Plane Brain: GE’s aircraft health management system can collect 10,000 parameters at any given time.

Thomson may be the closest thing to an airplane doc. He is the program manager for GE Aviation’s Integrated Vehicle Health Management system (IVHM), a sophisticated blend of hardware, software and data analytics that allows airlines and aircraft operators to run a healthy fleet and get the most out of their planes. “This technology fundamentally changes how aircraft fleets are managed,” Thomson says. “Not only we can see that something has occurred, we can see the data behind it and understand why.”

The roots of the system, which belongs to GE’s Industrial Internet offerings, go back more than 20 years. At the time, engineers started developing safety systems for helicopters servicing offshore platforms in the North Sea. “There was a big boom in offshore oil exploration and oil companies wanted to get people out to the rigs safely,” Thomson says.

Engineers then focused on planes and added sensors, data links, and other airborne hardware to monitor systems like the jet engines, landing gear and hydraulics, and beam the information to the ground. “The team has evolved it to the point where we collect 10,000 parameters at any given time,” Thomson says. “Like the brain, we have algorithms that decide what data should come down to the ground, when and over what link.”

If a flight proceeds according to plan, the plane beams down the vitals like, say, fuel usage, air pressure, and speed. The crew downloads more data after landing and sends it over Wi-Fi or cell towers to the cloud for analysis. GE has built a web service that allows the customers to mine and model the data.

But if there is a problem, the system springs to action. “If you have some kind of anomaly, all of the data sets can come down from the airplane while it’s still airborne,” Thomson says. “If we see that we had a trigger event, the team can back up 30 seconds and get the whole picture of what happened. The first thing you do is that you ask for more data and look at it.”

If the event requires an immediate fix, the system can help dispatch the right parts to the nearest airport and find a replacement plane to keep schedules in place.

The result: Fewer delays, fewer missed connections and less money spent on maintenance. The industry has noticed. The system recently received an award for “game-changing technology implementation” and the 2013 Manufacturing Leadership Summit.

Just what the doctor ordered.

Hancu, who is a physicist by training and MRI researcher at GE Global Research (GRC), couldn’t let her experience go. She went into her lab and designed an imaging solution that could one day save women like her from a similar ordeal. “I want the doctor to be able to look at all the images from an examination, and know for sure that everything is all right, or whether there is cancer,” she says. “Nobody needs or wants uncertainty at the end of a test.”

Quality Control: “I measure success by making an impact on somebody’s life,” says Ileana Hancu.

Hancu’s research could be a boon for women in groups for whom mammograms can often be inconclusive, such as those with dense breasts. Her goal is to help eliminate the uncertainties when this group is screened. “You want to find out the answer about the positive or negative at that moment in time,” she says.

Hancu grew up in Bucharest, Romania, in a scientific family. Her father was a computer programmer, her mother was chemist, and Hancu enrolled at the university to study optics. “There probably wasn’t a single MRI machine in the whole country at that point in time,” she laughs. After graduation in 1996, she moved to study physics the University of Pittsburgh and discovered magnetic resonance. A few years later, she had a doctorate in the esoteric field. “That’s why I ended up at the GRC,” she says. “There are few places in the world that could use that kind of expertise.”

Hancu was in the middle of MRI research when her cancer scare struck. MRI had been used in the clinic for breast cancer diagnosis, but standard MRI images produced too many false positives, or lesions, which turned out to be benign at biopsy.

A new MRI-based method meant to clarify the picture, called diffusion-weighted imaging, has “bounced around the literature for the past few years,” Hancu says. “But the diffusion-weighted images were kind of lousy,” she says. They were like a screenshot from a TV with a bad antenna, fuzzy and distorted. A doctor could not truly read them and use them to decide the fate of a patient.

Hancu has started experimenting with methods that would increase the MRI signal and the spatial resolution of the images so that she could see even smaller tumors. One of her approaches involves building the MRI equivalent of a high-resolution camera.

