Search Results (26,539)

Dynamic membrane bioreactors (DMBRs) are promising systems for continuous biohydrogen production because they enable effective biomass retention under short hydraulic retention time (HRT) conditions. In this study, a dynamic membrane bioreactor coupled with electro-fermentation (DMBR-EF) was operated for 59 days to investigate the effect of applied voltage on biohydrogen production and metabolic flux regulation. The reactor was sequentially operated at 0 (no applied voltage), 0.2, 0.4, 0.6, 0.8, and 1.0 V using glucose as a model substrate. The highest hydrogen production rate (HPR) and hydrogen yield (HY) were achieved at 0.2 V, reaching 15.35 ± 0.48 L H₂/L/d and 1.54 ± 0.05 mol H₂/mol glucose_added, respectively, which were 33.71% and 33.91% higher than those of the 0 V control. At 0.2 V, residual glucose and effluent volatile suspended solids (VSS) were minimized, while butyric acid (HBu) formation was enhanced and lactic acid (HLa) accumulation was suppressed. In contrast, voltages above 0.4 V reduced hydrogen recovery by shifting metabolic flux toward HLa, propionic acid (HPr), formic acid (HFo), and homoacetogenic pathways. Microbial analysis showed that Clostridium dominated under all conditions, but voltage application selectively altered the relative abundance and metabolic output of Clostridium-related amplicon sequence variants (ASVs). These results indicate that mild electrochemical stimulation at 0.2 V effectively enhances continuous biohydrogen production by promoting butyric acid-type fermentation, suppressing lactic acid accumulation, and reducing hydrogen loss through competing metabolic pathways in DMBR-EF systems. Full article

Ultra-low-temperature (ULT) refrigeration systems are widely required in applications such as biomedical storage, cryogenic processing, and advanced scientific facilities, where both exergy efficiency and economic performance are critical. In this study, a comprehensive analysis and optimization of the exergy and exergoeconomic performance evaluation of a triple-stage cascade refrigeration system under ULT refrigeration is presented using ethylene (R1150), ethane (R170), propylene (R1270), difluoroethane (R152a), propane (R290), and fluoroethane (R161). A detailed exergy/exergoeconomic analysis, along with an optimization procedure, is performed at both the overall system level and the component-wise level in order to determine the key performance indicators. Minimizing the total cost product rate for each evaporator and condenser temperature is achieved by optimizing the condensing temperatures of the low-temperature cycle and the medium-temperature cycle. The component-wise analysis reveals that major thermodynamic irreversibilities occur in the HTC compressor and throttling valves, with maximum relative exergy destruction reaching 24.71% in TV-3 and the highest exergy destruction ratio reaching 14.92% in the HTC compressor. From an exergoeconomic perspective, the evaporator exhibits the largest combined exergy destruction and capital investment, and operational and maintenance cost rate (up to 24.73 $/h), while the condenser shows the highest exergoeconomic factor (up to 14.24%). The system-level results show that R1150/R170/R152a has the best exergetic and exergoeconomic performance compared to the other refrigeration combinations within the temperature ranges. Compared with R1150/R170/R290, this combination increases exergy efficiency by up to 5.05% for evaporator temperature variations and up to 7.84% for condenser temperature variations while reducing exergy destruction by up to 7.50% and 11.14%, respectively. Furthermore, the same combination decreases the total exergy destruction cost rate and product cost rate by up to 11.15% and 7.88%, respectively. In contrast, R1150/R170/R290 generally exhibits the poorest overall performance. The findings demonstrate that the refrigerant choice is crucial in achieving better exergetic and exergoeconomic performance for ULTs under various evaporator and condenser conditions. Full article

