The aging behavior critically affects the safety and service life of lithium-ion batteries. To overcome the challenges associated with time-consuming and costly aging tests, this study proposes a physics-informed data-driven framework that integrates an electro-aging model with deep learning techniques to enable efficient prediction of temperature-dependent battery degradation. The electro-aging model is first employed to simulate battery aging behavior over a wide practical temperature range, instead of extensive multi-temperature aging experiments. Based on the data generated by the physical model and experimental aging data measured at 5 temperatures, a transfer learning model is developed. Through a fine-tuning strategy, the model can allow efficient estimation of long-term degradation using only a single initial constant-current charging cycle at a specific temperature point. Validation using sparse experimental data demonstrates that the proposed model achieves root mean square errors below 5.43 Ah for a 280 Ah cell, corresponding to only 1.94% of the nominal battery capacity across a temperature range of 10–75°C.

Abstract Battery fast-charging optimization is essential for maximizing service life. A major challenge lies in accurately predicting degradation under diverse charging conditions. Due to the high cost of experiments, data-driven prediction models often suffer from limited data. This results in poor interpretability and generalization of the models. This study proposes a novel approach to evaluate the interpretability of aging prediction models, aiming to assist model development. The approach proposes a new evaluation metric called the rank loss. This metric quantifies the degree to which model predictions on diverse charging conditions deviate from aging prior knowledge. The method was validated on a dataset comprising batteries across multiple charging conditions. Based on this, the interpretability of multiple neural networks was evaluated.

Battery fault diagnosis is a prerequisite for safety certification to electric vertical take-off and landing (eVTOL) aircraft. Unlike ground vehicles, eVTOL systems operate under a zero-failure tolerance regime, as even minor battery anomalies can compromise flight safety. However, existing diagnostic methods exhibit two fundamental limitations: their structural dependence on single-fault assumptions and the limited adaptability to high-rate electro-thermal coupling under aviation duty cycles. To address these challenges, this study proposes a structurally decoupled, model-based diagnostic framework for battery modules. A cell-level electro-thermal coupling model is established to capture dynamic electrical-thermal interactions under high discharge rates. Four minimal structurally over-constrained subsystems are constructed to generate analytically decoupled residuals. This design enables systematic isolation of seven fault categories, including short circuits, interconnection faults, and multiple sensor failures, while preserving diagnosability under concurrent fault conditions. Kalman filtering enhances state estimation robustness, and an adaptive threshold strategy accommodates varying operational regimes. Under single-fault scenarios, the proposed method achieves a detection rate of 93.88%, enabling isolation and parameter estimation of all seven fault types. More critically, under concurrent-fault scenarios in a 10S3P module with 4323 possible fault-pair combinations, the types of individual faults can be identified in 85.5% of cases, and in 98.3% of cases, the types can be narrowed down to fewer than three candidates.

Regulating battery charging is one of the most effective approaches to extending battery lifetime, which relies on accurate prediction of battery aging under diverse charging conditions. Data-driven methods have been widely adopted. To tackle the problem of limited data, physics-informed neural networks (PINNs) and data augmentation show strong potential for guiding data-driven models by leveraging aging-related physical knowledge. However, this often requires accurate physical aging models or laws. To overcome such a challenge, this study proposes a new method based on ensemble learning. A batch of multi-layer perceptron (MLP) based aging prediction networks are first trained using real data. Each network can predict average degradation rate over the next 200 cycles using one cycle charging data. By using charging data under multiple charging conditions generated by an electro-thermal model, the physical consistency of each network was further evaluated and ranked with respect to well established rules that can reflect the influence of various charging factors on aging. Based on the results, the most well-performing MLPs were selected. By using ensemble learning, the aging prediction can be obtained through combining predicted results from each selected MLP. An aging dataset covering 45 charging conditions was constructed for validation. Cross-validation demonstrated that the proposed method can reduce prediction root mean squared error (RMSE) by 7-17% on degradation rate across different charging conditions, simultaneously guaranteeing physical consistency.

