NSF CAREER Award To Develop Next-Generation Energy Diagnostics
MIE Assistant Professor Juner Zhu received a $630K NSF CAREER award for “Interpretable Electro-Chemo-Mechanical Analytics, Diagnostics, and Artificial Intelligence for Porous Electrodes.”
Abstract Source: NSF
This Faculty Early Career Development Program (CAREER) grant aims to advance national prosperity and secure national defense by developing new ways to understand, monitor, and manage complex energy storage and conversion systems. Today’s energy systems—such as batteries in electronics, the power grid, and defense technologies—are increasingly complex, yet they are often monitored using only electrical signals. This limited perspective makes it difficult to detect early signs of failure or to fully understand how these systems behave under real-world conditions. The proposed research introduces a new paradigm that integrates multiple types of information, including electrical, mechanical, and thermal signals, and uses artificial intelligence to interpret them together as a complete system. By enabling real-time health monitoring and early fault warning, this approach will improve the safety, reliability, and resilience of energy technologies. The outcomes of this project will support the energy, transportation, and defense sectors through the development of next-generation monitoring hardware and software, while also educating a new generation of engineers with interdisciplinary skills critical to the U.S. manufacturing and energy workforce.
The intelligence and resilience of advanced energy systems depend on the ability to manage complex interactions among electrical, chemical, and mechanical processes. However, few existing methods can directly analyze nonlinear and irreversible interactions across multiple physical fields. The long-term goal of this CAREER project is to establish a scientific foundation for understanding multi-field dynamic systems and to address a central question: whether the state and evolution of one physical field can be inferred, diagnosed, and controlled through measurements of another. To achieve this goal, the project will integrate experiments, theory, and computation to develop a unified reduced-order framework for electrochemical and mechanical coupling in porous systems. A new analytical methodology, mechano-electrochemical impedance spectroscopy, will be developed to directly measure coupled electrochemical–mechanical responses under controlled perturbations. Microstructure-resolved simulations will be employed to investigate how heterogeneity, nonlinearity, and irreversibility emerge from evolving porous architectures. These physics-based insights will be combined with machine learning techniques to enable system diagnosis and health assessment without reliance on long-term historical data. In parallel, the education and workforce development activities will tightly integrate research and teaching by introducing cross-linked instructional modules in dynamics and machine learning, aligning curricular content with industrial needs through partnerships and cooperative education, training students in entrepreneurship and technology translation through a student-led venture, and engaging the public through a community-facing initiative focused on battery safety, reliability, and end-of-life management.