Braking control in electric vehicles is crucial for enhancing energy efficiency through regenerative braking. However, conventional anti-lock braking systems (ABS) typically deactivate regenerative braking to prioritize stability, sacrificing energy recovery potential. This paper presents a novel hierarchical control strategy for dual-motor electric vehicles that enables cooperative regenerative and hydraulic braking throughout the ABS operating cycle. The strategy employs a Proportional-Derivative (PD) controller, optimized via Particle Swarm Optimization (PSO), for precise slip ratio regulation, coupled with a rule-based torque distribution mechanism. Simulations under diverse road conditions demonstrate that the proposed method maintains robust braking performance and stability. Crucially, it achieves a significant energy recovery of 0.26

External short circuits (ESCs) in lithium-ion batteries are often abrupt and difficult to predict in advance; therefore, rapid suppression at the early stage of ESC abuse is essential for preventing its escalation into thermal runaway and mitigate the resulting hazards. Low-pressure water mist provides a promising approach for cooling and heat-removal capability; however, its early-stage suppression behavior and parameter-dependent mechanisms under ESC conditions remain insufficiently understood. In this study, the effects of water mist flow rate, spray application surface, and intervention timing on the early-stage suppression of external-short-circuit-induced thermal runaway were experimentally investigated. The suppression behavior was further analyzed by considering the coupling between water mist deposition, evaporative cooling, and battery heat-accumulation regions. The results show that the suppression performance of water mist strongly depends on the matching between atomization conditions and the targeted battery surface. Under appropriate flow-rate and spray-surface configurations, water mist effectively prevented open flame and severe casing damage, and remained effective during the high-temperature pre-runaway stage. Propagation tests further demonstrated that water mist can delay or inhibit cell-to-cell thermal runaway propagation. These findings provide a practical basis for rapid early-stage mitigation of ESC-induced thermal runaway in lithium-ion batteries.

The widespread adoption of lithium-ion battery-powered electric vehicles has raised increasing concerns regarding battery safety under mechanical abuse conditions. However, mechanical abuse scenarios, such as battery extrusion, are highly diverse, making it impractical to conduct extensive destructive experiments and independent modeling for each specific scenario. In this work, a cross-scenario mechanical safety modeling framework for lithium-ion batteries is proposed based on transfer learning. Three quasi-static mechanical abuse tests, including flat-plate, rigid-rod, and hemispherical compression, are conducted on 18650 lithium-ion batteries. An equivalent mechanical model with a spring-damper parallel structure is employed to characterize the mechanical response and generate simulation data. Based on data from a single mechanical abuse scenario, a backpropagation neural network (BPNN)-based safety model is established to predict the maximum stress in the battery. The learned knowledge is then transferred to other mechanical abuse scenarios using a transfer learning strategy. The results demonstrate that, under limited target-domain data, the transferred models achieve stable prediction performance, with the average relative error controlled within 3.6%, outperforming models trained from scratch under the same conditions. Compared with existing studies that focus on single-scenario modeling, this work explicitly investigates cross-scenario transferability and demonstrates the effectiveness of transfer learning in reducing experimental and modeling effort for battery mechanical safety analysis.

During fast charging of lithium-ion batteries in electric vehicles, lithium plating severely degrades performance (e.g., capacity fade, impedance rise) and exacerbates safety risks (e.g., thermal runaway). To address this, we propose a voltage-driven adaptive charging strategy that dynamically regulates current profiles by leveraging the intrinsic correlation between terminal voltage and lithium-plating overpotential. This correlation is first quantified via an electrochemical-thermal-aging (ETA) coupling model. Our analysis reveals that polarization characteristics and plating susceptibility exhibit distinct voltage-range-dependent signatures (independent of ambient temperature or charging current), enabling plating suppression via real-time voltage feedback. Numerical analysis of the current-profile parameters quantitatively elucidated their influence on charging efficiency, guiding the optimization of voltage-threshold-based current regulation. Parameter sweeping confirms that the current profile consistently outperforms conventional constant-current constant-voltage (CC-CV) charging, as well as more advanced MCC and pulse charging strategies, across the entire parameter space. Experimental cycling tests demonstrate that, compared to CC-CV charging, the proposed strategy reduces the capacity loss caused by lithium plating by half and capacity retention rate increased by maximum of 6.8% while maintaining comparable charging time. Critically, this approach eliminates reliance on high-precision onboard state estimators or three-electrode modifications, making it compatible with existing battery management systems (BMS) in electric vehicles and offering a practical pathway to safer fast charging. A full-range mechanistic correlation between terminal voltage and lithium-plating overpotential is established using a validated ETA model;A model-free voltage-driven fast-charging strategy is developed, requiring no reference electrodes or cell modification;Verified fast charging strategy on lithium-ion battery in vehicle soutperforms CC-CV charging across the entire parameter space;37% lower polarization and 50% less plating-induced degradation without sacrificing charging efficiency.