The research work has already earned Hancu and her collaborator, Prof. Robert Lenkinski from the University of Texas Southwestern Medical Center, a five-year, $3.2 million grant from the National Institutes of Health. They plan to test their technology on patients in a clinical trial to help further evaluate its use in breast cancer diagnosis.

Hancu says that learning is not an academic exercise for her. “If you end up developing an imaging technique that will help save people from having biopsies, if you can develop a technique that can tell you with certitude whether you have cancer or not, I think that’s the final measure of success,” Hancu says.

“I measure success by making an impact on somebody’s life.”

“The worst thing is to have pain somewhere in your body and have no answer for it,” Potter says. In the 1990s, Potter started using MRI to get a better look at ailing joints only to face a new problem: metal joints, plates, and screws produced magnetic artifacts, which obscured a critical part of the image where the implant meets the tissue. She experimented with magnetic field settings, pushing the MRI machine to its limit. “Every single day I would look at images side by side and scratch my head trying to figure out why Mrs. Jones had knee pain,” Potter says. “But the machine can only go so far before it cries uncle and says ‘Sorry, Hollis, I can’t do any more for you.’”

Dr. Hollis Potter with GE’s Kevin Koch: “I didn’t want an ivory tower physicist being locked up in Waukesha,” says Dr. Potter (right).

As the population ages, the problem could crop up more and more. Recent studies estimated that there are more than 1 million knee and hip replacements surgeries performed in the U.S. and 250,000 in Europe.

In 2008, Potter decided that she needed a physicist to make progress. She gathered her 10 years of patient data and started knocking on industry doors. “I went to MRI manufacturers, went through my data, and showed [them] what we had found,” Potter says. “GE was willing to step up to the plate, roll up their sleeves, and try to solve this problem through imaging.”

Potter’s luck was about to turn. GE Healthcare in Waukesha, Wisconsin, had just hired Kevin Koch, a young physicist with a doctorate from Yale. “It was methods that I developed during my Ph.D. that allowed me to model the magnetic field distortions that the metals were introducing,” Koch says.

Potter was skeptical, at first. “I didn’t want an ivory tower physicist being locked up in Waukesha,” she said. “I asked him to come out and he spent a lot of time going through my data. I had him meet with orthopedic surgeons and understand how frustrated they are. I had him meet with the patients.”

Koch sympathized with the doctor. “She was very patient in the beginning,” he says. “I am sure that she wanted results and I was showing her physics explaining where the image artifacts were coming from.”

Before long, the team got their first break and filed their first patent. Koch wrote software that could predict distortions in the magnetic field and used it to correct the distorted images. But the approached turned out to be a blind alley. “It was not robust enough,” Koch says. “It was a little bit deflating.”

But then, about nine months into their research, Koch hit on another solution. He found a new way to capture the magnetic resonance signal. Rather than building the image from two-dimensional slices, kind of like reassembling a salami, Koch started adding up an arbitrary collection of 3-D blocks, or bins. The method, a set of software instructions that can work on any MRI machine, allowed the team to take the same image at multiple frequencies and fix the distortions. “It really opened up our eyes,” Potter says.

Potter says that the system, which GE calls MAVRIC SL, allows her to see damaged tissue around “MR conditional implants,” but also nerve impingement, the integrity of joints and bones fortified with stainless plates and screws, and other medical issues. “People with a recalled implant can come in and find out if their implant is at risk,” Potter says. “Whether they are symptomatic or not, they are worried about what’s happening inside their body.”

Potter says that her hospital has already imaged 3,000 patients with MAVRIC SL. She said that her work with Koch was “a perfect example of a collaboration between an academic site and an industry.”

“It gives people an answer for their pain,” Potter said. “From an emotional standpoint of dealing with illness, that’s huge.”

