Wang succeeds Judy Jeevarajan, Ph.D., who will remain with UL Research Institutes as vice president and distinguished scientific advisor. In his new role, Wang will direct research strategy and operations at the Electrochemical Safety Research Institute, an organization focused on understanding why electrochemical energy-storage systems fail and how those failures can be prevented. The institute investigates the links between battery chemistry, mechanical design, electrical control, manufacturing quality, and real-world operating conditions. Its work is intended to influence safer products, testing methods, engineering practices, and standards as energy systems become more densely packed and increasingly dependent on rechargeable batteries.

Before joining UL Research Institutes, Wang spent more than three decades at Pennsylvania State University, where he served as the William E. Diefenderfer Chair in Mechanical Engineering, professor of mechanical engineering, chemical engineering, and materials science and engineering, director of the Electrochemical Engine Center, and co-director of the Battery and Energy Storage Technologies Center. His research career has covered the full battery-development chain, from fundamental electrochemical reactions and heat generation to manufacturing, fast charging, system integration, and commercialization. That breadth has made him a prominent figure in a field where the performance of a battery cannot be separated from its thermal behavior, structural integrity, production consistency, and control software.

One of Wang’s most important contributions has been the development of electrochemical-thermal modeling methods for batteries. These models combine the movement of ions and electrons inside a cell with the heat produced by electrochemical reactions, electrical resistance, and transport limitations. As a battery charges or discharges, local variations in current density, temperature, and chemical concentration can create areas of stress that are invisible from the outside. Electrochemical-thermal models allow researchers to predict how these internal conditions evolve, helping engineers design cells and battery packs that operate within safer limits. The approach has influenced battery development across transportation, consumer electronics, defense, and stationary energy storage.

Wang also invented a self-heating, all-climate battery designed to maintain performance in extremely cold conditions. Conventional lithium-ion batteries can lose power at low temperatures because ion transport slows and the internal resistance of the cell rises. Charging a cold battery can be particularly hazardous, as lithium plating may occur on the anode surface instead of lithium ions being safely stored within the electrode structure. Wang’s technology uses the battery’s own electrical energy to generate controlled internal heat, bringing the cell rapidly to an operating temperature at which it can deliver power and accept charge more efficiently. The technology enabled electric buses to operate during the 2022 Winter Olympics and has since been commercialized for transportation, defense, and energy-storage applications.

His work on ultrafast charging also drew international attention after being recognized by The Guardian as one of the world’s leading science stories of 2022. Fast charging is not simply a matter of supplying more electrical current. High charging rates can produce heat, accelerate unwanted chemical reactions, and cause lithium ions to accumulate as metallic deposits on the anode. These deposits can reduce capacity and, in extreme cases, create internal pathways that trigger a short circuit. Wang’s research has explored how electrode architecture, thermal management, charging protocols, and cell chemistry can be coordinated to reduce these risks while shortening the time required to recharge a battery.

More recently, Wang has focused on lithium-metal and solid-state batteries, two technologies widely viewed as possible successors to today’s dominant lithium-ion systems. Lithium-metal anodes can store substantially more charge by replacing conventional graphite, potentially increasing energy density and extending the range of electric vehicles. However, lithium can form needle-like structures known as dendrites during charging. If dendrites penetrate a separator and reach the opposite electrode, they can cause an internal short circuit. Solid-state batteries replace the flammable liquid electrolyte used in many conventional cells with a solid ion-conducting material, but they introduce their own challenges, including interfacial resistance, cracking, contact loss, and mechanical instability. Wang’s research has examined the safety mechanisms behind these emerging systems and contributed to the design of batteries intended to be intrinsically safer rather than merely protected by external controls.

“Dr. Wang is among the world’s foremost authorities on battery technology and electrochemical energy systems,” said James J. Hudgens, Ph.D., president and chief executive officer of UL Research Institutes. Hudgens said Wang’s scientific leadership, entrepreneurial approach, and focus on battery safety made him especially qualified to lead the Electrochemical Safety Research Institute as demand for energy storage accelerates. Wang said batteries are fundamental to the future of transportation, infrastructure, and energy systems, and that he would work with colleagues across UL Research Institutes to advance research that improves the safety, reliability, and sustainability of energy technologies worldwide. His responsibilities will include expanding experimental and computational capabilities and strengthening partnerships with industry, government, universities, and standards-development organizations.

Wang’s appointment also brings an unusually extensive record of invention and technology transfer to a research institute whose findings are intended to inform public safety. He is a fellow of the National Academy of Inventors, the Electrochemical Society, and the American Society of Mechanical Engineers, holds approximately 140 issued patents, and has authored research cited more than 50,000 times. He earned bachelor’s and master’s degrees in mechanical engineering from Zhejiang University and a doctorate in mechanical engineering from the University of Iowa. Throughout his career, he has founded companies and helped move laboratory discoveries into commercial products. At UL Research Institutes, that combination of fundamental science, engineering, and commercialization could help close the gap between promising battery concepts and the safety requirements of technologies deployed at global scale. The institute, part of the nonprofit UL Research Institutes, conducts independent research across electrochemical safety, fire safety, chemical insights, materials discovery, digital safety, and research education, publishing findings openly to support safer standards, policies, products, and communities.