Accurate evaluation of fault severity and reliable estimation of the battery internal state are crucial for battery safety management in electric vehicles. While existing studies mainly focus on fault detection approaches, the assessment of fault severity following short-circuit occurrence remains insufficiently addressed. To address this limitation, this study proposes a novel integrated EKF-LSTM framework for concurrent estimation of short-circuit resistance and internal temperature, and for quantitative evaluation of short-circuit fault severity. This method performs real-time estimation of the battery SOC and short-circuit resistance. Subsequently, the internal temperature is estimated using the EKF algorithm, and an LSTM network is employed to predict short-term thermal evolution and assess fault severity. The experimental results demonstrate that the proposed method can accurately estimate short-circuit resistance and internal temperature after fault occurrence, and effectively evaluate the severity level of short-circuit faults, while the LSTM-based severity assessment achieves an accuracy of 98.2%. These results indicate that the proposed framework provides an effective post-fault safety assessment solution for short-circuit scenarios, which is of great significance for battery safety management.

Internal short circuits (ISCs) are among the most critical faults in lithium-ion batteries, as they can evolve silently and eventually trigger severe safety incidents. Existing studies primarily focus on ISC detection, while insufficient attention has been paid to distinguishing ISCs from other non-ISC abnormalities. This paper proposes a two-stage diagnostic framework based on kernel density estimation (KDE) and long short-term memory (LSTM) network. It is found that, in the KDE-transformed feature space, representative statistical characteristics can be effectively extracted, enabling not only reliable detection of ISCs but also discrimination from other non-ISC abnormalities. In the first stage, voltage distribution deviations are quantified using a characteristic distance metric for anomaly screening. In the second stage, multiple statistical features extracted from the KDE distribution are incorporated into the LSTM network to confirm ISC faults and distinguish them from other kinds of faults. Experimental validation based on nail penetration demonstrates that the proposed method achieves an overall detection accuracy exceeding 93%, while effectively differentiating ISC faults from excessive aging and voltage sensor faults. The proposed method enhances the reliability of ISC detection and provides an effective solution for battery safety monitoring in electric vehicles and energy storage systems.

Accurate diagnosis of micro short-circuits (MSCs) is essential for ensuring the safety of lithium-ion batteries used in electric vertical take-off and landing (eVTOL) aircraft. Unlike conventional electric vehicles, eVTOL batteries normally operate under high-rate discharge conditions, where strong polarization and rapid voltage variations are apt to mask the weak signatures of MSCs. To address this challenge, this study proposes an MSC diagnosis method based on multiscale voltage residual analysis. A battery model is first established to characterize the normal response under high-rate discharge, and the discrepancy between the measured and estimated terminal voltages is used to construct the model residual. Features describing the overall voltage evolution, residual statistical distribution, and multiscale residual fluctuations are then extracted. Specifically, the shadow region integral area and voltage–capacity slope are used to characterize the global voltage trajectory, while the residual mean and kurtosis quantify the systematic deviation and non-Gaussian fluctuation of the residual. Wavelet decomposition is further applied to capture the low- and high-frequency residual characteristics. After feature reduction, eight representative features are retained to establish the diagnostic model. Experimental validation under high-rate discharge conditions demonstrates that the proposed method can effectively identify MSCs despite interference from abnormal aging, thereby reducing the false alarms caused by feature similarity. This study provides a reliable approach for micro short-circuit diagnosis of eVTOL lithium-ion batteries under strong polarization and highly dynamic operating conditions.

