Managing Marine Energy and Monitoring Battery Health

In this episode, hosts Kerrin and Taylor dive into three stories with one goal: powering our world with curiosity, creativity, and just a dash of “What if it fails?” Listen in to explore:

• Updates to NREL’s risk management framework. Remember the HERO-WEC from last episode? Researchers are working with the team behind this device to help developers identify, prioritize, and address the challenges of designing and deploying wave energy converters.
• How NREL engineer Bri Friedman will deploy a new underwater wave energy converter. As she prepares to take the SURF-WEC to Hawaii, Friedman is embracing early challenges as a way to accelerate learning and share open data with the marine energy community. Plus,
• Insights into how researchers are fast-tracking battery diagnostics with AI. NREL’s new physics-informed neural network can predict battery health nearly 1,000 times faster than traditional models, combining the speed of AI with the accuracy of physics—and it’s available for free!

This episode was hosted by Kerrin Jeromin and Taylor Mankle, written and produced by Allison Montroy, Hannah Halusker, and Kaitlyn Stottler, and edited by James Wilcox, Joe DelNero, and Brittany Falch. Graphics are by Brittnee Gayet. Our title music is written and performed by Ted Vaca and episode music by Chuck Kurnik, Jim Riley, and Mark Sanseverino of Drift BC. Transforming Energy: The NREL Podcast is created by the U.S. Department of Energy’s National Renewable Energy Laboratory in Golden, Colorado. Email us at podcast@nrel.gov. Follow NREL on X, Instagram, LinkedIn, YouTube, Threads, and Facebook.

Transcript

[intro music, fades] 

Kerrin: Welcome to The NREL Podcast, brought to you by the U.S. Department of Energy’s national laboratory for energy systems integration. We’re highlighting the latest in advanced energy research and innovations happening at the lab. It’s Wednesday, June 25. I’m Kerrin Jeromin. 

Taylor: And I’m Taylor Mankle. Today, we’re going beneath the surface for a look at NREL’s updated risk management framework for marine energy technology and development, the ultimate playbook for keeping ambitious marine projects afloat and storm-ready.

Kerrin: Then, we’ll learn about engineer Bri Friedman, who’s putting a new wave energy converter to the test in the waters of Hawaii, embracing early challenges to help future devices succeed more reliably.

Taylor: And finally, we’re swapping surfboards for circuits with a story about batteries and artificial intelligence. NREL’s new physics-informed neural network can predict battery health nearly 1,000 times faster than traditional models, paving the way for smarter, longer-lasting battery storage.

Kerrin: Three stories, one goal: powering our world with curiosity, creativity, and just a dash of, “What if it fails?”

Taylor: Sounds like good science to me. Let’s get into it!

[music]

Kerrin: The idea of a wave energy converter (or WEC) may sound idyllic, bobbing on ocean waves all day or swaying underwater, quietly generating electricity for people living and working near the shore.

Taylor: Ah, it sure does. But in reality, it takes a lot of careful planning for salt water and electronics to function in harmony. That’s where a robust risk management plan comes into play. 

Kerrin: NREL researchers David Snowberg, Ritu Treisa Philip, and Jochem Weber recently released an updated framework to help marine energy developers better manage risks and improve the success of WECs—from technical challenges to permitting and everything in between.

Taylor: The report revises a 10-year-old framework with a new tool that helps developers spot what might go wrong in the design of a WEC, understand why, and decide what to fix first. It’s free and public, and it helps organizations focus their time and money where it matters most.

Kerrin: Snowberg said it’s not just about the technology—people matter just as much to a marine energy project. Things like staffing and stakeholder support can make or break a project’s success.

Taylor: Last time on the show, we covered the HERO WEC, a wave-powered desalination device that’s been tested in North Carolina’s Outer Banks. It turns out, Snowberg is now working with that same team to put NREL’s updated risk management framework into practice.

Kerrin: That’s right. The WEC world at NREL is a pretty small, tight-knit group. The HERO WEC team is using the framework to redesign their device with survivability top of mind, so it can withstand harsh ocean conditions like storms, waves, and corrosion.

Taylor: With this tool, NREL and the Department of Energy’s Water Power Technologies Office can guide the marine energy industry toward faster, more reliable devices.

[music]

Kerrin: Alright, we’re riding the wave into the next story of Bri Friedman, an engineer on NREL’s water powervalidation team, who’s capturing energy from the ocean, one SURF-WEC prototype at a time.

Taylor: As you could have guessed, SURF WEC is another one of our amazing acronyms and is short for Small Underwater Research Flap Wave Energy Converter. It’s a submerged marine device with a flap that sways with ocean waves to generate electricity. 

Kerrin: In the coming months, the SURF-WEC will go through a design review, where experts and stakeholders will evaluate whether it’s ready for deployment.

Taylor: If all goes well, Friedman and her team, partnering with the University of Hawaii, will send the SURF-WEC for a test run off the Hawaiian coast for up to a year. 