Now You See Me: This 65-year old patient developed severe pain lasting several months after a total hip replacement three years earlier. Unlike X-ray and conventional MRI images, MAVRIC SL showed clear evidence of an abnormal response indicative of an adverse tissue reaction. The patient was indicated for a revision surgery. Credit: Courtesy of Hospital for Special Surgery *

*These images were generated using the MAVRIC SL software feature and are representative of the quality of images that users should expect to generate. However, GE Healthcare is not always able to confirm whether the images are of MR Conditional implants or whether scanning was in accordance with the implant’s instructions for use. MAVRIC SL should only be used with MR Conditional implants and within the MR conditions specified for those implants.

But airline passengers can now relax and stockpile adrenaline for later adventures. Queenstown airport authorities have rolled out a new navigation system called Required Navigation Performance (RNP) that relies on satellites and software to bring planes safely to the gate.

Unlike traditional navigation systems that use ground-based beacons and other aids, upgraded Queenstown-bound aircraft carry a satellite receiver system that precisely tracks their position. The system allows pilots to safely and efficiently follow a predetermined route through local peaks, valleys and mountain ranges, and make a smooth landing.

Picturesque: The snow-capped Remarkable Mountains frame Queenstown’s airport. Source: NewZealandView.com

The RNP technology may also provide benefits outside mountainous airports. Many of the world’s busiest airports are operating near capacity. By switching to RNP, they could improve operations, productivity and efficiency.

The results from New Zealand are breathtaking. Since launch in November 2012, the system reduced the cumulative delay of Queenstown bound flights from 2,400 minutes per month to just 200 minutes per month. This can translate to significant fuel savings and emissions cuts.

GE estimates that the technology will help aircraft taking off and landing at Queenstown Airport to reduce carbon dioxide emissions by as much as 4.5 million pounds each year. That’s like taking 450 cars off the road for a year.

Eco-friendly bungee jumping, anyone?

Click to enlarge.

 

 

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

 

• 
• 
• 
• 
• 
• 

 

Jolesz thought that magnetic resonance imaging (MRI), which can see inside the body and also detect heat, could help. With the right machine he would be able to visualize temperature changes during the surgery and monitor the tumor treatment with heat. But he ran into another problem: such a machine did not exist.

Trifon Laskaris holds 200 patents. He helped revolutionize medical imaging.

A GE executive introduced Jolesz to GE engineer and medical imaging pioneer Trifon Laskaris. “Trifon designed a magnetic resonance machine (MRI) that was open vertically,with two magnetic rings like a double donut,” Jolesz says “We could image the patient and operate at the same time. Not only laser procedures could be done, but all types of open surgeries.”

Jolesz says that more than two decades later, Laskaris’ design “is still the best configuration” for magnetic resonance imaging during surgery. Jolesz and other doctors at Boston’s Brigham and Women’s Hospital have used it for more than 3,500 surgeries, including 1,400 craniotomies, brain biopsies and other neurosurgery procedures. Today, intraoperative MRI is widely used in neurosurgery and in other procedures.

Laskaris received a dozen patents for his work on the Brigham machine. He now holds 200 U.S. patents, a feat matched only by a handful of GE inventors. “Trifon’s work speaks for itself,” says Mark Little, head of GE Global Research and the company’s chief technology officer. “Without his decades of dedicated research into superconducting magnets, MRI technology would not be where it is today, a mainstay of hospitals around the world.”

Laskaris says that he liked playing with gadgets since he was a small boy growing up in Athens, Greece. “My father was a high school teacher and my mother was a seamstress,” he says. “One day her sewing machine broke down. I was just six years old, but I connected the pulleys, installed the little motor and put in the switches.”

Laskaris studied engineering at the National University of Athens. In the 1966, he answered a call from GE and came to the U.S. “At the time there was a big U.S. space program and many American engineers were going to NASA,” Laskaris says. “That drained a lot of talent from the industry.”

