Search Results (7,779)

A conventional resonant-inductive wireless charging system includes a power factor corrector (PFC) to maintain a high input power factor (PF) and low distortion of the input current (THDI). Although a conventional low-power wireless charging system with a PFC has relatively simple power electronic circuitry, its control stage is comparatively complex and expensive. This complexity arises because it relies on multiple feedback loops, as well as a radio communication link with complex communication protocols. As a result, the design complexity and development time are relatively high, and a highly qualified engineer with strong programming and communication expertise is needed. Some state-of-the-art solutions have eliminated the wireless communication link at the cost of increased size of the receiving side. To overcome these drawbacks, this paper proposes a simpler control and communication method that combines output voltage and current limiting with a low-latency wireless communication link transmitting 1-bit logic signals. This approach improves the cost-effectiveness of the control circuit, reduces system complexity, and keeps the receiving side compact, while maintaining performance comparable to conventional and state-of-the-art solutions. The proposed method is validated through simulations and experiments using a 60 W prototype. Results show that the power-factor-corrected wireless charging system with the proposed control and communication scheme achieves a THDI of 4.3%, a power factor of 0.99, high charging voltage accuracy (±0.5%), and satisfactory current accuracy (±9%). Full article

Maintaining consistent part quality remains a critical challenge in industrial additive manufacturing, particularly in laser-based powder bed fusion of polymers (PBF-LB/P), where crystallization-driven thermal instabilities, governed by isothermal crystallization within a narrow sintering window, precipitate defects such as curling, warping, and delamination. In contrast to metal-based systems dominated by melt-pool hydrodynamics, polymer PBF-LB/P requires monitoring strategies capable of resolving subtle spatio-temporal thermal deviations under realistic industrial operating conditions. Although machine learning, particularly convolutional neural networks (CNNs), has demonstrated efficacy in defect detection, a structured evaluation of heterogeneous modeling paradigms and their deployment feasibility in polymer PBF-LB/P remains limited. This study presents a systematic cross-paradigm assessment of unsupervised anomaly detection (autoencoders and generative adversarial networks), supervised CNN classifiers (VGG-16, ResNet50, and Xception), hybrid CNN-LSTM architectures, and physics-informed neural networks (PINNs) using 76,450 synchronized thermal and RGB images acquired from a commercial industrial system operating under closed control constraints. CNN-based models enable frame- and sequence-level defect classification, whereas the PINN component complements detection by providing physically consistent thermal-field regression. The results reveal quantifiable trade-offs between detection performance, temporal robustness, physical consistency, and algorithmic complexity. Pre-trained CNNs achieve up to 99.09% frame-level accuracy but impose a substantial computational burden for edge deployment. The PINN model attains an RMSE of approximately 27 K under quasi-isothermal process conditions, supporting trend-level thermal monitoring. A lightweight hybrid CNN achieves 99.7% validation accuracy with 1860 parameters and a CPU-benchmarked forward-pass inference time of 1.6 ms (excluding sensor acquisition latency). Collectively, this study establishes a rigorously benchmarked, scalable, and resource-efficient deep-learning framework tailored to crystallization-dominated polymer PBF-LB/P, providing a technically grounded basis for real-time industrial quality monitoring. Full article

Anomaly detection on tabular data is widely used in fraud detection, predictive maintenance, and medical screening. While heterogeneous ensembles combining multiple detection paradigms achieve strong performance, their computational cost limits deployment in latency-sensitive or resource-constrained environments. We propose KD-AnomalyNet, a teacher–student framework that distills anomaly knowledge from a high-capacity ensemble into a lightweight neural model for efficient inference. Beyond performance replication, we study how anomaly representations transfer during distillation. To this end, we introduce a noise perturbation analysis that serves as a diagnostic probe for representation stability without introducing additional trainable components. Experiments on ten benchmark datasets show that the distilled model preserves up to 98.5% of the teacher’s AUC-ROC on the nine capacity-sufficient datasets (84.7% mean retention across all ten datasets) while achieving 26–181× inference speedups. Our analysis reveals which forms of anomaly knowledge transfer reliably—global outliers (78% transfer) and isolation-based detection (88% retention)—and which degrade under compression—local outliers (20% transfer) and neighborhood-based detection (76% retention)—providing practical guidance for deploying distilled anomaly detectors. Full article

