The GE Brief: August 11, 2020
Nuclear power plants can produce huge amounts of electricity with a relatively small carbon footprint, but it comes at a cost — literally. Building, maintaining and operating those plants is expensive. Aiming to keep nuclear power competitive, the Department of Energy’s Advanced Research Projects Agency energy division, or ARPA-E, is seeking ways to reduce operations and maintenance costs down to $2 per megawatt-hour, paying particular mind to next-generation advanced reactors. That’s no small feat, but GE scientists have a few ideas for how to go about it.
The nucleus of the matter: GE Research, along with other institutions, is the recipient of a three-year ARPA-E grant to investigate how artificial intelligence and computer simulations can help monitor and maintain nuclear reactors. GE engineers have already been working with a couple of concepts that they hope will show results: One is the “digital twin,” a virtual representation of a machine, part or process that the company already uses to monitor jet engines, gas turbines and other assets and systems. The other is “humble AI,” a machine learning program that recognizes when certain data are missing and searches for more details, or asks for human input. The hope is to use such technologies to provide nuclear plants with the ability to predict when maintenance issues might arrive, thus leading to cost savings for the power plants of the future.
Learn more here.
The year was 1941. Across the Atlantic Ocean, World War II was raging, with German bombs raining down on London. Just outside of Boston, a group of GE engineers gathered to unwrap a secret package from England. The contents? Parts for the first jet engine successfully built and flown by the Allies — called the Whittle engine after its designer, Royal Air Force officer Frank Whittle. The engineers’ job? The U.S. War Department and the Army Air Corps had commissioned GE to rebuild and commercialize the engine. They had to quickly learn inside and out how the engine worked, create a plan to bring it to mass production — and help Britain win the war.
The highest heights: The work remained top secret; in fact, the group of engineers came to be known as the Hush-Hush Boys. When the likely last living member of the GE team, Joseph Sorota, spoke to GE Reports in 2016 — 10 months before passing away at the age of 96 — he recalled being warned by the FBI “that if I gave away any secrets, the penalty was death.” In the summer of 1942, Sorota and his colleagues loaded the first pair of working jet engines onto a railcar and shipped them to an Army field in the Mojave Desert, where they thrust America’s first jet plane, the XP-59, to a height of 6,000 feet — and launched the United States into the jet age. GE, too: Now the company makes engines 100 times more powerful than Sorota’s original. “It never dawned on me it was going to turn over the entire aircraft industry like it did,” Sorota said.
Click here for the awesome story of Joseph Sorota, the Hush-Hush Boys and a run of technological innovation that changed flight forever.
1. Go Figure
A team of MIT computer scientists designed a machine learning system to predict how well a COVID-19 vaccine might work among broad populations of people.
2. Computer, Virus
Researchers enlisted Summit, the world’s fastest supercomputer, to analyze the lung fluid of COVID-19 patients and help shed light into the virus’ bewildering array of symptoms.
3. Spine Science
At the University of California, San Diego, scientists successfully implanted neural stem cell grafts into mice with spinal cord injuries, then tracked the progress of the implants.
Read more here about this week’s Coolest Things on Earth.
— QUOTE OF THE DAY —
“This is a very futuristic, forward-looking program.”
— Abhinav Saxena, senior scientist at GE Research
Quote: GE Reports. Images: Getty Images.