Ep 253: Clinton Carter - Director, Fleet Modernization, Luminant
Modernizing Comanche Peak and More (0:38)
0:38-10:49 (Clinton Carter discusses how he was involved in modernizing Comanche Peak, building the largest industrial wireless network, and expanding the idea to other locations over the course of his 35 year career there.)
Q. How did you get into the energy and nuclear sector to begin with?
A. (0:42) Clinton Carter grew up in Kansas and worked for an engineering firm. He became fascinated with the technology behind nuclear engineering and the “nuclear magic” that happens to generate power. Clinton remembers when Wolf Creek, the only nuclear reactor in Kansas, was being built and found the technology interesting.
He completed his degree in Nuclear Engineering from Kansas State University. Shortly after, he went to work for the Comanche Peak nuclear power station that was under construction in Texas. The plant was a 2-unit 4-loop pressurized water reactor Westinghouse design. He started in an emergency planning role and worked his way through the plant in the 20 years he worked there, working on projects in operations, licensing, engineering, nuclear administration, reliability programs, and as senior reactor operator.
In Clinton’s first role in emergency planning he had to prepare for any potential risk, as it’s required by law to protect the health and safety of the public and respond to any design basis scenario. He spent his days writing emergency plan scenarios and running drill scenarios to demonstrate Comanche Peak’s readiness to the nuclear regulatory commission in order to receive their operating license and safely operate the plant. Those scenarios include responding to a leak from a reactor coolant system and needing to protect the core, dealing with offsite power outages, and evacuation procedures for both the plant and the public up to a 10 mile radius.
Clinton doesn’t believe the Emergency Planning Zone (EPZ) has been reconsidered after the Fukushima event because it’s a product of analysis identifying areas in most direct harm. After an accident like Fukushima or a similarly devastating event, long-term action would be taken up to 50 miles from the site.
Clinton had an interesting experience while he was in operations at Comanche Peak. He became involved in a new technology-based initiative to modernize internal plant processes, which lead to the development of the world’s largest industrial wireless network in one of the most complex environments in the world, a nuclear power plant. They used the infrastructure of almost 400 access points across the plant to enable more point-to-point communication, improve workforce efficiencies, allow mobile computers and install equipment condition monitoring through wireless sensing technologies. When that project wrapped up at Comanche Peak, the company wanted to expand the success beyond the plant so they built a corporate advanced monitoring and diagnostics center for their entire generating fleet throughout Texas, which they called the Power Optimization Center.
For the last 15 years, he’s been at the corporate offices working to scale the robust 24-7 monitoring facility to optimize each nuclear power plant’s safety, reliability, and performance. They are currently monitoring over 50 stations and over 100 operating units coast-to-coast.
It was a complex project to complete because nuclear regulations are designed to preserve the power plant environment and ensure that everything is in the state it needs to be to respond. Clinton needed to prove that installing a wifis system would never interfere with each particular component.
Nuclear Safety Requirements and What Needs to Change (10:50)
10:50-16:30 (Clinton Carter and Bret Kugelmass discuss how antiquated regulations slow innovation in the nuclear sector and what can be done to change the outlook of nuclear energy in the future.)
Q. At some point, are we too ridiculous with our requirements?
A. (10:52) Clinton thinks that is part of the challenge that the legacy nuclear industry is facing today. If you rewind a few decades, the message was that nuclear power was going to be too cheap to meter, and nuclear energy was not far off from that until disruptive technologies turned the tables and collapsed market prices. While nuclear is still relatively inexpensive by former standards, nuclear energy can’t sustain economic competitiveness with natural gas right now. Over time, as events have happened and the nuclear industry has learned, more and more regulations have been added. The nuclear industry is carrying 30 years of burden and it finds itself stuck in a place where it’s difficult to change.
In order to change that, the nuclear industry needs to look at innovative approaches to comply with regulations and engage regulators. Clinton offers this up as an example; For example, after 9/11, the nuclear industry made a tremendous investment in nuclear security. Even though the threat hasn’t materialized, most of the regulations are still in place. There are some initiatives now to unbundle some of those regulations, but the reality is that these changes take time.
In the interim, part of what they’re doing under their DOE remote monitoring project is looking at new and innovative ways to satisfy regulations. For example, operators currently take local temperature readings in over 100 rooms twice a day to verify temperatures are within a safe range. By installing wireless sensors, the same work can be done at the push of a button.
In America historically, the nuclear industry has done things in a manual fashion. Formal procedures and processes have been built to sustain what works, but now the opportunity is to take all of this manual maintenance and optimize the process to improve efficiency and drive costs down.