Modern manufacturing quality control demands intelligent, adaptive inspection systems capable of real-time contamination detection with minimal computational overhead. We present a five-agent vision framework for material-aware contamination detection in manufacturing environments. The system comprises: a Material Classification Agent that identifies contamination type (fiber, sand, or mixed), three Material-Specific Detection Agents, each employing patch-based CNNs optimized for their respective material with dynamic patch size selection (128 px, 256 px, 384 px), and an Adaptation Agent that monitors performance and eliminates consistently failing patch size configurations. This hierarchical architecture enables intelligent routing to specialized detectors and continuous refinement through performance-driven adaptation. The Material Classification Agent achieves 98% accuracy in contamination type identification. Material-specific agents demonstrate F1-scores of 0.968 (fiber), 0.977 (sand), and 0.977 (mixed) with real-time inference (2.40–11.11 ms per 512 × 512 image). The Adaptation Agent implements selective patch size elimination: configurations failing quality thresholds (F1 < 0.5) across multiple evaluation cycles are removed from the detection pipeline. On the synthetic test split used in this study, comparative evaluation against PatchCore, WinCLIP, and PaDiM shows 3–45× higher F1-scores with superior accuracy–latency trade-offs, validating the efficacy of specialized material-aware architectures for manufacturing contamination detection. Full article

Time delays are intrinsic to energy systems, arising from transport phenomena, communication latency, and control dynamics; however, their accurate modeling remains challenging, particularly under variable operating conditions. The most common delays are constant over time and are easy to model and simulate. However, simulation tools of time-varying delay systems rely on signal-delay representations that fail to enforce conservation laws, leading to unphysical results in applications involving mass or energy transport. This study develops a physically consistent mathematical framework for time-varying transfer delays that explicitly couples kinematic evolution with conservation principles through a dynamic gain term. A systematic classification is introduced, distinguishing between signal delays (information transfer) and transfer delays (physical transport), further categorized by the source of variability in time delay into Types R (variable extraction), W (variable supply), and M (variable medium). The proposed formulation was implemented in Simulink through newly developed functional blocks supporting all delay variants and validated against representative heat transport scenarios. Comparative analysis demonstrates that standard signal-delay models violate energy conservation by generating spurious energy, whereas the proposed transfer-delay formulation preserves physical consistency under variable-flow conditions. The framework provides a rigorous foundation for accurate modeling of district heating networks, renewable energy integration with power-to-gas systems, thermal storage, and smart grid communications, supporting the development of reliable control strategies essential for the ongoing energy transition. Full article

Nearest-neighbor classifiers are accurate and easy to deploy, but their memory footprint and inference time grow with the size of the reference set. This paper studies an evolutionary prototype selection strategy for k-nearest neighbor (K-NN) classification aimed at extreme, class-balancedreduction. A compact genetic algorithm (GA) evolves a fixed number of prototype indices per class drawn from a disjoint design partition; the selected prototypes are then used by a 1-NN classifier, with fitness defined as the number of correctly classified test instances. To address concerns about generality and baseline strength, we evaluate an experimental suite including synthetic 2D Gaussians ($\sigma =0.5$ and $\sigma =1.0$) and a 3D three-moons geometry, as well as public benchmarks spanning binary and multi-class settings and higher-dimensional data (Breast Cancer Wisconsin, Wine, Reduced MNIST/Digits 8 × 8, Forest CoverType with seven classes, and a 10D five-class spiral benchmark). We compare against K-NN baselines with $k\in \{1,3,5,7\}$ using all design samples, and include GA operator ablations (GA1/GA2/GA3). Each scenario is repeated over 30 independent runs, reporting mean ± std, min/max, per-run distributions, win/tie/loss counts, and non-parametric significance tests (paired Wilcoxon with Holm correction; Friedman where applicable). Across datasets, the GA-selected prototype banks—often orders of magnitude smaller than the full design set—match or improve accuracy, with frequent statistically supported wins against strong K-NN baselines, and in the hardest cases provide substantial compression with no loss relative to the best baseline. These results establish a reproducible baseline for extreme, class-balanced prototype reduction suitable for memory- and latency-constrained deployments and for fair comparison against more elaborate prototype selection methods. Full article