Micro short circuits (MSCs) are minor faults within lithium-ion batteries and represent an early stage in the progression toward safety failures and thermal runaway. Early detection of MSCs is crucial for enhancing battery safety. However, this remains technically challenging, as the voltage anomalies caused by MSCs can easily be obscured by complex and frequent charging/discharging cycles. In this study, a two-step diagnostic approach is proposed for detecting MSC faults. Firstly, internal resistances are estimated using the forgetting factor recursive least squares (FFRLS) method, and suspected MSC cells are identified based on the Hellinger distance calculated between each cell and a reference value. Secondly, the inverse Markov method is applied to analyze voltage transfer anomalies in the suspected cells, thereby confirming the faults. To support algorithm validation, MSCs are first induced through slight extrusion and the faulty cells are then operated in series with normal cells under the UDDS cycle, thereby establishing a labeled database containing both normal and MSC data segments. The effectiveness of the proposed approach is validated using this dataset, and experimental results demonstrate that the method can effectively identify MSC faults, achieving a precision of over 98.0%.

Abstract Traditional micro-short circuit diagnosis methods face limitations such as susceptibility to noise interference and dependence on representative fault samples. To overcome these challenges in when locating micro-short circuit in series-connected lithium-ion battery packs, this study proposes a micro-short circuit localization algorithm that combines real-time performance with a low false alarm rate. The algorithm follows a three-stage process: First, voltage sequence data are processed using a sliding window (window size = 100, stride = 10, sampling period = 1 second) to calculate the average voltage baseline (after outlier removal) and derive the voltage deviation sequence. Next, four core features are extracted and normalized: average negative deviation rate, cumulative negative deviation integral, maximum negative deviation value, and negative deviation duration. Second, an Isolation Forest model is trained using 60 normal samples from the first 10 windows. An anomaly score (ranging from 0-1) quantifying the probability of battery abnormality, is generated by calculating the average path length of feature observations within the forest. Finally, a continuous scoring validation mechanism is employed: a fault is confirmed present when the battery’s anomaly score exceeds the 0.65 threshold for 20 consecutive iterations. Experimental and simulation data were collected from six 120Ah series-connected battery packs operating under the Urban Driving Development Standard (UDDS) cycle, with micro-short circuit simulated by connecting external resistors of 5Ω, 15Ω, and 20Ω. These data validated the algorithm’s effectiveness. The algorithm successfully captures the voltage characteristics of micro-short circuit batteries, mitigates interference from battery inconsistencies and data noise, and accurately locates faulty cells. These results demonstrate the proposed algorithm’s capability to precisely identify micro-short circuit faults within series-connected lithium-ion battery packs, providing robust technical support for battery safety management.

Abstract To address the challenges of inefficient tuning and opacity in XGBoost-based battery temperature prediction, this work introduces a framework integrating Bayesian Optimization (BO) and SHapley Additive exPlanations (SHAP). The resulting model is highly accurate, achieving a Mean Absolute Error (MAE) under 0.001°C on the test set; efficient, reducing the Root Mean Square Error (RMSE) by ∼12% over baseline while saving ∼45% of tuning time versus grid search; and interpretable, identifying physically-consistent key features such as coolant inlet temperature. This framework serves as a robust paradigm for developing trustworthy predictive models for complex industrial systems.

OBJECTIVE:The DNA methylation urine test, a noninvasive early detection method for upper tract urothelial carcinoma (UTUC), is currently in full swing. This study aimed to systematically assess its diagnostic performance on UTUC. MATERIALS AND METHODS:PubMed, Scopus, Embase, and Cochrane were our main databases when searching articles published from January 2000 to December 2023. Sensitivity and specificity were study primary endpoints. I2 was used to evaluated heterogeneity, meanwhile subgroup and meta-regression analyses were adopted to investigated the source of heterogeneity. Sensitivity analysis was performed to evaluate the result robustness, while Deeks' funnel plot asymmetry test was for the publication bias. RESULTS:Nine studies with 1326 patients were included. The pooled sensitivity was 0.89 (95% CI: 0.83-0.93) and specificity were 0.91 (95% CI: 0.83-0.96). The area under the receiver operating characteristic curve was 0.96 (95% CI: 0.93-0.97). Substantial heterogeneity was found during the data synthesis, whereas the pooled results remained robust in the sensitivity analysis. None of the potential covariates-urine sample collection method, population, country, methylation test method, or tumor grade-could account for the heterogeneity. CONCLUSION:DNA methylation urine test is a promising method with high efficiency for UTUC early detection. Nevertheless, owing to the significant heterogeneity, more well-organized studies are warranted to further explore its diagnostic efficiency and application context.

