Search Results (277)

Real-time, precise monitoring of the number and distribution of indoor personnel is crucial for building safety management, operational optimization, and personnel scheduling. However, narrow entrances and high-density passageways often lead to missed detections, false positives, and tracking failures in pedestrian detection, thereby reducing cross-line counting accuracy. Additionally, edge devices deployed in practical scenarios frequently process multiple video streams simultaneously, resulting in computational resource constraints. To address these challenges, this paper proposes a lightweight, enhanced multi-object pedestrian tracking and counting method tailored for indoor scenarios by optimizing deep learning models. Firstly, modular optimizations are applied to the YOLOv8n model to construct a more lightweight detector, RL_YOLOv8, reducing computational overhead while maintaining accuracy. Secondly, correlated pedestrian auxiliary prediction and pedestrian position change constraints are employed to mitigate ID switching, tracking interruptions, and trajectory jumps in dense scenes. Finally, a buffer zone auxiliary counting strategy is designed to further reduce missed detections of pedestrians crossing lines. Experimental results demonstrate that compared to the original detection-and-tracking-based line-crossing counting method, the improved approach effectively enhances counting accuracy and real-time performance, better meeting the requirements of practical intelligent security and crowd monitoring systems. Full article

This study proposes a wide-input high-step-up DC–DC converter with high efficiency and high voltage conversion ratio. Two coupled inductors were adopted to achieve a parallel-charging and series-discharging energy transfer mechanism, and a voltage multiplier circuit was integrated to increase the voltage gain. The proposed topology uses a single pulse width modulation signal to drive two main switches synchronously, resulting in a low switch count and simple control circuit, concurrently achieving a wide input voltage range of 24 V to 48 V. The proposed converter comprises an active switched inductor combined with a voltage multiplier circuit, achieving a high voltage gain without relying on high duty cycle operation or high-turns-ratio design. The leakage energy of the coupled inductors was recycled through a passive-clamp circuit, effectively suppressing the voltage spikes of the switching devices and reducing their voltage stress. Finally, a 1 kW converter was implemented to verify the feasibility of the proposed topology through steady-state analysis, circuit simulation, and hardware experiments. The maximum efficiencies achieved were 94.7% and 96.2% at input voltages of 24 V and 48 V, respectively. Full article

Multichannel vortex beams serve as an essential physical source for enabling multi-spot laser processing and high-dimensional spatial multiplexing communications. We demonstrate a compact, flexibly tunable multichannel vortex beam generator using spatially structured bidirectional two-color pump vortex four-wave mixing in a single ¹³³Cs vapor cell. To enhance spatial multiplexing, both sides of the cell are utilized. By engineering the propagation directions and frequencies of five input beams, we establish a nonlinear interaction region that supports 16 concurrent phase-matching conditions, thereby enabling the parallel generation of up to eight vortex channels. The orbital angular momentum of the output beams follows deterministic algebraic rules, allowing for programmable control via tailored input orbital angular momentum combinations. Moreover, the channel count can be linearly tuned by selectively deactivating pumps—each switched-off pump reduces the number of output channels by two. This flexible control over orbital angular momentum states, together with channel count and spatial arrangement, establishes a highly integrated platform for on-demand vortex generation. This work highlights the potential of spatially bidirectional structured pumping in atomic vapor to expand optical dimensionality and enhance multiplexing capacity, paving the way toward multidimensional communications, quantum networks, and integrated photonics. Full article

The deployment of Discrete Variable Quantum Key Distribution (DV-QKD) in high-traffic, short-reach environments, such as intra-data center networks, is currently constrained by the saturation of single-photon detectors. While CMOS Single-Photon Avalanche Diodes (SPADs) offer a cost-effective solution, their Secure Key Rate (SKR) is limited by detector dead time. To the best of the authors’ knowledge, this work is the first to derive a generalized detection rate model for SiPMs that addresses the dead-time bottlenecks of gigahertz-rate quantum cryptography. While methods for managing deadtime via active optical switching have been proposed, our model quantifies the benefits of passive spatial multiplexing inherent in standard SiPM arrays. Furthermore, contrasting with models designed to optimize energy resolution or characterize nonlinear charge response to light pulses, our work focuses on maximizing the detection count rate. We derive exact detection rate models for both analog (paralyzable) and digital (non-paralyzable) SiPM architectures, incorporating correlated noise sources such as optical crosstalk and afterpulsing. Simulation results indicate that SiPMs can increase detection rates by over an order of magnitude compared to single SPADs. Full article