Smartphone time-to-empty (TTE) estimation is vulnerable to residual battery-drain effects after short-duration user activities, which can cause runtime predictions to fluctuate. This paper proposes a continuous-time battery-drain model that separates instantaneous and delayed drain effects in operating-system logs. The discharge process is decomposed into baseline drain, direct drain induced by the current usage state, and delayed tail drain triggered by transient events. Multi-exponential recursive states are used to represent network communication, location requests, background wake-ups, and related events as residual drain terms that decay over time. Non-negative sparse estimation is adopted to preserve the interpretability of the individual contributions. Experiments are conducted on a public smartphone-log dataset with strict separation between the training and test sets. The results show that the proposed method achieves stable TTE prediction performance on the test set, with short-horizon $R^2$ values above 0.963 and scenario-level MAPE of 2.42–5.03%. These findings indicate that explicit modeling of delayed residual drain can improve the accuracy and stability of smartphone remaining-runtime prediction. Full article

Road sign detection is crucial for highway maintenance but often suffers from sign loss, occlusion, and spatial misjudgments such as repeated local detections or mapping errors. To address these issues, this study proposes YOLO-DeepSort, a tracking and detection framework integrating a novel Spatial Multivariate Clustering Algorithm with GPS information (SCMCg). The YOLOv9 detector is augmented using Mixed Local Channel Attention (MLCA) and DualConv modules to enhance image feature extraction and contextual awareness while compressing the theoretical model volume. In the tracking phase, DeepSort combined with SCMCg employs Delaunay triangulation and hierarchical GPS constraints to refine spatial clustering and data association. Validation was conducted on a mixed dataset comprising the CCTSDB and self-collected images from a Ningxia national road. Experimental results indicate that the proposed model operates efficiently at 21.4 M parameters, achieving a precision of 97.8%, a mean Average Precision (mAP) of 91.2%, and a tracking Success Rate of 97.3%. Compared to the baseline YOLOv8-DeepSort, absolute improvements of 4.60% in precision and 4.20% in mAP were observed. The integrated framework effectively mitigates occlusion and tracking spatial errors, providing a robust and lightweight methodology for the automated condition assessment of intelligent transportation infrastructure. Full article

The newly constructed Luofengshan Tunnel obliquely underpasses the operating Hanyicun railway tunnel, requiring an integrated assessment of deformation response and service safety. This study combined field monitoring, ultrasonic shear-wave reflection imaging, and three-dimensional numerical simulation to evaluate track-bed settlement, lining deformation, floor integrity, lining stress, and structural safety. The complete third-party monitoring record showed that vertical deformation dominated the tunnel response. The final maximum cumulative values were −17.44 mm for manual lining settlement, −14.91 mm for automated lining settlement, and −11.30 mm for track-bed settlement; the corresponding maximum absolute values of horizontal displacement and convergence were 1.82 mm and 1.40 mm, respectively. Same-period comparisons showed that deformation increments decreased after breakthrough, and post-lining flood-season monitoring showed monthly rates below 1 mm/month. Ultrasonic imaging detected no evident voids or collapse within 2.0–2.2 m below the floor. Numerical simulation showed that incorporating measured initial deformation increased the maximum compressive stress from 4.95 MPa to 6.26 MPa, whereas subsequent underpass excavation and train loading increased it slightly to 6.28 MPa and 6.33 MPa, respectively. The final minimum safety factor was 3.18, indicating that the underpass impact remained controllable. Full article

Accurate state-of-charge (SOC) estimation is essential for reliable operation of vanadium redox flow batteries (VRFBs), yet many model-based methods treat the open-circuit-voltage (OCV)-SOC relationship as a fixed calibration curve. This study experimentally investigates condition-dependent SOC-OCV behavior using a laboratory-scale VRFB equipped with a bypass OCV cell. The bypass OCV method was validated against an intermittent discharge–rest method, with OCV differences below 5 mV. SOC-OCV characteristics were then examined under different electrolyte flow rates, cycling histories, electrolyte/component refreshing conditions, and a dynamic stress test profile. The results show that the SOC-OCV curve varies with cycling history and flow rate, while electrolyte/component refreshing and dynamic operation further modify the measured OCV response. Fixed-curve-based SOC inversion confirms that calibration mismatch can introduce substantial SOC estimation errors, with a case-specific maximum error of 11.36% observed when the initial post-preconditioning constant-current curve was applied to the cell after 50 cycles under the tested 96 mL min⁻¹ DST condition. These findings highlight the need for adaptive OCV correction and condition-dependent SOC-OCV mapping in practical VRFB SOC estimation. Full article

