Duke Energy Smart Healing Grids

Been following this for some time, impressive results so far.  Much more to be done. 

 Duke Energy Florida's smart-thinking grid improves response, power restoration for customers during Hurricane Ian

Smart, self-healing technology in Florida helped to automatically restore more than 160,000 customer outages during powerful storm

ST. PETERSBURG, Fla., Oct. 24, 2022 /PRNewswire/ -- As Hurricane Ian made its way across Florida, Duke Energy's grid improvements were already on the job helping to combat power outages from the storm.

Smart, self-healing technology helped to automatically restore more than 160,000 customer outages and saved nearly 3.3 million hours (nearly 200 million minutes) of total lost outage time.

"Hurricane Ian is a strong reminder of the importance of grid hardening and storm preparedness to help keep the lights on for our customers," said Melissa Seixas, Duke Energy Florida state president. "Self-healing technology is just one of many grid improvements that Duke Energy is making to avoid outages, restore service faster and increase reliability for our customers."

Much like the GPS in your car can identify an accident ahead and reroute you around the incident to keep you on your way, self-healing technology is like a GPS for the grid. The technology can quickly identify power outages and alternate energy pathways to restore service faster for customers when an outage occurs.

Additionally, self-healing technology provides a smart tool to assist crews in the field with power restoration after a major storm like Ian, helping reduce outage impacts and freeing up resources t help restore power in other locations.

Three days after Hurricane Ian exited the state, Duke Energy Florida wrapped up restoration of approximately 1 million customers. This progress enabled the company to deploy more than 550 Duke Energy workers from throughout Florida, as well as contractors from across the country, to help restore power for Lee County Electric Cooperative customers.

Crews worked in Cape Coral and Pine Island, one of the hardest-hit areas in the storm's path and completed power restoration for the majority of customers on Pine Island within approximately one week after arriving to the island.

Prior to Ian in 2022, smart, self-healing technology had helped avoid nearly 250,000 extended customer outages in Florida, saving around 285,000 hours (17.1 million minutes) of total lost outage time.

Duke Energy currently serves around 59% of customers in Florida with self-healing capabilities on its main power distribution lines, with a goal of serving around 80% over the next few years.

Self-healing technology video.

Self-healing grid system animation video.

Duke Energy Florida

Duke Energy Florida, a subsidiary of Duke Energy, owns 10,300 megawatts of energy capacity, supplying electricity to 1.9 million residential, commercial and industrial customers across a 13,000-square-mile service area in Florida.

Duke Energy (NYSE: DUK), a Fortune 150 company headquartered in Charlotte, N.C., is one of America's largest energy holding companies. Its electric utilities serve 8.2 million customers in North Carolina, South Carolina, Florida, Indiana, Ohio and Kentucky, and collectively own 50,000 megawatts of energy capacity. Its natural gas unit serves 1.6 million customers in North Carolina, South Carolina, Tennessee, Ohio and Kentucky. The company employs 28,000 people.  ... ' 

UFL Thermal Plume Simulation

Simulations of thermal plumes addressed.

HiPerGator Simulations 'Solve Mother Nature' to Address Real-World Problems

University of Florida News, Eric Hamilton

September 20, 2022

University of Florida (UF) researchers reproduced the turbulence and complexity of thermal wall plumes using the university's HiPerGator supercomputer. The complexity of air movements long made it impossible to perform detailed simulations of hot air rising along a vertical surface, but the supercomputer enabled researchers to track air movements on the sub-millimeter level. The simulation utilized 125 of the HiPerGator AI cluster's 140 nodes, each of which hosts eight graphics processing units and 128 central processing units. Said UF's Sivaramakrishnan Balachandar, "Using computers, we solve Mother Nature, and what the computer simulation gives us is unprecedented access into all the details of what happens inside. With our simulation, we can go into the wall plume and see every nook and cranny." 

Hacking Smartphones

Note this requires device proximity

 Hacking Device Can Secretly Swipe, Tap Smartphone Screen

New Scientist

Jeremy Hsu, August 31, 2022

Separate teams of researchers have developed devices for remotely hacking smartphone touchscreens. Both methods involve placing devices featuring an antenna for transmitting signals and a phone locator under a table. The locator deduces the position and orientation of a touchscreen device placed on the table, then the antenna sends electromagnetic signals imitating electric field disturbances caused by physical touch. The Invisible Finger method created by University of Florida researchers caused both iOS and Android devices to recognize the electromagnetic touches; the GhostTouch technique from scientists at China's Zhejiang University worked with multiple Android devices, but could not crack the iPhone 7 Plus or certain OPPO phone models. The hacks only work with the target device lying face down and positioned close to the antenna.

In NewScientist

Voice-assisted AI is Becoming smarter, but Consumers have Higher expectations

I see here is a post by a former school of mine ...   Some reasonable thoughts. 

