loop tuning is implemented, with this data automatically synchronized regardless of the changes personnel make to the process or the control loop.

Engineering units, tuning changes, loop tag numbers, process variable scale, and other PID options are automatically updated and reflected in embedded tuning tools when values change. This reduces the number of personnel needed to maintain loops and can reduce or eliminate the risk of data entry errors.

When using an external tool, the services required to copy the control system database, modify it, and set up replication increase cost, time and potential for errors. In contrast, embedded loop tuning tools come already installed in the control system, eliminating duplicate effort. Setting up an integrated system typically requires no more than the few clicks necessary to activate the software.

Figure 1: Properly tuned controllers lead to significant operational benefits. Engineering teams can use an embedded toolkit on an Emerson DeltaV control system to quickly identify loop dynamics and tune even the most complex loops in the plant. Courtesy: Emerson

Figure 1: Properly tuned controllers lead to significant operational benefits. Engineering teams can use an embedded toolkit on an Emerson DeltaV control system to quickly identify loop dynamics and tune even the most complex loops in the plant. Courtesy: Emerson

On-demand tuning in action in a power plant

For one power plant where nearly 50% of control loops were underperforming, operators found themselves spending far too much time attending to troublesome loops. The engineering team used the embedded toolkit on their control system to identify loop dynamics and tune even the most complex loops in the plant.

Using the process data and configuration details gathered and contextualized directly from the control system, the team created a structured approach to the identification of control performance problem sources and developed appropriate loop tuning parameters.

Implementing embedded control loop tuning significantly improved performance across the plant. Better PID tuning improved the plant’s control performance, and processes across the facility experienced gains in throughput, as well as reductions in energy cost and quality variation.

Process awareness drives smoother process changes

One of the primary factors causing control loop tuning to take up so much time is the manual effort needed to maintain process awareness and understand when changes impact production. If process dynamics change when production variables change, PID tuning parameters must as well to help ensure optimum performance.

The most advanced loop tuning tools use process learning to develop reliable process models as closed-loop control is executed. These process learning tools run in the background 24/7, identifying process models and comparing against the active model. When process dynamics change and deviate far enough from the existing model to cause a problem, adaptive tuning systems alert plant personnel to let them know further tuning is required. Operators or engineers can use the interface to update the tuning parameters based on the new process model.

Not only does adaptive tuning keep plants alert to necessary PID tuning changes without the need to commit personnel to manually monitor loop tuning, it also helps operations and process control personnel trend process performance. If the adaptive tuning software alerts operators to a need for new tuning too frequently, the team can quickly see something is changing in the process dynamics. They can then compare parameters of the process model versus current or past state variables to identify patterns and relationships, and to determine root causes of these issues.

Figure 2: A large polymer plant used the loop tuning toolkit embedded in its Emerson DeltaV distributed control system to reduce startup time and increase production rates. With another plant’s control system upgrade, adaptive tuning and control software embedded in DeltaV enabled controllers to maintain the desired response and reduce variability over the full operating range. Courtesy: Emerson

Figure 2: A large polymer plant used the loop tuning toolkit embedded in its Emerson DeltaV distributed control system to reduce startup time and increase production rates. With another plant’s control system upgrade, adaptive tuning and control software embedded in DeltaV enabled controllers to maintain the desired response and reduce variability over the full operating range. Courtesy: Emerson

Embedded tuning at a polymer plant

At a large polymer plant, poor control performance of the reactor temperature control system led to downtime when the safety system would trip the reactor (Figure 2). The problem was exacerbated by long reactor startups and frequent product formula changes, which caused long waiting periods for the temperature to stabilize.

Because different product formulas required different tuning, reducing the reactor trips was not as simple as creating a set-and-forget tuning strategy for the slurry loop. Instead, the team turned to the loop tuning toolkit embedded in the control system to measure the complex process dynamics of the reactor and its coolant systems for different product formula groups. Then, an adaptive system was used to manage the tuning for different product formulas.

The embedded tuning procedure improved process performance. The operations team reduced startup and product formula change time by 40% and increased production rates of some products by as much as 35%.

Tuning interacting loops

In some complex processes, manual tuning is not enough. In these processes, there is a high level of interaction among several PID loops. Or, in other situations, operators optimize processes by controlling against an active constraint and trying to push those parameters as close as possible to the minimum or maximum to achieve peak performance. In these and other complex control environments, teams typically turn to adaptive process control.

Adaptive control uses information from process learning, which identifies changes in the process models in relation to specific state variables. Adaptive control automatically changes the PID tuning parameters for multiple regions of a state variable to maintain the desired loop response, with no manual intervention required. Operators work in a more supervisory role, freeing them up for other tasks. The process is more stable, improving throughput and helping ensure consistent quality.

Adaptive control at a gas plant

In one large gas plant, quality specifications were not being met because key control loops overshot their setpoint or oscillated around it. Operators were forced to frequently intervene to make adjustments, or they kept the loops running in manual mode. Not only were process upsets causing alarms, trips, and opening of release valves, but cycling was also shortening the life of the plant’s control valves and other equipment.

As part of a control system upgrade, process control personnel implemented adaptive control. Using the adaptive tuning and control software embedded in their new control system, the controllers maintained the desired response and reduced variability over the full operating range.

Today, the plant’s loops stay in automatic mode and maintain performance over the entire operating range. Flow, temperature and composition are more accurately controlled, bringing improvements in product quality, yield, energy consumption, operator effectiveness and equipment lifespan.

Leveraging the power of the control system

Properly tuned control loops are critical to efficient, productive operations. In today’s environment of personnel shortages and increased performance expectations, performing manual PID tuning is no longer an optimal solution. Fortunately, the tool at the heart of operations, the plant’s control system, also contains and manages all the data necessary to maintain properly-tuned loops.

Moreover, the most advanced control systems already contain the tools needed to unlock the value in that data to drive faster and easier PID tuning as well as implement advanced control strategies to increase performance, even in the most complex environments.

By leveraging the robust software and contextualized data available in the control system, you can help improve quality, yield, and energy consumption, while freeing up the most knowledgeable personnel to focus on other essential tasks.

James Beall is a principal process control consultant at Emerson Automation Solutions. Edited by Mark T. Hoske, content manager, Control Engineering, CFE Media, mhoske@cfemedia.com.

KEYWORDS: PID, process optimization, control system tuning

LEARNING OBJECTIVES

CONSIDER THIS

Have you reconsidered strategies for control-loop optimization?


James Beall
Author Bio: James Beall is a principal process control consultant at Emerson Automation Solutions with 40 years of experience in process control and instrumentation. Prior to Emerson, Beall worked for Eastman Chemical Co. for 20 years. Areas of expertise include process instrumentation, control valve performance, control strategy analysis and design, advanced regulatory control and multivariable, model predictive control (MPC). He is a contributing author to “Process/Industrial Instruments and Controls Handbook,” Sixth Edition. He spends most of his time in plants implementing process control improvement projects in all process manufacturing industries.