Subject of Research: People

Article Title: Battery Safety Pioneer Chao-Yang Wang Appointed to Lead UL Research Institutes’ Electrochemical Safety Research Institute

Web References: Chao-Yang Wang, Ph.D.; Electrochemical Safety Research Institute; James J. Hudgens, Ph.D.

Image Credits: UL Research Institutes

Keywords

Battery safety, electrochemical energy storage, lithium-ion batteries, lithium-metal batteries, solid-state batteries, ultrafast charging, thermal runaway, battery research, electric vehicles, energy storage safety

At the core of this research is the fundamental understanding of the battery’s SOH—an indicator that represents the current condition of the battery in comparison to its optimal performance metrics. The SOH assessment is crucial for predicting a battery’s remaining useful life and ensuring that systems relying on these batteries can operate safely and effectively. Traditional methods of estimating SOH often involve complex empirical models that can be limited in accuracy and scalability, particularly as battery systems grow in complexity.

To address these limitations, Lyu and colleagues adopted a more sophisticated approach involving the integration of advanced neural network architectures. By utilizing a Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory (BiLSTM) network, paired with attention mechanisms, they aimed to significantly enhance the accuracy of SOH predictions. This combination allows the model to effectively capture the temporal dynamics of battery ageing and the intricate patterns embedded within the data.

Central to their method is the application of ICA—a statistical technique that separates a multivariate signal into additive, independent components. By employing this technique, the researchers were able to distill key features from the battery ageing data, discarding noise and focusing on the most informative signals related to battery health. This preprocessing step is critical, as it directly impacts the neural network’s ability to learn and make predictions based on clean, relevant input data.

The architecture of the CNN–BiLSTM–Attention network used in their study is particularly noteworthy. The CNN layers are designed to extract spatial hierarchies in the data, allowing the model to discern local patterns indicative of battery performance. Following this, the BiLSTM layers provide the ability to remember long-term dependencies in sequential data, which is essential given the time-series nature of battery performance metrics. The attention mechanism further refines this process, allowing the model to concentrate on the most significant features over others.

Through rigorous training and validation, the researchers demonstrated that their optimized network outperformed conventional SOH estimation techniques. The results indicated a marked improvement in accuracy, with the CNN–BiLSTM–Attention network achieving a prediction success rate that exceeded other established methodologies. This advancement is a significant leap forward, marking a new paradigm in the accurate monitoring of battery health.

The implications of this research are far-reaching, especially as energy storage solutions become increasingly critical in various sectors, including electric vehicles, renewable energy systems, and consumer electronics. By enhancing SOH estimation, the proposed methodology has the potential to extend the life expectancy of batteries, improve their safety profiles, and optimize their operational efficiencies.

Moreover, the integration of machine learning techniques into battery management systems represents a transformative shift in how battery health can be monitored and managed. As machine learning algorithms continue to evolve, they offer the promise of real-time monitoring and predictive maintenance capabilities that could further revolutionize battery performance management.

As a result, this study not only paves the way for more dependable battery health assessments but also highlights the critical intersection of machine learning and energy storage innovations. The findings elucidate how emerging technologies can be harmonized with traditional energy systems to foster a sustainable future.

The authors emphasize that while their model shows promising results, further research will be essential to validate its effectiveness across various battery chemistries and operating conditions. Continuous improvement in data collection methods and model training will be necessary to ensure that the optimized network remains applicable in real-world scenarios.

Future iterations of this research could also explore the capabilities of integrating other complementary machine learning approaches alongside the CNN–BiLSTM–Attention framework. By doing so, researchers may uncover even more intricate understandings of battery behaviour and health assessment methodologies.

Ultimately, Lyu et al.’s study marks a significant contribution to the field of battery technology, providing a fresh perspective on how machine learning can enhance SOH estimation. As batteries continue to power our world, innovations like this underpin the journey towards more intelligent and sustainable energy solutions.

Thus, as we look to the future of battery technology, the work of Lyu and his team serves as a beacon of progress, demonstrating the vital role that advanced computational techniques will play in fostering energy innovations that can withstand the test of time.

Subject of Research: Battery SOH estimation using an optimized CNN–BiLSTM–Attention network with ICA-Based ageing features.

Article Title: Battery SOH estimation via an optimized CNN–BiLSTM–Attention network using ICA-Based ageing features.

Article References:

Lyu, Z., Wang, H., Shi, W. et al. Battery SOH estimation via an optimized CNN–BiLSTM–Attention network using ICA-Based ageing features.
Ionics (2026). https://doi.org/10.1007/s11581-025-06933-7

Image Credits: AI Generated

DOI: 12 January 2026

Keywords: Battery health, SOH estimation, Machine learning, CNN, BiLSTM, Attention mechanism, ICA, Energy storage, Predictive maintenance.