In many organizations, business intelligence systems support analytical reporting and operational decision making. As data volumes grow and analytical tasks become more complex, architectures based on centralized processing pipelines increasingly face limitations related to scalability and timely response. These challenges motivate the development of alternative architectural approaches capable of operating efficiently in data-intensive environments. This study presents a modular multi-agent business intelligence framework that distributes analytical tasks across autonomous agents and applies lightweight reinforcement learning at the decision-making stage. The analytical workflow is decomposed into agents responsible for data collection, preprocessing, analytical modeling, and decision execution. Decision adaptation relies on localized policy updates driven by operational feedback, which avoids complex learning coordination and helps preserve system stability and interpretability. The proposed framework is evaluated using real-world transactional data from an electronic commerce setting. Experimental results show that the approach consistently outperforms centralized analytical pipelines and non-agent machine learning baselines in terms of processing efficiency, classification accuracy, and balanced classification performance. Threshold-independent evaluation further confirms stronger discriminative behavior across varying decision thresholds. In addition, stability analysis across repeated experimental runs indicates reduced performance variance and more predictable system behavior. These findings suggest that the proposed multi-agent business intelligence framework provides a practical and scalable alternative to centralized analytical architectures for data-intensive decision-support environments, while maintaining the robustness and transparency required in enterprise systems. The evaluation is limited to a single dataset and a classification task, and results should be interpreted within this scope. Experiments on the Online Retail dataset (UCI Machine Learning Repository) show an average accuracy of 0.89 ± 0.012 (baseline: 0.74 ± 0.029) and decision latency of 94 ± 9 ms (baseline: 137 ± 16 ms) across 10 independent runs, indicating stable behavior under repeated execution. Full article

While typical benign paroxysmal positional vertigo (BPPV) presents with reproducible patterns of nystagmus and vertigo during positional testing, atypical variants often deviate from typical patterns, making diagnosis more complex. Recognizing atypical BPPV is crucial to avoid misdiagnosis and inappropriate management. This study aims to describe the clinical spectrum of atypical BPPV, differentiate it from central positional vertigo, and provide practical diagnostic guidance for clinicians. A narrative review was conducted to explore the clinical spectrum of atypical BPPV. Findings indicate that it may present with vertigo without nystagmus, conflicting torsional components in bilateral cases, or persistent symptoms despite repositioning maneuvers. Canal switch and pseudo-spontaneous nystagmus have also been described. Although these variants may mimic central etiologies, the absence of consistent neurological signs supports a peripheral mechanism. Diagnosis relies on detailed assessment of nystagmus characteristics—such as latency, /duration, and direction—as well as the exclusion of red flags, like direction-changing nystagmus without head movement, vomiting, or non-positional ocular motor abnormalities. Atypical BPPV remains a diagnostic challenge and requires careful bedside assessment and clinical testing. Understanding these variants is essential for timely and appropriate treatment. When doubt persists and resolution with treatment does not occur, neuroimaging should be considered to exclude central pathology. Full article

The interoperability of electronic health records in Colombia faces a critical gap between the regulatory mandates established by the Colombian regulatory framework and the actual technical capacity of healthcare institutions to implement them. This article presents PIRE (Electronic Records Interoperability Platform), an open-source architecture that demonstrates the viability of end-to-end FHIR systems in the Colombian context. The main objective was to develop a platform capable of integrating health data from biomedical devices into an FHIR server, preserving clinical semantics through LOINC terminologies. The methodology followed an iterative development approach, implementing a HAPI FHIR server on AWS, a normalization application in Flask, and clinical visualization modules aligned with the FHIR Core CO Implementation Guide. The Bioharness-3 device was used to capture metrics on heart rate, respiratory rate, activity, and posture. The platform achieved a data normalization latency of 104–438 ms per record and 100% semantic validation against the FHIR Core CO profiles, validating compliance with Colombian IHCE specifications. It is concluded that PIRE constitutes a reproducible reference model for healthcare institutions that wish to implement interoperability as a cost-effective solution. Full article

