GE and friends show how to harness IoT

Sept. 30, 2015
Experts and end users demonstrate how they're using GE solutions like Predix to achieve peak asset performance
About the author
Jim Montague is the Executive Editor at Control, Control Design and Industrial Networking magazines. Jim has spent the last 13 years as an editor and brings a wealth of automation and controls knowledge to the position. For the past eight years, Jim worked at Reed Business Information as News Editor for Control Engineering magazine. Jim has a BA in English from Carleton College in Northfield, Minnesota, and lives in Skokie, Illinois.

The onrushing Internet of Things (IoT) and its industrial counterparts can be overwhelming topics to get your head around, but GE and its many partners explain it's still just about improving the efficiency and productivity of its customers.

Kate Johnson, chief commercial officer of GE Digital, led a team of experts and end users, who described some of the best ways to think about, understand and take advantage of IoT, and then demonstrated the best practices, IoT-based tools and GE solutions they're using to achieve new levels of asset performance and optimization. They presented "Asset Performance Management: The Power of Performance" today at Minds + Machines 2015 in San Francisco.

Dr. Bill Moreau, managing director of sports medicine for the U.S. Olympic Committee (USOC), reported that its three national clinics and 30,000 annual patient encounters rely on GE's imaging and digital health records, and help it give U.S. athletes the crucial edge they need at Olympic events.

"The difference between winning and competing—eating cereal versus being on the box—is very small, and so we're using evidence-based telemedicine to provide more timely and correct diagnostics starting with the first patient encounter," said Moreau. "Previously, many of their health records were on paper that had to be moved as freight, and it was useless for analysis. Now, we're using GE's technology to put electronic records in the cloud, and we can provide better care by predicting and managing care before it's needed, and USOC and GE are going to lead in bringing in neural models in the near future."

Lessons from the cloud

"Products are becoming services and services are becoming products, and the question is how can you make these new strategies central to your business?" NetSuite's Zach Nelson discussed the ongoing transformation of new business practices enabled by cloud technology.

To help other people and organizations consider and begin to initiate similar IoT programs, Zach Nelson, CEO of NetSuite, presented lessons his firm has learned from its 17 years of providing cloud-based computing services. NetSuite was founded in 1998, and now provides nearly 50% of all cloud-based ERP services, Nelson said.

"The world of bits and things is coming together everywhere, and so the first lesson is that every company is a cloud company," said Nelson. "We have 26,000 customers, and they all want to network, compete and leverage the IoT, and so it's important to think of yourself as a cloud company first."

Nelson's second lesson is that the cloud's power lies in data aggregation. "Aggregation is disruptive because it can deliver second and third insights based on data," he said. "In just the past several months our customers have added 25 million SKUs, run $150 billion in orders, and generated 17 million purchase orders and 115 million invoices. So, the fourth lesson is that the cloud enables entirely new capabilities, new models, new channels and new markets.

"NetSuite customers like FitBit and GoPro make physical products, but it's what users do with them that, for example, is turning GoPro into more of a new media company. Products are becoming services and services are becoming products, and the question is how can you make these new strategies central to your business? The key platform is the cloud because of its flexibility, extensibility and scalability; we're also excited about GE's Predix and what it will allow us and other users to do."

Next-gen APM

"How can industry get to today's asset performance management?" asked GE's Kate Johnson. "Many users know about machine health, but do you know everything about your equipment? Are you able to put it context? Can you do maintenance optimization and plan your workforce around that knowledge?"

Johnson added that GE's answer to these questions is Asset Performance Management (APM) powered by Predix. It can connect, capture data, monitor machine health, do advanced analytics, perform reliability management, and get the right information in the right hands at the right time. "There are a lot of challenges facing companies these days, and so they need a digital, industrial company like GE to help," she said.

For instance, Seth Bodner, general manager of cab electronics at GE Transportation, and Sameh Fahmy, executive sales leader of GE Transportation, reported on GE's Self-Aware Train program. To alleviate the derailments and disruptions that keep railroad professionals up at night and cost the industry 300,000 lost hours and about $1 billion, they showed how GE is filling in the gaps between traditional visual inspections and sensors in the search for broken rails, stressed overheating bearings and other difficulties.

"Even ultrasonic detectors may only check rails and other components once per month, and unseen problems can occur in between," said Fahmy. "Because locomotives are our flagship product, we've monitored their real-time health for a while now, but they're only one part of a train. Our aim is to make entire trains aware and self-aware. We already have our GoLink mobile data centers on trains, and now we're using them to gather more data, and check the health of railcars, rails and other infrastructure for functional effectiveness."

Bodner reported that GE has been working with Sperry on a locomotive-mounted rail integrity monitor that can check and report on the status of rails as the train goes over them, and pinpoint targets and locations for repair crews. GE is also working with Amisted Rail on real-time health monitoring for individual rail cars, and then sending that data via GoLink. Also, where GE's LocoVision video analytics tool was traditionally a passive data recorder, it too is being enlisted to help inspect equipment and even check signage on the fly.

"Of course, Predix cloud is what's going to tie all these inputs together, deliver insights to users for greater efficiencies and peak performance," said Bodner.

The Internet of Energy

Many energy generation and extraction applications are using similar IoT tools to their advantage. Jeff Ingraham, director of combustion turbine services at Duke Energy, reported that his firm presently has 50 gigawatts of power generating facilities, including 300 turbine assets, and so it's been focusing on asset performance and how the Internet can help for many years.

"We always want to identify issues, so we can respond and prevent catastrophic events," said Ingraham. "We don't want to find out about issues on short notice, and so we bring the same level of technology to maintenance that we bring to monitoring equipment. However, many failure modes don't have signatures that forewarn us much, and many events seem to come from nowhere. Unfortunately, a lot of maintenance is still static and based on calendars or financial schedules, and so we've needed a more dynamic solution like GE's APM suite."

Hugo Guerrero, engineering VP at Crestwood Midstream Partners LLC, agrees that APM's predictive analytics helps it run its Waukesha engines, compressors and other equipment more efficiently at its shale gas applications in and around West Virginia. "We had a lot of quick and dirty expansion driven by growth, and so we had a lot of disparate unconnected networks of assets," explained Guerrero. "We're finding a lot of ways we can use a platform like Predix to bring our assets together, predict failures, manage downtime, and coordinate with other systems. And, we're not just looking at engines and moving gas; we're looking at all of our connected assets. We think we're going to use Predix as a neural network that can deliver improved reliability to customers, as well as minimize downtime, reduce costs, and help make unprofitable wells profitable."


Similarly, Kay Saleeby, senior technology consultant for machinery at Exxon Mobil Production Co., added that his company is using GE solutions like Predictivity and Predix to manage asset performance and optimize operations at its new Papua New Guinea liquid natural gas (LNG) plant. This two-part, $19-billion project includes an LNG plant on the coast near Port Moresby and a remote natural gas plant at an elevation of 2,000-meters, which are connected by 700 miles of pipelines. PNGLNG began operations in 2014, and is expected to deliver abut 9 trillion cubic feet of natural gas during its 25-30-year lifetime.

"We shifted more to asset monitoring and detection when we noticed dry gas seals that were leaking, and GE was able to help us predict their life better and reduce downtime," said Saleeby. "We also had a tripped train, but we couldn't find the leak in an enclosure until GE's tools advised the site to check a C-ring manifold. The leak was found in brazed joint that wasn't under pressure, and so it might not have been found otherwise."

About the Author

Jim Montague | Executive Editor

Jim Montague is executive editor of Control.