Urban air mobility (UAM) can reshape city transit but concurrently introduces complex and unprecedented safety challenges. With regard thereto, this study involves a systematic examination of the state of UAM safety through a structured survey of the relevant literature; thereafter, a novel structured matrix-based classification and analysis framework (“UAM-SafeM”) is proposed. This framework is used to analyze safety from two complementary perspectives: the operational dimensions of UAM, encompassing flight, airway, operational, and societal safety; and the core analytical process, wherein risk sources, assets, and mitigation technologies are examined. Through mapping of the process onto the operational architecture, this approach ensures a comprehensive examination. A thorough review of potential risks and state-of-the-art mitigation technologies is then conducted across the four safety layers, with findings from a wide body of literature considered. Critically, UAM-SafeM underscores that UAM safety is shaped not only within individual layers, but also by interactions between layers, including risk propagation and mitigation interdependencies. On the basis of these relationships, research gaps are herein identified, and a matrix-informed roadmap is proposed to support scalable, safe, and socially acceptable deployment.

Contemporary battery systems demand extended durability and high safety. Inspired by biological self-healing, self-healing strategies have shown promise in mitigating chemical degradation and physical damage. This review summarizes self-healing materials and mechanisms reported for battery applications, comparing their effectiveness. Importantly, we find that current research focuses on individual components, while device-level holistic self-healing remains largely unexplored. Distinct from prior reviews that concentrate on advancing self-healing technologies, this article maps self-healing needs to representative application scenarios and outlines a vision for intelligent battery systems. This review provides guidance for material selection and proposes a paradigm shift: transforming batteries from static energy containers into active energy systems.

Accurate state of charge (SOC) estimation is critical for fully utilizing the capacity of lithium-ion batteries (LIBs) in smartphones and preventing unexpected shutdowns. However, sluggish solid-phase diffusion and intensified polarization at low temperatures cause voltage responses to differ from those at room temperature, leading to large SOC estimation errors when using conventional equivalent circuit models (ECMs). To address these limitations, based on the solid-phase diffusion mechanism, this study systematically analyzes the differences in constant-current discharge behavior at low versus room temperatures, and reveals the respective failure mechanisms of conventional ECMs in the high SOC region and the end of discharge (EOD) region. Building on these findings, an improved ECM is proposed, comprising a physically interpretable Voigt element that accurately captures solid-phase diffusion kinetics, and an adaptive resistor REOD that accurately captures the sharp increase in solid-phase diffusion polarization and the corresponding steep voltage drop in the EOD region. Moreover, the parameters of the improved ECM evolve dynamically with temperature and SOC, enabling the model to accurately reflect electrochemical dynamics at low temperatures. Experimental results show that, at -10 degrees C, an extended Kalman filter based on the improved ECM rapidly corrects initial SOC errors of up to 10% across four dynamic profiles used in smartphones, achieving root mean square error below 2%. The improved ECM helps enhance SOC estimation accuracy, robustness, and interpretability with low computational cost, offering strong potential for smartphone battery management systems.

Long-term warning of lithium-ion battery failures provides a critical window to reduce safety risks in electric vehicles (EVs). However, achieving long-term warning remains a significant challenge due to highly complex EV operation conditions and the subtle signatures of early-stage failures. To address this, a novel unsupervised clustering-based long-term warning method is proposed in this study. Firstly, core battery features are extracted using daily-scale windows. The voltage sorting average feature is processed through an adaptive window filtering to enhance the detection of fault signatures. Secondly, four highly separable fault features are constructed through derivative processing of the core battery features. Together with the longitudinal outlier average, these features constitute a five-dimensional feature matrix. Thirdly, a comprehensive determination mechanism is designed to utilize peak-to-plateau separation to obtain feature weights, thereby avoiding feature dilution caused by equal-weight averaging. Then, unsupervised clustering is employed to prevent score overestimation and false alarms that may arise from feature averaging. Finally, the proposed method is validated using 30 real-world EV operation datasets, demonstrating high accuracy and long-term warning capability. This study provides a highly generalizable and practical solution for battery safety warning in real-world EV environments.

