Lithium-ion batteries power electric vehicles and grid storage, but their internal health remains largely invisible. A new review published in the Journal of Zhejiang University–SCIENCE A examines how nondestructive sensing and failure diagnosis can turn hidden aging signals into actionable health information. The paper, with DOI 10.1631/jzus.A2600100, was authored by researchers at Zhejiang University’s State Key Laboratory of Chemical Engineering and proposes a unified framework linking degradation mechanisms to measurable physical signatures.
Battery aging involves complex chemical, mechanical, and thermal processes—such as solid electrolyte interphase growth, particle cracking, and lithium plating—that can lead to internal short circuits and thermal runaway. Conventional battery management systems track only voltage, current, and surface temperature, but these external signals are distorted by polarization, side reactions, and delays. Advanced imaging techniques like synchrotron and magnetic resonance offer mechanistic insight but are limited by scale, cost, and speed for real-time use.
The review organizes nondestructive diagnostics into four families: surface-attached sensors (thermocouples, RTDs, fiber Bragg gratings), implantable sensors (MEMS, thin-film strain gauges, optical fibers), in situ integrated designs (embedded in current collectors, separators, or packaging), and noncontact methods (magnetic-field imaging, acoustic probing, gas analysis, electrochemical impedance spectroscopy). Each approach has trade-offs. Surface sensors are low-cost but suffer from signal delays under high-rate operation. Implantable sensors improve fidelity but face long-term stability and vibration fatigue challenges. In situ integration offers minimal intrusion, while noncontact methods provide system-level insight without direct contact.
To make sense of these heterogeneous signals, the review discusses feature extraction techniques like incremental capacity analysis, differential voltage analysis, and differential thermal voltammetry. These can be interpreted using physics-based models (pseudo-two-dimensional, single-particle) and data-driven approaches such as physics-informed neural networks, transformers, and CNN-LSTM architectures. The authors emphasize feature selection, dimensionality reduction, cloud-edge collaboration, and standardized interfaces to distinguish normal battery “breathing” from lithium plating, gas generation, and microcracking.
The implications for industry are significant. The framework could enable earlier thermal-runaway warnings, more accurate state-of-health and remaining-useful-life estimates, and smarter fast-charging. In electric vehicles, it may allow predictive maintenance and cell-to-pack safety monitoring. For grid storage, it could improve fleet-level reliability, second-life assessment, and fire prevention. Low-cost strategies using existing voltage, current, and temperature signals combined with cloud-edge computing could ease deployment. However, sensor stability, manufacturing compatibility, data standardization, bandwidth, and cost remain key barriers.
The authors stress that no single sensing modality can fully capture battery degradation. The real advance comes from fusing surface, implanted, in situ, and noncontact signals with physics-informed algorithms. This sensing–algorithm co-design could make batteries more predictable, reliable, and durable. As the field shifts from external observation to direct internal perception, it promises to move laboratory advances into scalable battery systems, benefiting electric transportation and stationary storage alike. Additional information is available at http://chuanlink-innovations.com.