Temperature is a critical indicator for safety monitoring of lithium-ion batteries. However, due to uncertain thermal diffusion from the cell interior to the surface, surface measurements cannot accurately or promptly reflect internal states. To address this limitation, this study proposes a novel internal temperature estimation method applicable under both normal operation and short-circuit fault conditions. A synergistic framework is developed based on the online sequential extreme learning machine, enabling simultaneous internal temperature estimation and fault diagnosis. An innovative experimental setup was established using controlled external short circuits (ESC) and shallow, slow nail penetration to trigger internal short circuits (ISC). Experimental results demonstrate that the proposed method achieves high accuracy, with maximum errors of 1.0963 degrees C, 3.0876 degrees C, and 2.2119 degrees C under normal, ESC, and ISC conditions, respectively. Moreover, the method successfully distinguishes between ESC and ISC based on distinct internal temperature dynamics, confirming its capability for reliable fault classification. These results highlight the method's promise for real-time fault detection and safety monitoring in lithium-ion battery systems.

In the health monitoring and safety assessments of concrete structures, ultrasonic non-destructive testing (NDT) technology has become an indispensable tool due to its non-destructive nature, efficiency, and precision. However, when used in inspecting irregular concrete surfaces, traditional planar ultrasonic transducers often encounter energy loss and signal attenuation induced by poor interface coupling, which significantly reduces the accuracy and reliability of the test results. To address this problem, this article proposes a point-contact dry coupling ultrasonic transducer solution, which enables efficient acquisition of ultrasonic signals within concrete without the need for couplants. By combining an array imaging system with a total focusing algorithm, this study not only significantly enhances the convenience and signal-to-noise ratio (SNR) of concrete ultrasonic imaging, but also opens new pathways for ultrasonic NDT technology in concrete.

Understanding lithium-ion battery failure under mechanical abuse is critical for safety. While continuous compression is studied, the effects of intermittent loading and its interaction with State of Charge (SOC) are less understood. This study experimentally investigates the mechano-electrochemo-thermal failure of 18650 NMC cells under continuous versus intermittent compression across various SOCs (20 %-80 %), monitoring mechanical, electrical, and thermal responses. Results revealed complex, non-linear failure mechanisms. High SOC consistently increased failure susceptibility (earlier failure, lower deformation tolerance). Intermittent compression proved significantly more hazardous than continuous loading, particularly at high SOC, accelerating internal damage and causing earlier, more severe thermal runaway, often initiated by identifiable soft short circuits (multi-stage voltage drops). Crucially, stiffness exhibited complex, mode-dependent behavior: nonmonotonic under continuous compression (peaking at SOC 20 %) versus a generally monotonic increase under intermittent compression, highlighting loading pattern influence. The anode was the primary heat source during thermal runaway, and anomalous behavior was noted at SOC 60 %. These findings highlight the limitations of standard continuous crush tests and the need to incorporate intermittent loading in safety assessments. The complex, mode- and SOC-dependent mechanical responses provide critical data for robust battery pack design (considering stiffness variations and fatigue) and inform Battery Management System (BMS) failure precursor detection.

Precisely estimating the remaining mileage of electric vehicles is highly important for vehicle control and battery recharging determinations. Remaining mileage estimation (RME) is a technique difficulty in practice since it is impacted by many factors, including the battery state of charge (SOC), state of health (SOH), ambient temperature, and traffic condition, etc. In this study, an online RME method is proposed based on dual extended Kalman filter (DEKF) and extreme gradient boosting (XGB) algorithms. Firstly, the battery SOC and SOH are co-estimated based on DEKF with considering the impacts of ambient temperature. Secondly, the current traffic condition are analyzed by using a historical data segement, and then the energy consumpation rate is predicted by XGB algorithm. The XGB algorithm's accuracy under the varying length of data segment is analyzed for determining the proper algorithm parameters. The presented method is evaluated by a simulation study. The results under several typical driving cycles indicate that the precise RME can be achieved with the maximum error less than 1.2%. The method is expected to be useful in providing credible mileage estimation in electric vehiecle applications.