At GE, Laskaris started developing software simulating cooling flows inside massive power generators for nuclear power plants. But he quickly moved to GE Global Research (GRC) and started working on magnets and superconductivity, a physical phenomenon that drops electrical resistance to zero in extremely cold metals. “When you power up a supercooled magnet, it can produce the same magnetic field for a thousand years with no more power required. You can do so many cool things with it,” he laughs.

Things like building an MRI machine. In 1983, a team of GRC engineers developed the world’s first full-body MRI, and Laskaris helped design the machine’s 1.5 tesla magnet. “We started by imaging grapefruits,” he says. But his magnet has since become the industry standard. Today, there are some 22,000 1.5 tesla MRI machines working around the world, generating 9,000 medical images every hour, or 80 million scans per year.

But Laskaris, now 69 years old, is pushing on. Liquid helium used to cool down the magnets is becoming scarce and his MR team is working on designs that need just a fraction of the fluid. His first machine 30 years ago used 5,000 liters of helium. His latest design in development is projected to need no more than 10.

GE Transportation’s Super Tugger tows an assembled truck of an Evolution Series locomotive. Kumar holds 200 patents, including patents covering axles in such trucks.

Invetor Ajith Kumar atop a GE Evolution Series locomotive.

Then a train came into view and started braking. Standing amid the screeching and noise, Kumar saw up close how a compressed air cylinder applied the friction brakes. “I thought, if you can use it to apply that kind of force on the wheel, surely it could lift the axle,” Kumar recalled.

Armed with this insight, Kumar, a GE consulting engineer, went into his office and filed a patent, which is something that happens to him a lot. In the 36 years he has worked at GE Transportation in Erie, he’s had 200 other Eureka moments – big and small, in his car and at the supermarket – that led to patents. (“I don’t know where it comes from,” he laughs. “But it’s not generally at work.) He lags behind Thomas Edison’s 1,093 U.S. patents but the number puts him on par with the GE’s most prolific modern innovators.

Many of Kumar’s patents are individual but others are shared with colleagues. “Just an idea won’t do the job. You have to have people work on it, too,” Kumar said. “You have to have the implementation and that involves a lot of teamwork.”

Kumar’s drive to improve train technology was a surprise career choice. He joined GE in 1977 from Stanford University, where he studied digital controls. He soon filed his first patent for reducing the current on a transmission line but getting the same amount of power. “The first few are exciting,” he said of his patents.

Some of his greatest achievements came in pursuit of technology that hasn’t yet materialized. Take the hybrid locomotive, which would combine a diesel engine with batteries that capture waste energy from braking – a sort of Prius on rails. “A braking locomotive puts out a lot of kilowatts. Almost a small town could use that,” Kumar said.

Storing that power requires batteries that last 20 years, add only a bit of cost to the vehicle, survive intense vibration and endure temperatures greater than the boiling point of water. Kumar had an idea.

Kumar and his team always start by reviewing and experimenting with existing technology, then seeing what else they can do. Eventually the team hit on metal halides, a material that would be solid and inactive at room temperature but, when melted, becomes a battery. The technology has evolved to become the Durathon battery, which stores as much energy as other battery chemistries but takes up far less space. The Durathon does not yet serve on locomotives, but telecoms use it for backup power and wind farms store excess energy in the batteries.

Another innovation that grew out of the hybrid locomotive project was the Trip Optimizer, a software and big data system that tells a locomotive when to brake and apply power in anticipation of hills and curves in the track. The platform helps trains improve efficiency and save large quantities of fuel. It was another idea that took an unexpected turn and yielded great things.

“We don’t have a hybrid locomotive today but we have two new products, Trip Optimizer and the battery, which hundreds of people work on. So you have no idea where it’s going to go.”

That’s Epic: Ships using GE pump jets will supply Petrobras oil and gas platforms located 180 miles off the coast of Brazil.

“We took the motor and put it in an external pod so it’s now in the water,” says Paul English, marine leader at GE Power Conversion. “Like a jet engine, it has fixed stator vanes inside a nozzle. The vanes straighten the water flow and guide it across the impeller blades. The blades get good water to attack and throw out the back. The result is a more efficient engine with better thrust.”