Although incremental capacity (IC) curve analysis is promising for early internal short circuit (ISC) detection, its diagnostic accuracy degrades significantly across a wide resistance range, especially for low-resistance events dominated by leakage currents. To overcome this limitation, we propose a threshold-adaptive ISC diagnostic framework that dynamically integrates quantitative resistance calculation (for high resistance, R ≥ 100 Ω) with Gaussian process regression (GPR)-based leakage current analysis (for low resistance, R < 100 Ω). Validated on 20 Ah LiFePO4 batteries, this approach achieves <6% error for 100–300 Ω and <8% error for <100 Ω (after GPR correction), demonstrating robust, implementation-ready solutions for real-world battery safety monitoring.

Lightweight structural materials with high strength and multifunctionality are urgently needed for advanced applications in aerospace, automotive, and sustainable engineering. However, conventional composites often suffer from limited mechanical performance, anisotropic mechanical behavior, or non-sustainable resource. Inspired by the Bouligand structure found in arthropod cuticles, we report a scalable approach for fabricating high-performance nanocellulose composites with a biomimetic gradient helical organization. By employing a programmable assembly process involving aligned cellulose nanofiber layers and an optimized interfacial matrix, we successfully constructed a dense, macroscopically isotropic bulk material. The multi-level hierarchical design promotes efficient energy dissipation through mechanisms such as microcrack deflection, interlayer sliding, and dynamic hydrogen bonding across scales. As a result, the composites exhibit outstanding mechanical performance, achieving a tensile strength of 649.9 MPa, fracture toughness of 192.1 MJ·m-3, and puncture resistance of 178.4 N/mm-values that substantially exceed those of leading natural and synthetic structural counterparts. Moreover, the material demonstrates multifunctional characteristics, including tunable structural coloration, effective electromagnetic interference shielding, and exceptional stability across extreme temperatures. This work establishes a versatile and sustainable platform for the development of advanced structural materials suited for demanding environments such as spacecraft shielding, robotic systems, and next-generation vehicular technologies.

Abstract Facing growing demand for high-capacity, long-life lithium-ion batteries (LIBs), particularly those targeting over 10,000 cycles, scientific verification and precise assessment of their theoretical lifespan are critical challenges for system reliability and safety. Accelerated aging testing is an effective approach for efficient, short-term lifespan assessment, whose key lies in ensuring consistent aging mechanisms under accelerated conditions and the precise definition of corresponding boundaries. This study focuses on temperature as a critical accelerating stress factor, systematically investigating the aging failure mechanisms of LIBs under both low-temperature and high-temperature environments. Primary innovations include: applying characterization techniques to reveal low-temperature aging phenomena dominated by reversible and irreversible lithium plating and identify its critical boundary; and elucidating consistent high-temperature aging mechanisms within the 25°C to 85°C range. Based on these mechanism boundary findings, a multi-stress accelerated aging test protocol spanning low-temperature and high-temperature ranges was designed; furthermore, grounded in the Arrhenius equation, a kinetic dependency model relating battery capacity fade rate to temperature was developed. This successfully validated the feasibility of extrapolating battery service life under normal 25°C operating conditions using three-month accelerated aging data, thereby confirming the rationality of the boundary design. This study establishes a mechanism-driven aging boundary research framework, which helps reduce the time consumption and cost of traditional lifespan testing, providing technical basis for lifespan assessment and safety optimization of energy storage systems.

The accurate prediction of battery capacity degradation trajectories is essential for ensuring device safety; however, existing analytical models often overlook the uncertainties of future operating conditions. An analytical approach that incorporates future operating condition variability into capacity degradation trajectory prediction is presented. A dual-exponential model is used to model the nonlinear degradation behaviors, and Box-Cox transformation with variable coefficients is applied to describe the trajectory's dependence on future operating conditions. In addition, particle filtering is employed to dynamically predict the capacity degradation trajectories and confidence intervals under random future conditions. Experimental results show that the proposed approach achieves a median root mean square error below 0.172A center dot h using only the first 25 cycles, exhibiting a 52.8% improvement in accuracy over existing methods.