Short circuits are a prevalent fault in lithium-ion battery applications, leading to severe safety consequences; therefore, the rapid diagnosis is imperative for improving battery safety. External and internal short circuits exhibit similar characteristics but have significantly different evolution results. This study introduces a diagnostic approach based on Discrete Wavelet Transform (DWT) that promptly identifies and categorizes short circuits using only three seconds of post-occurrence data. The diagnostic method encompasses a wide range of short-circuit resistances, from 10 mS2 to 10 S2. Subsequently, a fault severity assessment method based on empirical formulas is established. For online applications, an enhanced extreme learning machine is utilized to estimate short-circuit resistance and to predict the potential peak temperature, failure time, and remaining capacity. The proposed method is evaluated through experiments, including internal short circuits, external short circuits, sensor faults, and normal battery conditions. The tests demonstrate that the method can achieve a diagnosis accuracy of 95.7 % while the estimation error for short-circuit resistance is less than 3.9 %.

The immunosuppressive tumor microenvironment (TME) plays a crucial role in the progression and treatment resistance of melanoma. Modulating the TME is thus a key strategy for enhancing therapeutic outcomes. Previousstudies have identified clonidine (CLD), an α2-adrenergic receptor agonist, as a promising agent that enhances T lymphocyte infiltration and reduces myeloid-derived suppressor cells within the TME, thereby promoting antitumor immune responses. In this study, we discovered that CLD reshaped the melanoma immune microenvironment, facilitating T-cell activation and exerting antitumor effects. However, the high doses of CLD required for effective TME modulation pose significant toxicity concerns, limiting its clinical applicability. To address this, we employed the controllable cavitation-on-a-chip (CCC) platform to formulate CLD-loaded liposomes and optimize their size. This approach aimed to enhance the precision and efficacy of drug delivery while reducing systemic side effects. Our results demonstrated that size-specific CLD liposomes, particularly those at 50 nm, significantly improved tumor growth inhibition and immune cell infiltration within the TME. Moreover, these optimized liposomes mitigate adverse effects associated with high-dose CLD treatment. This study indicates the potential of CCC-optimized CLD liposomes as a safer and more effective melanoma therapy, highlighting the critical interplay between liposome size control and therapeutic outcomes in cancer treatment.

The improvement of battery management systems (BMSs) requires the incorporation of advanced battery status detection technologies to facilitate early warnings of abnormal conditions. In this study, acoustic data from batteries under two discharge rates, 0.5 C and 3 C, were collected using a specially designed battery acoustic test system. By analyzing selected acoustic parameters in the time domain, the acoustic signals exhibited noticeable differences with the change in discharge current, highlighting the potential of acoustic signals for current anomaly detection. In the frequency domain analysis, distinct variations in the frequency domain parameters of the acoustic response signal were observed at different discharge currents. The identification of acoustic characteristic parameters demonstrates a robust capability to detect short-term high-current discharges, which reflects the sensitivity of the battery’s internal structure to varying operational stresses. Acoustic emission (AE) technology, coupled with electrode measurements, effectively tracks unusually high discharge currents. The acoustic signals show a clear correlation with discharge currents, indicating that selecting key acoustic parameters can reveal the battery structure’s response to high currents. This approach could serve as a crucial diagnostic tool for identifying battery abnormalities.

Optical imaging and phototherapy in deep tissues face notable challenges due to light scattering. We use encoded acoustic holograms to generate three-dimensional acoustic fields within the target medium, enabling instantaneous and robust modulation of the volumetric refractive index, thereby noninvasively controlling the trajectory of light. Through this approach, we achieved a remarkable 24.3% increase in tissue heating rate in vitro photothermal effect tests on porcine skin. In vivo photoacoustic imaging of mouse brain vasculature exhibits an improved signal-to-noise ratio through the intact scalp and skull. These findings demonstrate that our strategy can effectively suppress light scattering in complex biological tissues by inducing low-angle scattering, achieving an effective depth reaching the millimeter scale. The versatility of this strategy extends its potential applications to neuroscience, lithography, and additive manufacturing.