Industrial Internet of Things is out there, but ‘work needs to be done.’
Industrial Internet of Things (IIot) is nebulous term bandied about in just about any conversation when it comes to automation, when in reality, the level of communication already conducted on a daily basis for the past decade or so has been a precursor to the advanced technology on the verge of becoming the industry norm.
With sensors becoming ubiquitous and data rushing through the enterprise like a raging creek after a torrential spring rainstorm, properly using Big Data and putting everything in context becomes paramount for users.
“Today, we need to hire a slew of mathematicians to help us out,” said Chet Mroz, chief executive at Yokogawa North America. “There are about three times more data points coming in.” There will be even more data coming online as Mroz said there will be direct integration of the distributed control system to subsea.
While IIoT is not a complete package yet, there are companies employing pieces to garner additional knowledge and productivity. “There is a lot of IIoT out there, but there is a lot of work that needs to be done,” said Brian Phillippi, one of the managers on the C Series IO products at Austin, Texas-based National Instruments. “A lot of companies already are taking in a lot of data.”
Left: A screwdriver is the driving force behind the former way of detecting a leak in a pump. Right: A screen shot showing NOV’s leak detection system. Images from NOV.
Holding Back IIoT
There are some big issues, however, preventing full adoption of the IIoT.
One is finding the right people with the correct skill set. “Some of the hottest jobs right now are the data scientists that can make sense of all the data,” Phillippi said. “Another big one in terms of adoption of the IIoT is going to be security. Every week now, we are hearing about some kind of hack going on and I think the industry as a whole has to get better about embedded security. The third thing causing a lot of problems of IIoT is the ability of everything being able to communicate on a single standard. On the commercial IoT they have a really good way to communicate with the regular Internet. That won’t cut it on the industrial side of things because of the determinism that is needed for the industrial environment.”
“IIoT is just about taking a thing, automating that thing and adding value to the Cloud and the Internet,” said Jamie Smith, National Instruments’ director of embedded systems product management, quoting a conference keynote speaker discussing the IIoT. “If you boil it down to those three simple steps, I think a lot of companies can start. They can look for things they want to automate and then they can add value through the cloud and the Internet.”
A case in point is condition monitoring and big analog data analytics that allows for a predictive maintenance plan.
Monitoring Mud Pumps
National Oilwell Varco (NOV) is a perfect example. The oil services giant was dealing with trying to solve issues with mud pumps that circulate drilling fluids. Mud pumps stabilize pressure and support the well during the drilling process and drilling fluids provide friction reduction and a means to remove cuttings. NOV created a leak detection system for hex pumps. The hex mud pump relies on six pistons, six suction valves, and six discharge valves. A rotating asymmetric cam drives the six pistons, according to Pål Jacob Nessjøen, formerly with NOV and now with Kelda Drilling Controls.
“Previously, they were listening to these pumps and trying to understand if there was a leak,” Phillippi said. “Sometimes they would put a screwdriver on the pump and listen to the vibration on the screwdriver. When you are talking about these big pumps that cost thousands and thousands of dollars, that wasn’t a good enough solution for them. NOV implemented a condition-based monitoring system so they can, not only know when they have a problem, but be able to pinpoint where exactly in the pump the seal is actually leaking and having a problem.”
Learning about a valve issue often occurs when the leaks are so severe they induce large discharge pressure fluctuations and create washout damages. “When a severe leak is detected, we localize it manually by listening to the fluid modules while the pump is running, but it is difficult to uniquely localize the leak and distinguish between a suction valve leak and a discharge valve leak,” Nessjøen said.
When it comes to valve leaks, they often develop quickly so manual detection gives very little time to prepare for changing out a problematic valve. In addition, if the source of the leak is not clear, finding out where it came from ends up being time consuming.
That is where a remote system to detect and determine the location of the leak comes into play.
Using a now-patented system, Nessjøen and NOV’s Age Kyllingstad found during a vibration monitoring project for hex pumps, it was possible to detect leaks using accelerometers. NOV engineers recorded vibrations at different locations, on the pump and on the discharge line, along with suction pressure, discharge pressure, and pump speeds for different pump conditions.
When at one point they noticed a vibration signature changed during a 15-minute time frame, they knew a leak was developing.
That was the beginning of a true predictive maintenance situation. They were then able to perform more tests and they determined discharge valve 2 had a severe leak.
Based on that experience and some further testing, they ultimately included this condition-based maintenance system as a standard feature on all hex pumps. They developed the system as a stand-alone module to add to the existing hex pump control system. It consists of the following components: Accelerometers (one per valve block), a proximity sensor picking up pump speed and phase, a discharge pressure sensor, an embedded monitoring system, signal processing software and alarm logics.
Now they are able to remotely verify leaks detected automatically by signal processing. The operator can do this by: Viewing and interpreting the vibration signals directly from graphs; selectively listening to the recorded acceleration signals as audio signals to hear the leak sound; checking to see if the mean discharge pressure is stable or dropping, and seeing if the lowest pressure harmonics are growing.
Based on the field experience of the new leak detection system, NOV found the leak detection method offers advantages over current practices, including:
- High sensitivity for early leak detection and localization
- Remote, continuous, and computer-based pump monitoring
- Increased safety through less human exposure to hazardous environments
- Multiple leak detection and localization (in hex pumps)
- Reduced maintenance time and cost because leaky valve(s) are localized before the valve exchange jobs start
- Easy to retrofit existing pumps because accelerometers can be attached by glue, magnets, or tape
IIoT is continuing to grow offshore as more operators are understanding and finding opportunities to incorporate more knowledge and productivity into the system.
“There is more adoption now and it is on the condition monitoring side of things to begin with,” Phillippi said, “but we also see the value of that on the control side as well.”
Gregory Hale is the Editor and Founder of Industrial Safety