English says traditional screw propellers produce drag by “spilling” water around the screw tips to the front of the propeller. “When you look over the aft end of a ferry, you see a lot of churning water,” English says. “That’s basically wasted energy. Instead of pushing the water backwards, which is ideal, you are wasting energy on making it roll.” The stator and impeller, a fancy propeller enclosed in a nozzle, greatly reduce the churn.

The pod design also eliminates complicated transmission gears, cuts maintenance, and improves efficiency. “The shaft comes out the back end of the pod and straight into the impeller,” English says. “There are no gearbox [energy] losses at all. We’ve got rid of it.”

The pump jet was originally used in submarines, jet skis and high-speed surface vessels. But GE adapted the technology so that it can now power large supply ships.

GE workers are already making 17 pump jets for eight offshore platform supply vessels, including four ships that will supply deep sea oil and gas platforms operated by Petrobras and located some 180 miles of the coast of Brazil.

The new pods were designed for maximum speed of 16 knots, the oil and gas industry standard. They will work in combination with GE’s data-driven dynamic positioning system, which can keep ships virtually stationary on high seas without an anchor. “The ship algorithms gather location, water current speed and other data, and the computer calculates what thrusts it needs and its direction,” English says. “The pods can turn around the vertical axis and hold the ship at a particular angle. You don’t need a rudder.”

If only Ulysses had a pump jet. He could set his ship on autopilot, his crew could skip the wax earplugs, and they could all enjoy the Siren song together.

The ear sends the sound through an umbilical cable to a seaborne control room filled with computers and engineers dissecting every stray whoosh and hiss. They are listening for oil leaks, odd vibrations, strange cracks and other signs of trouble. “An acoustic signature from a piece of equipment is like a fingerprint from a human,” says Fabian Dawson, sales manager from GE Measurement and Control. “No two leaks or pumps are going to sound the same.” Call it Shazam for the seabed.

Dawson says that the system holds a library more than 100,000 sounds. “We can filter out the background noise, like marine life, and listen only to the things that we want to,” Dawson says. He says that the system is 10,000 times more accurate than traditional “mass balance” systems that measure differences in the amount of oil and gas flowing through the pipes to detect leaks.

GE engineers designed the eardrum of the system, which looks like a giant birdcage, from special crystals that respond to sound wave vibrations and convert them into electricity. (Engineers call this effect “piezoelectricity.”) One ear can listen to sounds within a 1,600 foot radius.

But the system does more than that. An array of attached carbon rods can detect changes in the electromagnetic field generated by electrical cables, pumps, motors and other electrical equipment, and spot ground faults or defective isolation. “You can determine the rpm of a compressor from the acoustic signal, and then you can determine how hard it is working from the electrical signal,” Dawson says. “Taken together, they will tell you what the efficiency is.”

Workers deploy the cage by lowering it down the side of a ship from a crane. They use ROVs to secure it to a piece of subsea equipment or to the sea floor. “It’s a simple, X-marks-the-spot procedure,” Dawson says.

The system is already working at some 130 sites in the North Sea operated by Statoil, ENI, and Shell, and off the coast of Africa. GE has now introduced it to American customers at the 2013 Offshore Technology Conference held this week in Houston.

The system is using powerful software and algorithms to make sense of the sounds. But like an elusive remix you’ve heard at a party, phantom noise can still creep in. “You are looking at correlations between the pieces of the equipment emitting both the electrical and the acoustic signals, and sometimes there is a signature there that you can’t trace.” Dawson says. “We call them rogue sounds.”

When that happens, the GE teams goes straight to the source, the engineers who built the equipment. “They are the experts,” Dawson says. “They know what their machines should sound like. They can recognize the rogue harmonics.”