The aim of this study was to evaluate the effect of Sida hermaphrodita harvest timing on biomass composition, properties, and methane fermentation performance. In addition, an energy–economic assessment was performed for biomass obtained at different stages of the growing season. The economic assessment assumed CHP electrical and thermal efficiencies of 38% and 47%, electricity and heat prices of 0.18 and 0.05 EUR/kWh, respectively, month-specific agrotechnical costs, and OPEX equal to 30% of total energy revenue. The biomass exhibited clear seasonal changes, transitioning from a material with high bioavailability during the summer period to a structurally more recalcitrant substrate in the autumn and winter months, as indicated by increasing lignification and fibrous fraction contents. The highest CH₄ production yields, ranging from 300 to 320 mL/g VS, and maximum production rates of up to 33.5 mL/g VS·d were obtained between June and August. In December, the CH₄ yield decreased to 180 ± 9 mL/g VS, accompanied by a substantial deterioration in kinetic performance. Despite the relatively stable theoretical methane potential, which ranged from 405 to 430 mL/g VS, its conversion efficiency declined from 77.1% in the summer period to 41.9% in the winter period. Regression analysis confirmed the key influence of the C/N ratio and total solids content, with model fits reaching R² values of 0.74–0.80, while the structure of lignocellulosic complexes had a less pronounced but still relevant effect. The maximum CH₄ production per unit cultivation area, approaching 3380 m³/ha, was achieved in July–August, reflecting a balance between high specific methane yield and biomass productivity. At the same time, the results demonstrated that the maximum biomass yield did not translate into the highest energy and economic performance. The highest net economic return, 1789 ± 330 EUR/ha, was obtained in July, despite biomass yield being 13.6% higher in September. These findings indicate a seasonal decoupling between biomass yield and energy performance, highlight biomass quality as a critical determinant of anaerobic digestion efficiency, and support harvest-date optimization as a low-cost strategy for the practical use of S. hermaphrodita in agricultural biogas plants. Further long-term continuous and semi-continuous studies are required to validate process stability and performance under industrial operating conditions. Full article

This work presents a new Battery Thermal Management System (BTMS) concept utilizing a metal foam manifold cold plate (MFMCP), developed specifically to meet the rising thermal dissipation needs of lithium-ion batteries. By coupling a manifold flow design and high-conductivity metal foams (characterized via SEM), the system significantly enhanced heat transfer and temperature uniformity. A Local Thermal Equilibrium (LTE) model evaluated thermal–hydraulic performance across varying Reynolds numbers (40–200), foam structures (PPI and porosity), and Al₂O₃ nanofluid concentrations. Results indicated that an optimal foam structure (95 PPI, porosity of 0.905) yielded a significant surface temperature reduction of 15 K. Although the Al₂O₃ nanofluid provided an additional cooling effect of 0.3 K, its pressure penalty lowered the performance evaluation criterion (PEC); the highest initial PEC of 2.57 was established when employing the pure base fluid without any functional additives. A Multi-Layer Perceptron (MLP) model was developed to expedite design validation, achieving high predictive accuracy as evidenced by an R² of 0.983 and an MRE of 1.8%. Subsequent optimization via a genetic algorithm (GA) further increased the maximum PEC by 12% to 2.88. Equal pumping power system simulations indicated that the MFMCP design achieved a peak cell temperature of 304.35 K, outperforming traditional designs which peaked at 308.33 K. Furthermore, transient tests at 1C to 2C discharge rates confirmed consistent cooling improvements, demonstrating the MFMCP’s promising potential for dynamic operational conditions. Full article