We have many examples where other industries have moved forward with technology without the regulatory overhead that the nuclear power industry has to manage. The nuclear industry is dedicated to preserving nuclear safety. If that takes time, it takes time but they will never compromise reactor safety for economic benefit.
A New Industry Demonstration and the Role of Machine Learning in Nuclear (16:30)
16:30-28:30 (Clinton Carter shares the Utility Service Alliance’s $14 million industry demonstration funded in part by the Department of Energy. Clinton and and Bret discuss if there is a necessity for AI or machine learning in the nuclear industry.)
Q: What happens next in your career? When did you move on?
A: (16:34) Clinton was contacted by the president of Utility Service Alliance about three or four years ago. The industry was looking at how to respond to the emerging economic crisis. He began work on an initiative called Delivering the Nuclear Promise together with other members of the industry and the Nuclear Energy Institute. The Utility Service Alliance began asking if they could take the learnings from Clinton’s work on power plant monitoring from Luminant to modernize, centralize, and ultimately drive costs out of the business. The project evolved into meetings with the Department of Energy to seek funding to move in that direction. They built a framework and strategy for how to modernize the fleet using advanced technology and offered the Utility Service Alliance as an industry demonstration, making their findings available to the entire industry after the project.
The project is just getting started. In May 2019, the Utility Service Alliance was awarded a grant from the Department of Energy towards the $14 million dollar cost share industry demonstration. They are currently in contract negotiations with the Department of Energy but expect those negotiations to be completed in the next few weeks, implementing the project at the beginning of 2020.
The objective of the project is to demonstrate how to apply advanced technologies to nuclear businesses and nuclear programs in power plants through the use of advanced field sensing technology that emulate human senses to minimize walk around inspections, data collection, and analysis. They will use cameras, video cameras, infrared sensors, acoustics, and vibration monitoring technologies to emulate human senses. Modern sensing technologies exceed the ability of human monitoring and they are inexpensive compared to a full-time employee.
They are also working with Idaho National Labs, a lab that has many impressive technologies already ready for demonstration in the nuclear environment. Together, they’d like to develop machine learning algorithms to take the data from the field sensors to identify when something warrants human attention.
Clinton says the project is really a three phase initiative. First, basic monitoring and alarm technology. He’s been doing this for 15 years now at Luminant. Now, we’re at a point as an industry where we need to take the next step to derive further value. Secondly, they’ll create a shared services technology platform for USA members. As they develop modules that will pull in data from the plant, analyze them, and deliver results, they’ll learn things through the machine learning algorithms.
Bret discusses his concern in using AI and machine learning for the nuclear industry because the regulator requires that it understands the technology going into each plant. AI and machine learning as understood today inherently undermines that because we literally don’t understand what happens inside the computer. He fears that a rush to bring the technology to fruition welcomes an expensive and prohibitive NRC review.
Clinton will be at NRC tomorrow to have a similar conversation with their R&D department. He encourages Bret to view the machine learning as supplemental monitoring. They are still complying fully with regulatory requirements capably and confidently, but this is above and beyond that to learn something about the environment. It’s not required by regulation to learn that information, but Clinton believes that it will improve the nuclear industry to understand. When nuclear regulations were created, this type of technology didn’t even exist. It wasn’t considered as a possibility. Predictive technologies is the right place to be for satisfying PM requirements and taking credit for monitoring the plant currently, but it may not always be.
The Importance of the Legacy Fleet (28:13)
28:13-32:06 (Clinton Carter shares the importance of the legacy fleet of existing nuclear generating plants to ensure the future of nuclear and the nation.)
Q: Clinton, what are some other problems you see in the industry throughout your career?
A: (28:44) Clinton believes we need to take a closer look at the larger nuclear industry. He sees congress promoting and supporting R&D, and enhancing regulatory processes to support next generation reactor technology and small modular reactor technologies. Clinton finds that fantastic, but worries that he doesn’t see any legislation whatsoever focused on existing nuclear generating fleet. Clinton says the legacy fleet has achieved a safety level that is unparalleled and has never been matched by any other industry in the world. Simply because of economics, ¼ of the generating fleet is most likely going to shut down if nothing changes in the very near future. He says it may be sexy to invest in next generation technology but here you’ve got a fleet of generators that is the envy of every other industry and every other nuclear operation around the globe. Clinton believes that it is important to preserve current nuclear systems to allow next generation nuclear time to develop, test, and scale up. He finds this important to the future of our country, not only the nuclear industry.