That focus paid off. Today, Quantcast’s servers are processing 30 petabytes of advertising and ad auction data per day. (For comparison, humanity’s entire written record takes up about 50 petabytes.) Quantcast has 340 employees, offices across the U.S. and in Europe, and reported revenue of $100 million.

This puts Quantcast squarely in the U.S. middle market, a group of 200,000 companies with annual revenues between $10 million to $1 billion that has long been the unsung hero of the U.S. economy. Middle market companies like Quantcast employ 43 million Americans and account for one third of U.S. jobs. The group generates $3.8 trillion in economic output, enough to make it the world’s fourth largest economy after the U.S., China, and Japan.

The segment powered through the financial crisis and added 2.2 million jobs between 2007 and 2010. But it still needs support. That’s why GE Capital, a major lender to the middle market, partnered with Slate, a leading online magazine, to launch Roadshow for Growth. The roadshow is a six-month, multi-city bus tour designed to stimulate debate and economic research, and generate stories about innovation and successful business ideas and make the middle market stronger.

Slate, as a pioneer of web journalism, will help GE tell the middle market stories. Slate Chairman Jacob Weisberg will be on the roadshow bus, as well as top GE Capital executives Mike Pilot, the unit’s chief commercial officer and Dan Henson, president and CEO of GE Capital Americas.

The partners say that the roadshow will allow them to hold town hall meeting and debates across the country, drop in on customers, and talk to entrepreneurs and employees in their communities about issues like domestic and overseas competition, innovation, regulation, and healthcare. They hope that the talks will provide a real-time feedback. Slate publisher Matt Turck said that the roadshow was designed to amplify the middle market’s voice so that it “will also be heard on Capitol Hill.”

The tour starts today in Kansas City, and the roadshow bus will make stops in the Mid-West in St. Louis, Indianapolis, Chicago and elsewhere, before hitting New York and Los Angeles. Our infographics have the details.

Click to enlarge.

 

 

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

Trifon Laskaris’ double-donut design “is still the best configuration” for magnetic resonance imaging during surgery, says Dr. Jolesz.

 

• 
• 
• 
• 
• 
• 
• 
• 
• 
• 

 

CFM’s LEAP engines include for the first time components made from advanced ceramic composites and parts manufactured using 3-D printers.

The company completed design freeze on the LEAP-1A and LEAP-1C sister engines in June 2012. The first full LEAP engine, a LEAP-1A for the Airbus A320neo, is on schedule to begin ground testing in September of this year.

CFM said that it would release detailed LEAP-1B engine design drawings over the next six months. The company plans to start manufacturing parts later this year, build the first engine in early 2014, and proceed with flight testing and certification over the next two years. The first Boeing 737 MAX powered by the LEAP-1B is scheduled to enter service in 2017.

Boeing estimates that the world aircraft fleet will double in size over the next 20 years to some 40,000 planes. Much of the growth will come from single-aisle next-gen planes like the 737, the A320neo, and COMAC’s C919, the LEAP’s target market.

Southwest, Lion Air, AirAsia, Virgin America, Quantas and dozens of other airlines have already placed orders for more than 4,500 LEAP engines.

The LEAP builds on the world’s bestselling jet engine, the CFM56. (Nearly 25,000 CFM56 engines have been delivered to date.) The short-haul market, where these engines serve, demands durability and quick turnaround between takeoffs and landings. The LEAP adds the fuel efficiency of long-haul engines for wide-body aircraft.

New additive manufacturing techniques and 3-D woven carbon fiber composite fans helped cut the LEAP’s weight. Brand new materials from GE Global Research called ceramic matrix composites (CMCs) allow engineers to raise the temperature and pressure inside the engine and make it more efficient. “Because of our innovation, we don’t need to use complex systems such as gearboxes and other tricks,” says Dale Carlson, manager of advanced programs at GE Aviation.

Besides saving millions in fuel costs, the LEAP also generates fewer emissions and less noise, compared to current CFM engines.