The resilience of Intelligent Transportation Systems (ITSs) heavily relies on distributed Battery Energy Storage Systems (BESSs) deployed in harsh, unattended highway environments. Traditional maintenance strategies often fail to account for the dynamic feedback between battery aging, environmental stress, and maintenance response latency. This study proposes a system dynamics (SD) framework to evaluate and optimize the resilience of these critical power infrastructures. By modeling the nonlinear interactions among sensor data, controller logic, and remote discharge terminals, we simulate the system’s dynamic behavior over a 36-month lifecycle. The results reveal a critical “scalability threshold”: when battery pack quantity exceeds 40 units, the system’s self-healing time increases disproportionately, degrading resilience. Furthermore, the study identifies 384 V as the optimal “Resilience Topology Voltage”, offering the fastest recovery speed by balancing thermal stability with consistency management efficiency. These findings provide theoretical guidelines for configuring BESS capacity and optimizing remote maintenance protocols to ensure uninterrupted highway operations. Full article

Reliable and real-time environmental monitoring is essential for controlling pollution and protecting public health. However, conventional station-based measurements are expensive and often lack spatial and temporal resolution. This paper proposes a low-cost multimodal environmental monitoring system. Experiments verified that thin-film thermocouples exhibit near-linear voltage–temperature characteristics ($R^2>0.99$). Integration of the AI data pipeline substantially enhances monitoring accuracy: the proposed fusion strategy reduces relative error to approximately 2.3% under typical noise conditions, with a correlation coefficient of 0.79 between predicted and observed PM2.5 values. This research provides a scalable blueprint for edge-deployable environmental monitoring. A thin-film thermocouple with a fast response time is used as a temperature sensor and is statically calibrated against a K-type reference. To improve dynamic tracking and reduce measurement noise, a Kalman filter-based fusion strategy is employed, which is then compared with weighted averaging and Bayesian fusion. Simulation-driven validation is performed for thermocouple linearity, PID-based temperature control, micro-signal filtering and system-level latency and robustness. The results demonstrate that thin-film thermocouples exhibit near-linear voltage–temperature characteristics (R² > 0.99) with Seebeck coefficients ranging from 40.92 to 42.08 μV/°C, close to the theoretical K-type value of 42.87 μV/°C. The proposed fusion strategy reduces relative error to ~2.3% under typical noise conditions, enabling stable, real-time processing with near-second latency for 10,000-point batches. This study summarizes the design considerations for selecting and calibrating sensors and for achieving AI robustness in the presence of drift and faults. It provides a scalable blueprint for edge-deployable environmental monitoring. Full article

This narrative literature review explores the clinical use of Ketamine as part of an untested hypothetical model framework for bridge therapy for acute suicidality. Long-term suicide rates continue to increase in the United States and in many other countries, creating a pressing public health challenge with a variety of treatment considerations. Existing standard-of-care SSRI therapeutics have a delay between administration and symptom relief at 2–6 weeks, leaving a so-called danger zone of about 1–3 months of risk for suicidal follow-through behaviors. Ketamine, a potent NMDA (N-methyl-D-aspartate) receptor antagonist, has recently seen widespread interest in both regulatory and clinical settings for increasing neuroplasticity and alleviating depressive symptoms. Ketamine’s mechanism-of-action through mTORC1 is much faster than SSRI’s downstream transcriptional regulation, leading to quicker relief of suicidal symptoms and the removal of the danger zone lag period. The current literature suggests that a controlled, supervised subanesthetic dose of Ketamine in a clinical setting has low risks of addiction or abuse, distinguishing therapeutic uses of Ketamine from recreational uses. While the biological efficacy of Ketamine is established, this conceptual review focuses on a possible initial hypothetical framework of a “Bridge Protocol.” We searched PubMed, Google Scholar, The Cochrane Library, and PsycINFO (January 2000–December 2025) to synthesize evidence regarding SSRI latency, acute Ketamine protocols, and post-discharge safety. We conclude that while promising, the proposed Ketamine Bridge Therapy requires rigorous longitudinal validation and sustained clinical studies before it can be safely used and experience widespread adoption. Full article

