Start State-Based Control with Baby Steps

As you introduce batch control philosophies to your continuous processes, you might want to start small—with a small part of the operation that could use it most—to make sure you have the buy-in you need for further expansion.

The amount of moves operators had to make during state transitions dropped significantly after automating the process.
The amount of moves operators had to make during state transitions dropped significantly after automating the process.

There are all kinds of reasons that operations reliant on continuous processes might want to borrow a few pages from batch processing to improve their operations. In researching our September feature on state-based control, I found that a key motivator is safety.

“Safety is a huge driver—being able to reduce risk and improve safety and have a reliable system in place without adding cost,” says David Funderburg, technology manager for ABB. So as state-based control begins to grow in certain pockets, areas like high-risk compressors are often a target. “They follow state-based control-type concepts to get those units running,” he says. “But it may not be a philosophy across the entire plant.”

Another place an operation might get started is in alarm management. For example, Funderburg describes, a unit could have four states, and it could do alarming conditionally based on those states. “You don’t want an alarm to go off if the unit is not even running,” he explains. “When the unit is running, the control system is enabling those alarms, and the operator’s only getting the alarms that make sense.”

Funderburg points to the alarm example as a baby step an operation could take just to get started with state-based control. “They will be able to simplify the alarm strategy just by enabling and changing alarm limits,” he explains. “That could be their first step in state-based control. Then they could dabble in certain areas to do more procedural control.”

Wherever you start, it’s often important to start small because it can affect so many different parts of the plant, and there could be barriers to change in any number of those—engineering, maintenance, operations, even business.

“Part of the key is doing a small sample or slice test and seeing touch points within the organization, whether maintenance or safety,” Funderburg says. “It’s important to get buy-in early on. It might be easy for you to get general agreement, but until you see something working on a small scale, you’ll see skepticism from engineering groups. So getting a small solution prototyped or slice tested is a good exercise, especially if it’s fundamentally different than current control philosophies.”

The main article on this topic describes a Chevron initiative that started small at just one base oil hydroprocessing unit. Consistency in transitions from one process state to another improved significantly, along with transition times and product yields. (See the chart above to see how significantly operator load dropped using an automation program.)

Since starting small at just one unit, Chevron has since expanded the initiative to other large installations (crude unit, sulfur recovery unit, complete startup and shutdown of the lube hydroprocessing unit) and small installations (molecular sieve dryer, vessel dump and charge).


Companies in this article
More in Control