The leakage and dispersion behavior of CO₂ in the ground injection stage of oilfield CCUS systems is highly complex under high-pressure, low-temperature, and continuous-operation conditions, while quantitative criteria for emergency control distances remain insufficient. To address this issue, a liquid CO₂ storage tank in an oilfield CO₂ cyclic injection station was selected as the research object. A 1:1 two-dimensional and three-dimensional computational model was established. A two-stage simulation framework was developed in this study, where PHAST was first employed for rapid multi-scenario consequence screening, followed by three-dimensional CFD validation using KFX under representative high-risk scenarios. The proposed framework was used to systematically investigate CO₂ leakage and dispersion characteristics under various leakage aperture sizes, ambient temperatures, and wind speed conditions. Risk zoning and risk quantification were further conducted based on high-concentration CO₂ toxicity thresholds of 10%, 15%, 20%, and 30%. The results show that the leak aperture is the dominant factor controlling dispersion consequences. As the aperture increased from 5 mm to 50 mm, the maximum dispersion distance increased from 8.8–9.4 m to 132–146 m, and the maximum cloud footprint increased from 4.8–5.4 m² to 2204.1–2599.6 m². Higher temperature enhanced CO₂ dispersion capacity, whereas wind speed exerted a dual effect involving near-field dilution and downwind transport. The KFX three-dimensional simulation indicated that buildings and equipment arrangements induced near-ground flow separation and local accumulation. Under a 25 mm leak and an average wind speed of 4.6 m·s⁻¹, the cloud approached full coverage of the injection station within approximately 300 s and tended to spread beyond the site boundary. Liquid CO₂ leakage was accompanied by flashing, dry-ice formation, and sublimation, leading to a leakage-rate pattern characterized by an initial decrease followed by an increase and gradual stabilization. This indicates that emergency response should account for the ice-plugging and de-plugging process. The risk zoning results show that the downwind region can be divided into fatal, severe-injury, minor-injury, adverse-reaction, and relatively safe zones. A risk level on the order of 10⁻⁶ still existed at 100 m from the leakage source. It is recommended that CO₂ concentration monitoring and alarm systems be deployed within 100 m of the station, warning signs be placed near the maximum impact boundary of 146 m, and 300 s be used as a critical time window for coordinated evacuation between the plant area and surrounding residential areas. The findings provide a technical basis for safety-distance verification, monitoring layout design, and emergency-plan development for ground injection systems in oilfield CCUS projects. Full article

Trajectory prediction is a critical component of autonomous driving systems, enabling vehicles to anticipate future motion behaviors and perform safe decision-making in dynamic traffic environments. While recent trajectory forecasting methods achieve state-of-the-art prediction accuracy, many operate as black-box systems and are evaluated primarily on high-end computing platforms, limiting their interpretability and practical deployment feasibility in resource-constrained autonomous driving systems. To address these limitations, this work proposes an interpretable and edge-deployable spatio-temporal trajectory prediction framework for autonomous driving. The proposed architecture integrates a Temporal Convolutional Network with Multi-Head Self-Attention (TCN–MHSA) in ActorNet for selective temporal modeling, a Lane Graph Attention Network (LaneGAT) for structured spatial reasoning, and a multi-stage FusionNet for actor–lane interaction. To improve model interpretability, a comprehensive Explainable AI (XAI) evaluation framework is introduced, including temporal sensitivity analysis, interaction-aware perturbation studies, spatial influence analysis, and gradient-based feature attribution methods. These analyses provide insights into how the model captures temporal motion dependencies, neighboring vehicle interactions, and environmental context during trajectory prediction. To improve the robustness of the interpretability analysis, temporal sensitivity was additionally evaluated over 100 validation scenes, demonstrating that recent observations consistently exert the greatest influence on trajectory prediction, while neighboring interaction effects gradually diminish with increasing spatial separation. Furthermore, practical real-world deployment feasibility is investigated on the NVIDIA Jetson Xavier NX platform using edge-aware optimization strategies, including mixed-precision inference and graph-complexity reduction techniques for efficient resource-constrained inference, achieving 125.74 ms latency at 12.86 W. Additional edge deployment comparisons with HiVT and SIMPL approaches under identical hardware conditions demonstrate that the proposed framework provides a more favorable balance between computational efficiency and embedded deployment performance. Experimental evaluation on the Argoverse 1 dataset demonstrates a minimum Average Displacement Error (minADE) of 0.90 m, a minimum Final Displacement Error (minFDE) of 1.50 m, a Miss Rate (MR) of 0.19, and DAC = 0.95, while establishing an accuracy–deployability operating point under embedded hardware constraints with low power consumption and practical inference throughput. Full article