Abstract High-precision prediction of lithium-ion battery degradation trajectories is a crucial foundation for achieving optimal lifespan and ensuring safety management. However, due to the strong nonlinearity of the aging process and the coupled influence of the initial state of health (SOH) and operating conditions, traditional point-by-point prediction methods suffer from inherent limitations such as long-term error accumulation and insufficient adaptability to varying situations. To address this, this study proposes a dimensionality reduction modeling and reconstruction method for capacity degradation trajectories based on key degradation feature points. By predicting four future key SOH states, it achieves high-precision reconstruction of capacity degradation trajectories and remaining useful life (RUL) prediction under arbitrary initial aging conditions. Specifically, this study constructs a deep hybrid neural network framework that integrates a 3-layer convolutional neural network (CNN), a 2-layer bidirectional long short-term memory network (BiLSTM), and a 5-layer fully connected network (FCN). This framework enables current SOH estimation and future key degradation point prediction based on segmented operational data. By combining the piecewise cubic Hermite interpolating polynomial (PCHIP) method, the reconstructed degradation trajectory adheres to physical principles. To further enhance the model’s cross-domain generalization capability, the study innovatively introduces deep domain adaptation (DDA) theory, enabling the method to adapt to different operating conditions and battery system applications effectively. The proposed approach was validated using data from 77 lithium iron phosphate (LFP) batteries under various cycling test conditions. The results demonstrate that the maximum error in SOH estimation does not exceed 3%, and the root mean square error (RMSE) for future key SOH point predictions remains below 6%.

Thermal management is critical for stationary battery energy storage systems (BESSs). This study proposes a prediction-based thermal management strategy for stationary BESSs, leveraging the operational predictability of BESSs to enable preemptive cooling regulation. Numerical investigations across four seasons demonstrate that, compared with the 60 s ON/60 s OFF intermittent control, the proposed prediction-based thermal management strategy reduces the overtemperature duration by approximately 80% under extreme summer conditions, lowers the peak battery temperature by 4.07 K, and eliminates temperature-difference exceedance. Continuous airflow modulation reduces the average of the seasonal maximum rates of temperature rise and drop by 38.6% and 75.1%, respectively, and decreases the standard deviation of temperature-difference change rate by 54.3%. Moreover, the proposed strategy achieves 67.1%-99.9% fan energy savings across all conditions. By preempting thermal responses and avoiding excessive cooling, the proposed strategy simultaneously enhances battery thermal safety and longevity without consuming excessive energy.

The construction industry faces an urgent need to develop sustainable alternatives to conventional energy-intensive materials such as concrete and ceramics. However, many existing alternatives are often constrained by insufficient mechanical strength, lack of recyclability, poor thermal insulation, or limited fire resistance. Drawing inspiration from historical building techniques that utilized plant fiber-mineral binder systems, this study overcomes their inherent limitations through advanced material design. We report a high-performance, low-carbon building bio-mineral composite fabricated under mild conditions by integrating a thermally induced hardening amorphous calcium carbonate mineral binder with waste wood sawdust. The synergistic interaction between these components endows the composite with outstanding mechanical properties, including a specific compressive strength of 36.87 MPa (g-1 cm-3) and a specific compressive modulus of 10.73 GPa (g-1 cm-3), substantially exceeding those of most conventional construction materials. Furthermore, the mineralized porous structure contributes to a UL-94 V-0 flame-retardant rating and a low thermal conductivity of 0.06 W (m-1 K-1). By bridging ancestral natural wisdom with modern materials science, this work not only delivers a high-performance material but also establishes a sustainable architectural pathway that "learns from yet surpasses antiquity," offering a practical and scalable solution for greening the construction industry.

Cellulose nanocrystal (CNC) photonic films have emerged as sustainable alternatives to traditional pigments by harnessing cholesteric nanostructures for vivid, fade-resistant colors. However, scalable fabrication of continuously functionalized structural color patterns directly from CNC dispersion remains a grand challenge due to limitations in conventional printing techniques and coassembly approaches. Here, we present a bioinspired sequential nanofluidic-assisted photonic patterning (SNAPP) technique that leverages preformed cholesteric CNC films with three-dimensional helical nanochannels as self-regulated pathways for spontaneous ink diffusion. Mimicking nature's helical Venturi effect, this technique enables universal integration of diverse functional components (molecules, polymers, and nanoparticles) to manipulate photonic band gaps or impart stimuli-responsive functionalities while preserving long-range chiral order. By combining mask-guided patterning with capillary-driven transport, we achieve full visible-spectrum structural colors, humidity/thermal-responsive patterns, and fluorescent carbon dot integration with 4-fold enhanced emission intensity and circularly polarized luminescence. The resulting films retain angle-dependent iridescence and polarization selectivity, while exhibiting exceptional environmental stability, biocompatibility, and degradability. This multifunctional platform enables dual-channel optical encryption with orthogonal authentication modes (structural color, fluorescence, thermochromism, and circular polarization), positioning it as a high-security anticounterfeiting solution for pharmaceutical applications. The SNAPP technique overcomes the fundamental limitations of traditional methods, offering a scalable, versatile route to functional photonic materials with programmable dynamic responses.

