A new generation of jet engines, locomotives, MRIs and other big machines loaded with sensors generating gigabytes of data and linked in networks will become more efficient to operate, easier to deploy, and more accessible to innovators, according a new report on the industrial internet published by O’Reilly Media and sponsored by GE. “The barriers between software and the physical world are falling,” the report says. “It’s becoming easier to connect big machines to networks, to harvest data from them, and control them remotely.”
The report points out that “the same changes in software and networks that brought about decades of Silicon Valley innovation are now reordering the machines around us.” It says that new, web-like software interfaces can manage the underlying complexity of machine data, “making it possible for innovators without specialized training to contribute improvements to the way the physical world works.”
Two years ago, GE tapped Silicon Valley talent and opened a $1 billion research center in San Ramon in the Bay Area. Engineers at the center are writing code to harvest data from GE machines, build machine networks, and turn big iron like turbines and locomotives into brilliant iron. “Any machine that registers state data can become a valuable sensor when it’s connected to a network, regardless of whether it’s built for the express purpose of logging data,” the reports says. “Once a system of machines is brought together on a network, it’s easy to add new types of intelligence to the system, and to encompass more machines as the scope of optimization expands.”
One such network profiled in the report is the grid. Dennis Sumner, senior electrical engineer at Fort Collins Utilities in Colorado told O’Reilly that a $36 million investment in advanced electricity meters could pay off in 11 years just from operational savings. But since the meters can read electricity usage every 15 minutes, a 2,880-fold increase compared to human meter-readers, they also provide the utility with tremendous resolution and tools to detect outages immediately. “Previously, we didn’t know what was going on at the customer level,” Sumner said. “Imagine trying to operate a highway system if all you have are monthly traffic readings for a few spots on the road.”
Utilities blending power from traditional generation with renewables need to know how much wind electricity they can count on and when. Many wind turbines are already packed with sensors and can talk to each other like a flock of birds. “We have advanced forecasting algorithms that give us power output predictions based on the data they are receiving,” says Vic Abate, vice president of GE’s renewable energy business. “With this technology you are able to say, I am going to give you 70 megawatts over the next 15 minutes and with 99 percent accuracy.” Customers like Boston’s First Wind have already signed up for such technology.
Norfolk Southern is using GE’s Trip Optimizer software as “a kind of autopilot for locomotives,” and Movement Planner software as an “advisor and controller.” Deborah Butler, Norfolk Southern’s chief information officer, told O’Reilly that her railroad has seen a 6.3 percent reduction in fuel usage and 10 to 20 percent increases in velocity from the software.
This is just the start. Besides crunching data from locomotives and signals, Norfolk Southern has used helicopters to map its rail network. “We know where every tree is growing beside the track,” she said. “We aren’t even beginning to use that data in the way we could.”
The full report is available online.
The Power of One Percent: Networks of “brilliant iron,” like this aeroderivative gas turbine, could save power companies $66 billion over the next 15 years by cutting fuel use by just 1 percent. The GEnx jet engine can collect and analyze 5,000 data points every second, detect problems, and optimize performance. Speeding up trains by just 1 mph could save a railroad $200 million in annual capital and operating expenses.