Extreme hydro-meteorological events are intensifying under climate change, disproportionately disrupting last-mile healthcare in flood-prone geographies. In this study, flood-adaptive primary care clinics (FAPCCs) integrated with islandable smart microgrids and a rapid-deploy medical technology stack (MedTech) are developed and evaluated to ensure continuity of essential services (triage, maternal and child health, vaccination cold-chain, minor procedures, diagnostics, and telemedicine) during fluvial, pluvial, and coastal flooding. Evidence on resilient health facilities, microgrid architectures, distributed energy resources, and modular clinical systems is presented in a multi-layer systems design: (1) a modular, amphibious, and elevatable clinic chassis; (2) a photovoltaic–battery–diesel hybrid system with demand-aware energy management; (3) redundant connectivity long-term evolution/fifth-generation, satellite, and very high frequency; (4) a rapid-deploy MedTech kit including point-of-care diagnostics, low-temperature cold-chain, negative-pressure isolation, and sterilization modules; and (5) flood-aware logistics using unmanned aerial vehicle/unmanned surface vehicle. A mixed-integer linear programming sizing is formulated and dispatched with a continuity-of-care reliability metric that couples energy availability to clinical throughput. Simulation across three archetypal sites (peri-urban delta, inland riverine, coastal estuary) shows that FAPCCs achieve the service availability of higher than 99.5% across 7-day grid outage scenarios while reducing fuel use by 62–81% relative to diesel-only baselines, maintaining vaccine temperatures within 2–8 °C with <0.1% thermal excursion time, and sustaining telemedicine quality of service with <150 ms median uplink latency in hybrid networks. A life-cycle cost analysis indicates a 7.1–9.8 year discounted payback from fuel displacement and avoided service loss. Deployment playbooks and policy guidance are also proposed for Ministries of Health and Disaster Agencies in monsoon-impacted regions. Full article

Accurate forecasting of nutrition supply–demand dynamics is essential for reducing resource wastage and improving equitable allocation. However, this task remains challenging due to heterogeneous data sources, cold-start regions, and the risk of information leakage in spatiotemporal modeling. This study presents a leakage-aware multimodal machine learning framework for nutrition supply–demand forecasting. The framework integrates temporal, spatial, and contextual information within a unified architecture. It combines self-supervised temporal representation learning, causal time-lag modeling, and few-shot adaptation to improve generalization under limited or previously unseen data conditions. Heterogeneous inputs include epidemiological, environmental, demographic, sentiment, and biologically derived indicators. These signals are encoded using a PatchTST-inspired temporal backbone coupled with a feature-token transformer employing cross-modal attention. Spatial dependencies are explicitly modeled using graph neural networks. Hierarchical decoding enables multi-horizon forecasting with calibrated uncertainty estimates. Model evaluation is conducted under strict spatiotemporal hold-out protocols with explicit leakage detection. All synthetic signals are excluded from testing. Across geographically and temporally disjoint datasets, the proposed framework consistently outperforms strong unimodal and multimodal baselines. It achieves macro-F1 scores above 99.5% and stable early-warning lead times of approximately 9 days under distribution shift. Ablation studies indicate that causal time-lag enforcement and few-shot adaptation contribute most strongly to performance robustness. Closed-loop simulation experiments suggest potential reductions in nutrient wastage of approximately 38%, response latency of 19%, and operational costs of 16% when deployed as a decision-support tool. External validation on fully unseen regions confirms the generalizability of the framework under realistic forecasting constraints. Full article

Casing collar locator (CCL) measurements are widely used as reliable depth markers for positioning downhole instruments in cased-hole operations, enabling accurate depth control for operations such as perforation. However, autonomous collar recognition in downhole environments remains challenging because CCL signals are often corrupted by toolstring- or casing-induced magnetic interference, while stringent size and power budgets limit the use of computationally intensive algorithms and specific operations require real-time, in situ processing. To address these constraints, we propose Collar Recognition Nets (CRNs), a family of domain-specific lightweight 1-D convolutional neural networks for collar signature recognition from streaming CCL waveforms. With depthwise separable convolutions and input pooling, CRNs optimize efficiency without sacrificing accuracy. Our most compact model achieves an F1-score of 0.972 on field data with only 1985 parameters and 8208 MACs, and deployed on an ARM Cortex-M7-based embedded system using the TensorFlow Lite for Microcontrollers (TFLM) library, the model demonstrates a throughput of 1000 inferences per second and 343.2 μs latency, confirming the feasibility of robust, autonomous, and real-time collar recognition under stringent downhole constraints. Full article