Battery fault diagnosis is a prerequisite for safety certification in electric vertical take-off and landing (eVTOL) aircraft. Unlike ground vehicles, eVTOL systems operate under a zero-failure tolerance regime, where even minor battery anomalies can compromise flight safety. However, existing diagnostic methods exhibit two fundamental limitations: their structural dependence on single-fault assumptions and limited adaptability to high-rate electro-thermal coupling under aviation duty cycles. To address these challenges, this study proposes a structurally decoupled, model-based diagnostic framework for battery modules. A cell-level electro-thermal coupling model is established to capture dynamic electrical–thermal interactions under high discharge rates. Based on structural analysis, four minimal structurally over-constrained subsystems are constructed to generate analytically decoupled residuals. This design enables systematic isolation of seven fault categories, including short circuits, interconnection faults, and multiple sensor failures, while preserving diagnosability under concurrent fault conditions. Kalman filtering enhances state estimation robustness, and an adaptive threshold strategy accommodates varying operational regimes. Under single-fault scenarios, the proposed method achieves a detection rate of 93.88%, enabling complete isolation and parameter estimation of all seven fault types. More critically, under concurrent-fault scenarios in a 10S3P module with 4,323 possible fault-pair combinations, 85.5% are uniquely identified and 98.3% are reduced to fewer than three candidates. These results demonstrate that structural residual design, rather than data-driven pattern recognition, is essential for achieving certifiable multi-fault diagnosability in next-generation eVTOL battery systems.

State of charge (SOC) is a critical parameter for the efficient and safe operation of lithium-ion batteries (LiBs). Currently, accurate SOC estimation heavily relies on battery open-circuit voltage (OCV) characteristics and its steepness relative to SOC. However, challenges arise when the electrical model is inaccurate or the OCV-SOC curve exhibits a flat region. To address these issues, mechanical properties have been widely integrated with traditional electrical characteristics to enhance SOC estimation accuracy. Nevertheless, the shape of the expansion force-SOC curve directly influences the estimation accuracy. This study provides a detailed analysis of the adverse effect of the plateau region and nonmonotonicity in the force-SOC curve on SOC estimation, and proposes a novel electrical-mechanical hybrid SOC estimation method. The online identified OCV is employed to determine the relative positions of the true SOC and estimated SOC, thereby guiding the direction of SOC feedback correction and the adjustment of the weighting coefficients for the electrical and mechanical estimators to achieve more accurate and robust SOC estimation. Experimental results of ternary and lithium iron phosphate batteries under dynamic operating conditions show that the adverse effects of the plateau region and nonmonotonicity in the force-SOC curve are effectively eliminated, and the maximum absolute error, root mean square error and mean absolute error are all below 2%.

With the evolution toward longer-life lithium-ion batteries, accelerated aging tests are critically required for life assessment. However, there has been no accelerated ageing test for high-capacity prismatic cells, whose increased dimensions induce challenges, such as pronounced thermal gradients, uneven current distributions, and heterogeneous mechanical stresses. To bridge the gap, this work adopted a framework combining acceleration factor analysis, half-cell aging mode assessment, ultrasonic scanning, electrolyte composition analysis, and SEM characterization to evaluate the acceleration effects and elucidate the multiscale degradation mechanisms of 280 Ah large-format prismatic cells under coupled thermal-electrical stresses. It has been found that the ambient temperature for accelerated aging tests can be increased from 55°C to 85°C, which can increase the time-based average acceleration factor from 8.21 to 22.36, peaking at 27.06, and consequently reduce the testing duration to roughly 3.7