Highway–rail grade crossing (HRGC) incidents in the United States declined substantially for several decades before stabilizing in recent years. Understanding this persistence is important because future safety improvements may depend on identifying locations where incident occurrence remains resistant to further reduction. This study developed an integrated framework to characterize persistent HRGC incident environments using 50 years (1976–2025) of Federal Railroad Administration incident records. Trend, structural-break, variance, and stationarity tests were first applied to determine whether the historical decline transitioned into a distinct persistence regime. A county-level persistence index (PI) was then developed to quantify the combined effects of incident burden and resistance to decline during the plateau period. Distributional analysis characterized the statistical behavior of the PI, while global and local Moran’s I statistics evaluated its spatial organization. Explainable machine learning methods were subsequently used to identify incident characteristics associated with elevated persistence. The results identified a statistically significant regime change around 2010. Prior to 2010, incidents exhibited a strong declining trend, whereas the subsequent period displayed a statistically significant but substantially weaker decline, lower variance, and behavior consistent with a persistence regime characterized by a markedly attenuated rate of improvement. The PI followed a strongly right-skewed distribution that was best represented by a bounded heavy-tailed unit log-logistic model, indicating that persistence is concentrated within a relatively small subset of counties. Spatial analysis revealed significant positive spatial autocorrelation (Moran’s I = 0.180, p = 0.001) and geographically coherent clusters concentrated primarily in the southeastern United States and several major freight-oriented regions. Explainable machine learning models identified train-operating characteristics, warning device contexts, movement patterns, and temporal conditions as key attributes associated with high-persistence counties. The findings demonstrate that the post-2010 incident plateau is sustained disproportionately by a limited number of geographically concentrated environments and provide a framework for supporting more targeted safety interventions. Full article

Water-intensive industries in the Mediterranean face supply stress and decarbonisation pressure simultaneously. We develop an operations-research decision-support framework that treats the firm as one node of a small industrial resource cluster and prices the cost and carbon-equivalent emissions of five alternative supply trains—municipal water, river water, groundwater, rainwater harvesting and brewery wastewater reuse—within a multi-objective Linear Program. Each train carries engineering-grounded expenditures, energy intensities and grid emissions, and a weighted-sum scalarisation is solved daily for 365 days under three managerial scenarios. On a Cretan microbrewery whose 2022 demand of 5250 m³ is met from the municipal network, the balanced and cost-focused scenarios coincide on a single optimum that cuts the Levelised Cost of Water by 25.3% and emissions by 40.7%, while the eco-friendly scenario yields a 19.3% cost and 51.7% emissions reduction. LP duality, shadow prices and an extended sensitivity programme (diversification, capacity, grid factor, discount rate, RO recovery and demand profile) turn the optimisation into a decision-support package: optimal daily allocations, shadow-price signals on capacity and demand, and robustness diagnostics for capital planning, dispatch and risk management. Results are site-specific, but the framework and its diagnostics transfer in structure to clusters sharing the same convex-polytope source geometry; transposition to energy cooperatives is future work. Full article