Reliable state-of-charge (SoC) estimation is crucial for safe and efficient battery management. However, it is challenging in practice. Terminal-voltage sensitivity becomes weak in open-circuit-voltage (OCV) plateau regions. Model uncertainty also persists at practical sampling periods. To tackle this issue, this paper proposes a discrete-time, model-based SoC estimation framework. This framework combines a dual-polarization equivalent-circuit model with a tuning-light sliding-mode observer. It is specifically designed for digitally sampled battery management systems. The modeling stage includes: (i) a discrete-time DP representation suitable for embedded use, (ii) a shape-preserving PCHIP reconstruction of the OCV–SoC curve and its derivative, and (iii) an effective-slope regularization mechanism that maintains non-vanishing output sensitivity even in flat OCV regions. On top of this structure, a boundary-layer SMO is developed with output-error shaping, model-driven gain scaling, and simple bias-compensation terms based on integral correction and leaky Coulomb counting. A discrete-time Lyapunov analysis is conducted directly on the surface dynamics. This analysis shows finite-time reaching to the boundary layer and a practical limit on the steady-state error that depends on the sampling period, disturbance level, and boundary-layer width. Numerical tests on a DP model identified from experimental data indicate that the proposed method achieves SoC accuracy similar to a switching-gain adaptive SMO. The results confirm the benefits of a model-centric design. The discrete-time formulation and convergence proof, which do not depend on high sampling rates, provide robustness advantages over traditional sliding-mode methods. The proposed method also performs better than a tuned EKF in plateau regions, requiring much less tuning effort. Full article

The rise of carbon emissions from fossil fuel-based power generation has intensified the need for efficient and low-carbon energy systems. The global CO₂ concentration has risen from 285 ppm in the pre-industrial era to nearly 420 ppm today, and this contributes to a 1°C increase in average temperature. Therefore, in this article, a hybrid photovoltaic–thermoelectric generator (PV–TEG) system integrated with a reduced-switch multilevel inverter (MLI) is proposed. This enhances renewable energy utilization and power quality. The proposed PV–TEG model recovers waste heat from PV modules, which yields an overall efficiency improvement of approximately 2–8% compared to standalone PV systems. Further, the proposed MLI operates in symmetric (seven-level) and asymmetric (11-level) modes using eight switches. The system develops high-quality stepped output voltages with a minimum component count. Simulation work is performed, and the results show a peak output voltage of ±220 V with Total Harmonic Distortion (THD) of 7.2% under R-load and reduced THD below 5% under RL and variable load conditions. The integrated system demonstrates improved efficiency, reliability, and suitability for sustainable power generation and rural electrification. Full article

Management of heavily treatment-experienced people living with HIV (HTE-PWH) remains challenging due to long antiretroviral therapy (ART) exposure and limited treatment options. We conducted an observational, real-world-data study on HTE-PWH in active care at the “Victor Babeș” Hospital, Bucharest, receiving doravirine (DOR)-based salvage regimens combined with dolutegravir (DTG) or bictegravir (BIC). Epidemiological, clinical, and laboratory variables were analyzed according to HIV acquisition mode and salvage regimen. Sixty-nine PWH were included; 57.9% male, with a median age of 36 years. Median ART duration before switch was 21.5 years. HIV was acquired parenterally in childhood (PM) in 64.7% cases. Salvage regimens included BIC/FTC/TAF + DOR (50.0%), 3TC/TDF/DOR + DTG (35.2%), and 3TC/DTG + DOR (14.7%). The median nadir CD4 count was 37 cells/µL, and the median viral load at diagnosis was 5.24 log10 copies/mL. Switching was performed for regimen simplification (n = 32) or non-adherence-related virological failure (n = 37). At switch, 53.6% had detectable viremia. Viral suppression was achieved in 68.3% at 6 months and 75.0% at 12 months. Individuals with PM infection were younger and had longer ART exposure than those with heterosexual acquisition. DOR-based salvage regimens combined with DTG or BIC were effective in adherent HTE-PWH, particularly those with extensive ART histories. Full article

As integrated circuits face increasingly stringent demands regarding power consumption, area, and stability, integrating novel spintronic devices with computing architectures has become a crucial direction for breaking through traditional computing paradigms. In the paper, switching mechanism of Magnetic Tunnel Junctions (MTJs) under the synergistic effect of Voltage-Controlled Magnetic Anisotropy (VCMA) and the Spin Hall Effect (SHE) is investigated. VCMA-assisted switching SHE-MTJ device is adopted, and a macrospin approximation model is established based on the Landau-Lifshitz-Gilbert (LLG) equation to systematically analyze its dynamic characteristics. The research demonstrates that applying VCMA voltage pulses with appropriate amplitude and width can significantly reduce the required spin Hall current density and pulse width for switching, thereby effectively minimizing ohmic losses and Joule heating. Furthermore, by incorporating a thermal fluctuation field, voltage-controlled SHE-MTJ device with stochastic switching behavior can be constructed, obtaining an approximately sigmoidal voltage-probability response curve. This provides an ideal physical foundation for stochastic computing and neuromorphic computing. Based on the above established fundamental discovery, an in-memory computing architecture supporting binarized Convolutional Neural Networks (CNNs) is proposed and designed in the paper. Combined with the lightweight network SqueezeNet, this architecture achieves a Top-1 recognition accuracy of 72.49% on the CIFAR-10 dataset, with a parameter count of only 1.25 × 10⁶. This work offers a feasible spintronic implementation scheme for low-power, high-energy-efficiency edge-side intelligent chips. Full article