Kerrin: “Up to” are the key words there, Friedman says. Her team actually expects the device to fail in its first year—but don’t worry, that’s all part of the plan! They’ll learn from what goes wrong and share those lessons with the broader marine energy community. 

Taylor: That’s why the team plans to make all of their data—from lab tests to ocean deployment—publicly available on the Marine and Hydrokinetic Data Repository, which is an online  collection of curated and diverse water power datasets built for accessibility and collaboration.

Kerrin: And for Friedman, the SURF-WEC is just the latest step in a journey that began long before her time at NREL. Her path to marine energy started back in middle school science fairs. That love for hands-on learning led her to high school robotics: her favorite after-school activity and the spark that solidified her path toward engineering.

Taylor: She followed that spark to Virginia Tech, where she earned both her bachelor’s and master’s degrees in mechanical engineering. As an undergrad, she interned at NREL through the Science Undergraduate Laboratory Internship program, helping develop a system to test silicon solar cells. 

Kerrin: And it was in her master’s work where Friedman’s water connection first began, when she used drone imagery to build flood models for low-resource areas in Malawi with the African Drone and Data Academy.

Taylor: Her team captured high-resolution aerial images of a refugee camp to map areas that were vulnerable to flooding. They validated the model by comparing their predictions to homes that had already collapsed due to flooding.

Kerrin: Wow. And in 2021, Friedman joined Pacific Northwest National Laboratory as a postgrad researcher, where she explored how drones and small energy devices could support ocean monitoring.

Taylor: That work brought her back to NREL in 2023, where she now focuses on marine energy full time—contributing to projects like SURF-WEC, as well as broader efforts to test, improve, and share wave energy technologies.

Kerrin: Friedman first studied WECs in her 2017 internship at NREL, and now, she might be deploying the SURF-WEC in Hawaii. Talk about a full circle moment, right, Taylor?

Taylor: Definitely.

Kerrin: Friedman says the collaborative spirit of her team is what keeps her grounded here at NREL. 

[music] 

Kerrin: And speaking of strong teams, our last story comes from NREL’s battery lifespan team, where researchers have developed an AI model that checks up on a battery’s internal state without having to take it apart.

Taylor: That’s right. Everyday devices powered by lithium-ion batteries—like smartphones, laptops, essentially anything cordless—face wear and tear over time. Charging, discharging, and even just sitting in different temperatures all slowly break batteries down over time.

Kerrin: But each battery cell is its own complex system of chemical reactions and physical changes, and it’s tough to track those changes quickly and accurately. Most of the time, you can’t really tell what’s going on inside a battery without cracking it open.

Taylor: Now, to address that challenge, the team developed a new AI model. It’s a physics-informed neural network—basically, a deep learning process trained to recognize complex patterns, kind of like how our brain works. It helps researchers get a clearer picture of how batteries degrade over time, without having to take them apart, and it could make battery health checks way more efficient.

Kerrin: Though, this isn’t the team’s first time using machine learning to improve battery performance. They’ve created older techniques, such as the Single-Particle Model and the Pseudo-2D Model, which are widely used and accepted approaches for understanding a battery’s internal health. 

Taylor: The problem is, those models require massive amounts of computations and take a ton of computing power. That makes them too slow for real-time diagnostics. So, NREL researcher Malik Hassanaly proposed a surrogate approach that combines the predictive power of artificial intelligence with the rigor of physics-based modeling.

Kerrin: The result is a two-part study published in the Journal of Energy Storage last summer. It shows how researchers trained and tested their AI model on a wide range of internal battery properties. The model is open source, meaning any battery manufacturer or consumer can download the files and run the AI model on their own batteries.

Taylor: What makes this new model such a game-changer is that it blends physics with AI. Traditional neural networks are great at spotting patterns in data, but they don’t really understand the physical rules that govern how batteries work.

Kerrin: Physics-informed neural networks solve that problem. They’re built to follow those scientific laws while they learn, which means they can predict what’s going on inside a battery with way more accuracy. And because they’re much faster than traditional models, they could make real-time battery diagnostics possible at scale.

Taylor: Now, the team is testing the AI model on batteries from NREL labs, aiming to make it reliable for everyday use. Future work will improve its accuracy and flexibility, so it can handle more battery types and real-time demands.

[music]

Kerrin: Woo! Alright, Taylor, we’ve gone from the ocean floor to the inside of batteries—all in one fabulous episode.

Taylor: Not bad for a few minutes of science reporting. 

Kerrin: I know, I know. Well, that’s a wrap for today’s episode of The NREL Podcast. Thanks everyone for joining us—we’ll be back in two weeks with more news from NREL.

Taylor: This episode was written by Hannah Halusker and adapted from NREL news articles published in May and June 2025 by Jana Wiegand, Tara McMurtry, and Rebecca Martineau. Our theme music is written and performed by Ted Vaca and episode music by Chuck Kurnik, Jim Riley, and Mark Sanseverino, of Drift B-C. This podcast is produced by NREL’s Communications Office.