Analysis: When will wireless be ready for control functions?
The conventional wisdom is that wireless instrumentation and networking in process environments will be limited to secondary systems until users feel it has proven itself, assuming it moves out of those roles at all. Today, wireless technology is fine for level measurements on a tank farm, vibration readings on a compressor, and other low criticality monitoring operations. But what will it take for companies to feel confident enough to recommend it for actual control functions? I asked companies that deal with control systems what they might want to see to answer that question positively.
“Before users will become fully comfortable with the idea of using wireless technology for closed-loop process control, their comfort level will have to gradually increase in stages,” says Hesh Kagan, director of technology marketing at Invensys Process Systems . “The first stage is for users in industrial process plants to become confident that they can effectively manage wireless networks to provide industrial-grade availability, robustness and security for open-loop process monitoring applications. This is often already the case for condition monitoring applications and open-loop process control.
“Once the management issue is behind them, the next step is for users to be confident that the wireless standards, such as ISA SP-100, have evolved enough so that device interoperability and co-existence are no longer issues. This will set the stage for users to truly feel comfortable with the idea of relying on wireless technology to do closed-loop process control. This can take the form of wireless fieldbus (to valves and actuators), or wireless communication at the control bus level between loosely coupled controllers (with limited peer to peer), or both.”
Bob Karschnia, vice president of technology for the Rosemount division of Emerson Process Management sees the issues of control depending on three basic components: high quality measurements, reliable communications, and robust control systems. “If the physical measurement is poor, everything else in the sequence of control is compromised,” he says.high percentage of reliability. Without a system that is designed to handle unplanned disruptions, gracefully, there can be no consistent communications, which most control algorithms require to function properly. Finally, a robust control system needs to be in place that can handle the data andcommunicate it to the final control element.
“Process control has performed all of these functions over wire for many years.
Honeywell Process Solutions has designed its OneWireless system deliberately with control capabilities in mind. In that respect the company is probably out ahead of most users. "It's not a question of if, but rather a question of when,” says Dave Kaufman, director of business development. “Remember when no one would run a plant with a PC? General wireless control will come with time as users get more comfortable with the security, availability, and reliability designed into these solutions. Given this, it's important to install a wireless network already designed for the possibility of control."
Gene Chen, Honeywell’s product marketing manager for wireless adds, “The control application also has to make sense. End-users will not rip out perfectly good wires just to use wireless. They will use wireless control where is it too difficult to use control wiring, such as in harsh environments and remote areas.”
The HART Foundation has been developing its wireless protocol, and in discussions of that, its engineers have been very cautious about the idea of actual control functions. “To me it all boils down to risk,” says Ron Helson, executive director. “In a traditional chemical plant environment where the risk of failure of a particular control function could result in catastrophic consequences to people or even communities, I don't think we'll be doing wireless closed loop control of those functions any time soon. Perhaps never. Safe and effective plant operation is number one always. Control functions in those applications are done over wire today and I believe they will be done over wire for the foreseeable future.”
Wally Pratt, chief engineer for the HART Foundation doesn’t expect a defining event. “Process control with wireless will be a quiet revolution,” he suggests. “The hype and market posturing will fade as significant sales of wireless field devices begin. Market leaders care more about selling boxes than whether wireless is used for control or not. User education and reliable, continuous, consistent operation will, over time, give the industry confidence. Experimentation will then begin with wireless-based control.
What will you have to see to be convinced? Send your ideas to PWelander@cfemedia.com .
Join a webcast on Wireless in Process Manufacturing , September 5 at 11:00 am EDT. If you miss it, you can listen to the archived version within a few days.
—Peter Welander, process industries editor, PWelander@cfemedia.com , Control Engineering Weekly News