Filled-polymer coatings enable functional surfaces for selective metallisation, wetting control and local conductivity, but pulsed-laser texturing is often limited by process non-uniformity caused by scan kinematics and plume shielding. Here, we develop a three-tier numerical workflow for nanosecond pulsed-laser surface treatment of a thermoplastic coating containing glass microspheres (baseline case: PLA matrix with V_f = 0.20; spheres represented via an effective optical transport model). Tier 1 predicts spatially resolved ablation depth under raster scanning, using an incubation law and regime switching (no-removal/melt-limited/logarithmic ablation/blow-off) coupled to a dynamic shielding factor. Tier 2 computes the 1D transient (pulse-averaged) temperature field and the thickness of the thermally softened layer. Tier 3 models post-pulse capillary redistribution of the softened layer to estimate groove reshaping. The simulations show that scan overlap and shielding dynamics dominate groove homogeneity more strongly than average power alone: under identical average power, variations in local pulse count and shielding lead to significant changes in depth statistics and regime fractions. The workflow produces quantitative maps and summary metrics (mean depth, P5–P95 range, uniformity index and regime fractions) and demonstrates how controlled reflow can smooth peaks while preserving groove depth. These results provide a predictive tool for laser parameter selection and process optimisation prior to experimental trials. Full article

We present WM-Classroom v1.0, a pedagogical multi-species agent-based model (ABM) designed for educational purposes in predator–prey–harvest systems. The model embeds a predator, two prey breeds, and human harvesters on a homogeneous 50 × 50 grid with weekly time steps, implementing random movement, abstract energetics, prey consumption, reproduction, legal harvest with species-specific cut-offs and seasons, optional predator control, and a poaching switch. After basic technical checks (energetic calibration, prey composition, herbivore viability), we explore the consistency of the model under illustrative scenarios including no hunting, single-prey harvest, hunter-density and season-length gradients, predator removal, and poaching. In the no-hunting baseline (n = 100), mean end-of-run abundances were 22 deer, 159 boar, and 45 wolves, with limited extinction events. Deer-only harvest often drove deer to very low end-of-run counts (mean 1–16) with extinctions in 2–7/10 replicates across cut-offs, whereas boar-only harvest showed higher persistence (mean 11–74) and boar extinctions occurred only at the lowest cut-off (3/10). Increasing hunter numbers or season length depressed prey and could indirectly reduce wolves via prey depletion. Legal predator control reduced predators as designed, while poaching had little effect under the implemented rules. Because interaction and prey-choice rules are simplified for transparency, outcomes should be interpreted as conditional on model assumptions. WM-Classroom v1.0 provides a didactic sandbox for courses, professional training, and outreach, with extensions (habitat heterogeneity, age/sex structure, probabilistic diet/kill success, and calibration/validation) outlined for future versions. Full article

This paper proposes a dynamic voltage restorer (DVR) based on a new three-phase multilevel inverter (MLI). An integral component of DVRs is the power electronic converter. At medium-to-high voltage levels, MLIs are the ideal converters for DVR applications because lower voltage-rated switches are used to generate high voltages, thus minimizing power losses. The proposed three-phase MLI generates 15 levels of load voltage per phase, using a reduced component count: eight lower-rated semiconductor power switches, four primary DC voltage sources, two auxiliary DC sources, and eight driver circuits per phase. Additionally, each phase features a low-frequency transformer with voltage-boosting and galvanic isolation capabilities. The switching sequence of the proposed MLI is simpler to execute using fundamental frequency control; this methodology provides reduced switching stress and reduced switching losses as merits. Structurally, the proposed MLI is less complex and thus scalable. The proposed DVR, based on three-phase MLI, efficiently offsets power quality problems such as voltage swell, voltage sags, and harmonics for balanced and unbalanced loads. The operational performance of the proposed DVR-MLI is verified by a simulation, using PSCAD software and an experimental prototype. Full article

