- August 29, 2016
By Cory Fogg, Automation.com
Efficient predictive maintenance has been a goal of the process industry for ages. This article discusses the efforts of Flowserve, with NI, HP and PTC, to create a effective solution for the process control industry
By Cory Fogg, Content Editor, Automation.com
$20 billion lost in unplanned downtime. That number represents a massive loss for the continuous process industry in 2015. It’s also what drove an impressive predictive maintenance collaborative project by Flowserve, in collaboration with National Instruments (NI), Hewlett Packard, and PTC as demonstrated at this year’s NIWeek in Austin.
Efficient predictive maintenance has been a goal of the process industry for ages, but as NI Marketing Manager, Monitoring Solutions, Brett Burger explained, “You had to throw your hands up, because there wasn’t an economical way to do it.” That is, until the age of IIoT and Industry 4.0 enabled a solution. Flowserve’s VP of R&D Eric van Gemeren described the solution as, “The convergence of new technologies with ubiquitous data access, to allow end users to not just monitor a critical few pieces of equipment, but the important many pieces of equipment that really drive uptime in a process plant.”
NI's Brett Burger (right) discusses the applications of the new predictive maintenance solution as Flowserve's Eric van Gemeren looks on
Leveraging IoT for Real-Time Predictive Maintenance
Burger, who worked with Flowserve and the ongoing initiative for condition monitoring, sat down with Automation.com to describe the solution and how it would impact the process control industry. The predictive maintenance solution combines NI’s LabVIEW 2016 development software, as well as their recently-released CompactRIO controller, with PTC’s Thingworx analytics engine and the HPE Edgeline IoT System, all connected together. Demonstrated by Burger and van Gemeren, on a Flowserve pump, LabVIEW and CompactRIO gathered real-time measurements from the connected sensors. This data was then fed into the Thingworks Analytics engine, which converted the data into operational insights. “Through this data, over time, the software learns how that machine operates,” described Burger. All of this was accomplished using the computational power of the Hewlett Packard edge computing device. In the demonstration, Burger and van Gemeren used some impressive augmented reality to show how the software was then able to use a green, yellow and red system to show how the pumps systems were running, and, if there was a fault, show exactly where the fault is, and how long they had until the fault would shut the pump down. Further, this solution was designed to be scalable, from just the one pump used in the demonstration, to the hundreds of pumps, motors, fans and gearboxes utilized across an industrial plant.
NIWeek Predictive Maintenance Presentation w/ Flowserve
A development environment designed specifically to accelerate the productivity of engineers and scientists, LabView 2016 delivers upgraded data communication, enabling users to pass data between loops with a single wire, and increased integration with open-sourced platforms.
This platform-based, Wi-Fi enabled controller sits on the edge with the asset, connecting with over 100 different types of sensor to give the user insight into all the asset’s raw data. “CompactRIO,” Burger describes, “Will connect any sensor with any type of feature or insight to help you run your business better.”
Automates the process of analyzing the raw data, and extracting the useful insights from the data. The engine enables particular pattern and anomaly detection, as well as simulation as part of the predictive analytics effort.
The integrator of the solution, the HPE Edgeline System enables the connectivity and drives the computing power of the solution, while transferring data, securely through the cloud, to the connected devices.
A Direct Financial Impact for Process Control
Burger used a clever analogy to describe just how this technology would impact the process control industry, “This technology is kind of like the tire pressure monitoring system (TPMS),” likened Burger, “10-15 years ago cars didn’t have this, and you only knew about a tire problem after it blew. Today, almost every car is going to have a TPMS.”
There is lots of room for this technology to make an impact in the process control industry. Burger said that while a plant would rarely give exact details on downtime, some recent studies estimated uptime around 75-80%. With the direct impact of process uptime on company revenue, “Even adding a few minutes of uptime could save companies millions,” exuded Burger, “That’s why this technology is so exciting.”
Flowserve's Eric van Gemeren (left) demonstrates augmented reality capabilities of their predictive maintenance solution as NI's Brett Burger explains how it works
Constant Connectivity = Streamlined Efficiency
The bottom line with this technology is that it could make technicians’ lives a lot easier. Though the demonstration used a tablet as the augmented reality device, Burger suggests that one day this technology will be meshed into technicians’ hard hats giving them an augmented reality headset, in a similar manner to Google Glass. As Burger describes, “This technology will guide technicians through the steps of repair, reducing error, making them more efficient, and driving more efficient production across the plant.” Further, he explained that, with the HPE Edgeline moving the data through the cloud, management, data acquisition teams, and technicians can log-in and exercise control either at the plant, or outside of it.
While it is uncertain just how much of the $20 billion downtime deficit this predictive maintenance solution will help the process industry recoup, it is clear that Flowserve, NI, Hewlett Packard and PTC are among those driving the IIoT technology to improve plant uptime.
Related ArticlesLearn More
Did you enjoy this great article?
Check out our free e-newsletters to read more great articles..Subscribe