Internet-of-Things (IoT) devices in smart homes are vulnerable to passive traffic fingerprinting, where an adversary captures encrypted IEEE 802.11 frames and identifies devices using MAC-layer metadata such as packet sizes and inter-arrival times. Existing defenses based on padding, traffic shaping, or synthetic cover traffic can remain vulnerable because artificial timing signatures are detectable by machine learning classifiers. This paper proposes a game-theoretic framework for evaluating Wi-Fi MAC-layer cover-traffic injection defenses. We introduce donor-based mimicry injection, in which the access point injects a replica of a paired device’s authentic traffic into each device’s stream. We compare donor mimicry with fixed-rate, exponential, and uniform synthetic baselines across 198 scenario instances (156 unique defender configurations) and eight classifiers using 10-fold cross-validation. Donor mimicry at 100% bandwidth overhead reduces the best attacker’s balanced accuracy to 33.5%, whereas synthetic methods at equal overhead reach 93.9%, showing that behavioral realism, rather than injected volume alone, drives effectiveness. Modeling the interaction as a finite two-player zero-sum game yields a mixed-strategy Nash equilibrium with game value 0.247 within the evaluated strategy space; a deployable deterministic defense holds the best pairing-unaware attacker to 25.9% balanced accuracy, near the four-class random baseline of 25%. A pairing-aware robustness analysis shows that an attacker who can orient the donor-induced identity swap recovers near-baseline accuracy, so the four-class protection presumes pairing secrecy and the durable effect is pair-level anonymity. The defense operates at the access point and requires no IoT device modifications. Full article

To improve the energy conversion performance and long-term structural stability of stay vane mixed-flow chemical pumps used for industrial residual pressure recovery, this paper establishes a coupled numerical framework of computational fluid dynamics (CFD) and finite element structural analysis (FEA). The internal flow evolution, radial hydraulic excitation, transient pressure oscillation and impeller mechanical bearing capacity are systematically investigated under three typical flow states: partial load 0.7 Q_d, design condition 1.0 Q_d and overload 1.2 Q_d. The results show that the flow inside the pump is smooth and there is no obvious backflow or separation under the rated working condition, and the energy conversion efficiency is the best. When operating under partial discharge, boundary layer separation and recirculating secondary vortices easily emerge inside the pump passage, which drastically elevates hydraulic energy dissipation. Meanwhile, operating load exerts a remarkable influence on the impeller’s radial hydraulic load and transient pressure oscillation intensity. The radial force and the pressure pulsation amplitude at the impeller outlet are the largest under the small flow condition, and the force is the most stable under the rated working condition. Blade passing frequency dominates the frequency components of transient pressure fluctuations. The maximum von-Mises stress on the impeller concentrates at the filet where blade roots connect with the hub, and this peak value hits 86.3 MPa under partial-load low-flow operating status. Calculated stress values for all three flow rates satisfy the structural safety criteria. The outcomes of this numerical investigation can offer reliable technical support for hydraulic performance optimization and structural dimension design of this type of mixed-flow chemical pump. Full article

This paper proposes a dynamic intelligent method for voltage violation management in high-renewable-penetration distribution networks. The method employs a dual-agent architecture: DERMS_Agent coordinates task scheduling, data management, and computational resource allocation, while Solution_Agent performs three-phase unbalanced power flow calculation and MIQP-based voltage violation joint optimization. Four key technical contributions are presented. (i) An asymmetric nodal admittance matrix is developed to incorporate transformer tap-phase-shift and capacitor branches within a unified formulation. (ii) Five categories of analytical sensitivities are systematically derived, covering transformer tap, phase shift, and capacitor compensation effects for both voltage regulation and harmonic suppression. (iii) A three-parameter MIQP joint optimization model is constructed with voltage deviation minimization as the objective and three-phase unbalance and resonance avoidance as constraints. (iv) A two-stage hybrid solution strategy combining Ipopt continuous relaxation with Gurobi neighborhood enumeration is designed to achieve real-time solvability. Validation on a real 10 kV feeder with 91 transformer areas over 768 time sections (8 days) demonstrates a 95.6% voltage violation resolution rate within the first three polling cycles and an average single-section solution time of 0.83 s, satisfying the real-time requirements of 15 min operational control cycles. Full article