Accurate fruit counting is essential for yield evaluation and automated management in greenhouse eggplant production. This study presents a lightweight detection and counting method based on an improved YOLOv5s–DeepSORT framework. To reduce computational cost while preserving accuracy, we replace the YOLOv5s backbone with MobileNetV3, insert an Efficient Channel Attention (ECA) module to enhance discriminative fruit features, and substitute the neck C3 block with C2f to strengthen multi-scale feature fusion. Compared with the original YOLOv5s, our improved YOLOv5s increases precision by 2.3% while reducing the number of parameters and FLOPs by 37.0% and 50.9%, respectively. For counting, we integrate DeepSORT with a counting-zone strategy that increments the count once per target when the bounding-box center first enters the counting zone, thereby mitigating identity switches (ID switches) and suppressing duplicate counts. Experimental results demonstrate that the proposed method enables accurate and real-time eggplant fruit counting in complex greenhouse scenes, providing practical support for automated yield assessment on inspection robots. Full article

This paper presents a nine-switch inverter for brushless operation of wound-field synchronous machines with excellent rotor flux regulation freedom. The manufacturing cost of permanent magnet machines is high due to the instability of rare-earth magnet prices in the global market. Moreover, conventional wound-field synchronous machines (WFSMs) have problems with their rotor brushes and slip-ring assembly, wherein the assembly starts to malfunction in the long run. Furthermore, recently, some brushless WFSM topologies have been investigated to eliminate the problems associated with rotor brushes and slip rings, but they have either a high cost due to a double-inverter, or low flux regulation freedom due to a single inverter (−i_d). The proposed nine-switch topology achieves a low cost by using a single inverter with nine switches and excellent flux control through three variables (−i_d, i_q, and i_f), making it highly suitable for wide-speed applications. In the proposed topology, the machine’s armature winding is divided into two sets of coils: ABC and XYZ. A 12-slot and 8-pole machine stator is wound with armature winding coils ABC and XYZ, creating six terminals for injecting currents and two neutrals from each ABC and XYZ coil set. The current to the ABC and XYZ coils is supplied by a nine-switch inverter. The inverter is specially designed to supply rated currents to the ABC winding coils and half of the rated current to the XYZ winding coils. The number of turns of the ABC and XYZ winding coils are kept the same so they produce the same winding function. However, the current in the XYZ winding coils is half compared to that of the ABC winding coils, which creates an asymmetrical airgap magnetomotive force (MMF). The asymmetrical airgap MMF contains two working harmonics, i.e., fundamental MMF for torque production and an additional sub-harmonic MMF component for rotor field brushless excitation. The rotor field is controlled by the difference in current of the two armature winding coils: ABC and XYZ. The proposed topology is validated through theoretical analysis and finite element simulations of electromagnetic and flux regulation. A 2D finite-element analysis is performed to verify the idea. The proposed topology is capable of establishing a 9.15 A dc current in the rotor field winding coil, which consequently generates a torque of 7.8 N·m with a 20.30% torque ripple. Rotor field flux regulation was analyzed from the stator ABC and XYZ coils current ratio ζ. The ratio ζ is analyzed as 2 to 1.3; subsequently, the inducted field currents were 9.15 A dc to 4.8 A dc, respectively. Full article

This paper proposes a novel seven-level switched-capacitor multilevel inverter featuring a shared front-end DC-link structure that achieves triple voltage gain with reduced component count. A distinctive feature of this design is its inherent capacitor voltage self-balancing capability, thereby eliminating the need for complex control algorithms typically associated with multilevel converters. Moreover, the topology demonstrates particularly significant advantages in three-phase implementations, where a single DC source, front-end switching devices, and capacitors can be shared across all phases—thus substantially reducing component count and system complexity compared to conventional designs. Additionally, this paper proposes an improved carrier-based modulation strategy for this topology requiring only a single triangular carrier, along with a systematic method for determining optimal capacitance values. Through detailed comparative assessment against state-of-the-art switched-capacitor seven-level inverters, the superior performance characteristics of the proposed topology are clearly demonstrated. Finally, simulation results under various operating conditions are presented and subsequently validated through experimental testing on a laboratory prototype, confirming the practical viability of the proposed solution. Full article

This paper presents a data-driven probabilistic framework for analysing power system faults using Monte Carlo simulations. The study evaluates the operational reliability of multiple high-voltage switchgear topologies—including single-busbar systems, double-busbar systems, and ring-type configurations—by modelling the stochastic behaviour of disconnectors, circuit breakers, busbars, and withdrawable switching elements with bypass arrangements. Realistic unavailability parameters derived from statistical reliability data are used to generate fault intervals for each device, enabling the simulation of millions of operational scenarios and capturing both full and partial outage events. The proposed methodology quantifies outage probabilities, identifies critical components, and reveals how devices count, switching logic, and system redundancy influence overall resilience. Results show significant reliability differences between topologies and highlight the importance of optimized substation design for fault tolerance. The developed probabilistic framework provides a transparent and computationally efficient tool to support planning, modernization, and predictive maintenance strategies in transmission and distribution networks. Findings contribute to improved fault diagnosis, enhanced grid stability, and increased reliability in both conventional and renewable-integrated power systems. Full article