Voice-assisted AI is becoming smarter, but consumers have higher expectations  by University of Florida

When Huan Chen started researching how consumers think about and perceive artificial intelligence in marketing communications, she was surprised at a major finding: They understand marketing and think it's unavoidable today.

"Marketing companies and organizations will take advantage of new technology," said Chen, and consumers understand that, just like they now have no or low expectations of privacy. "They know these kinds of devices can cache data all the time, (and) if you buy this device, you know you sacrifice your privacy."

Chen, an associate professor in the Department of Advertising, with Telecommunication Professor Sylvia Chan-Olmsted and graduate students Julia Kim and Irene Sanabria, recorded 20 consumers using AI devices and then conducted in-depth interviews to gauge their perspective. Her findings are presented in the paper "Consumers' Perception on Artificial Intelligence Applications in Marketing Communication."

Most people have a general understanding of artificial intelligence used in devices like Amazon's Alexa and Google Home, Chen said. In comparing AI to humans, "they think of AI as smart and intelligent but not as sophisticated or complicated enough. Even though they communicate and interact with AI, they still don't think AI can do complicated or sophisticated tasks. They think human beings still need to do that kind of task."

Some people wish AI devices could master more functions to help them with their daily lives, Chen said. For now, how they are using voice-assisted AI focuses on five aspects:

Function: Consumers understand and appreciate the various functions AI can offer, like "Alexa, play music" or "Find a recipe."

Communication: "Consumers have a love/hate relationship with AI," she said. "They are frustrated, especially if the device couldn't recognize what they are talking about."

Adaptation: How does AI adapt to users during communications and interactions, and vice versa? "The voice-assisted devices using AI have become smarter and smarter to know users better. At the same time, human beings also adjust and change. One man figured out a way to give an order, to talk to Alexa to make that communication more effective. That's a mutual adaptation."

People are cautious about their relationships with AI.

Privacy: Privacy is still an issue, Chen said, but not as much as it once was. Technology may be desensitizing some people, who now think their data is being collected everywhere, all the time. Others "do not think they are so important that people will do something with the data."  ... ' 

UF Acquires NVidia DGX System

Notable former connection working with new Nvidia System. 

UF becomes first U.S. university to acquire cutting-edge NVIDIA DGX A100 system, advancing its artificial intelligence initiative - News - University of Florida

Moving forward on its sweeping vision to transform the future of education through artificial intelligence (AI), the University of Florida is collaborating with technology company NVIDIA to acquire the world’s most advanced AI system to boost the performance of UF’s powerful supercomputer.

UF today announced it will be the country’s first higher education institution to acquire the new NVIDIA DGX™ A100 — the world’s most advanced AI system.

The new systems are scheduled to arrive at UF at the end of May and mark a significant step in the university’s bold initiative to become a national leader in the application of AI, an expansive plan that will elevate UF in research, teaching, and economic development. The initiative includes a commitment from UF to hire 100 faculty members specifically focused on AI, in addition to the 500 new faculty hired across disciplines -- many of whom will integrate AI into their teaching and research.​ UF is also embarking on a unique plan to infuse AI across academic majors, creating a next generation AI-enabled workforce and democratizing a technology that has the potential to solve some of the globe's most formidable challenges.   ... " 

Simulating Molecular Motion with Neural Networks

Have now seen several examples of using neural nets to simulate complex systems by generating a model from data.    Thought provoking example I am following up on.

Researchers Cast Neural Nets to Simulate Molecular Motion 
Los Alamos National Laboratory News
By Nancy Ambrosiano      July 2, 2019

The U.S. Department of Energy's Los Alamos National Laboratory (LANL), the University of North Carolina at Chapel Hill, and the University of Florida demonstrated that artificial neural nets can be taught to encode quantum mechanical laws that define molecular motion, potentially advancing simulations across many disciplines. Said LANL's Justin Smith, "We can now model materials and molecular dynamics billions of times faster compared to conventional quantum methods, while retaining the same level of accuracy." The researchers developed a machine learning technique to build empirical potentials—atomic dynamics descriptions that follow classical physical and Newtonian laws—from data collected about millions of compounds. The transfer learning technique can be applied to new molecules in milliseconds .... " 

Researchers cast neural nets to simulate molecular motion
Machine learning allows quantum mechanics to be efficiently applied to molecular simulations for drug development, detonation physics and more .... " 


Publication:   J. S. Smith, B. T. Nebgen, R. Zubatyuk, N. Lubbers, C. Devereux, K. Barros, S. Tretiak, O. Isayev, A. E. Roitberg, “Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning,” Nature Communications 10.1038/s41467-019-10827-4 (2019)  ..... " 

Technical paper:  https://www.nature.com/articles/s41467-019-10827-4