Search Results (3,025)

Prediction models for polymer injection molding quality often degrade due to shifts in operating conditions caused by variations in melting temperature, cooling efficiency, or machine conditions. To address this challenge, this study proposes a drift-aware dynamic quality-monitoring framework that integrates hybrid-feature autoencoder (HFAE) drift detection, sliding-window reconstruction error analysis, and a mixed-feature artificial neural network (ANN) for online quality prediction. First, shifts in processing parameters are rigorously quantified to uncover continuous drifts in both input and conditional output distributions. A HFAE monitors reconstruction errors within a sliding window to promptly detect anomalous deviations. Once the drift index exceeds a predefined threshold, the system automatically triggers a drift-event response, including the collection and labeling of a small batch of new samples. In benchmark tests, this adaptive scheme outperforms static models, achieving a 35.4% increase in overall accuracy. After two incremental updates, the root-mean-squared error decreases by 42.3% across different production intervals. The anomaly detection rate falls from 0.86 to 0.09, effectively narrowing the distribution gap between training and testing sets. By tightly coupling drift detection with online model adaptation, the proposed method not only maintains high-fidelity quality predictions under dynamically evolving injection molding conditions but also demonstrates practical relevance for large-scale industrial production, enabling reduced rework, improved process stability, and lower sampling frequency. Full article

The grounding grid of a substation is a crucial component for ensuring normal operation. However, since it is buried underground for long periods, it is highly susceptible to electrochemical corrosion. This corrosion leads to a reduction in its grounding performance, and severe corrosion may endanger the reliable operation of high-voltage equipment and secondary relay-protection equipment, as well as the safety of personnel. In this paper, the electromagnetic field analysis method is used to conduct simulation modeling of the grounding grid. A different-frequency current is injected into the grounding grid to study the variation law of the surface magnetic field distribution when corrosion occurs to different degrees at different positions in the grounding grid. Through the analysis of the evolutionary characteristics of the magnetic field distribution, the corrosion-induced breakages in the grounding grid are located and a comprehensive state evaluation is carried out. The results show that when a fault occurs in a conductor at the same position, the variation amplitude of the surface magnetic field gradually increases with increased corrosion. Based on this finding, an online monitoring algorithm for the location of corrosion-induced breakages and state evaluation of the grounding grid is proposed. A comprehensive evaluation model is constructed by combining the grounding resistance value and corrosion characteristic value to accurately locate the fault. Full article

Physical activity (PA) is recommended in pregnancy and postpartum to support mental and physical well-being. However, little is known about the association between pregnancy and postpartum PA and interrelated factors in PA engagement. The objectives of this study were to (a) measure and understand PA engagement in pregnancy and postpartum and how related variables (i.e., work status, number of children, time since birth, PA during pregnancy) are associated with postpartum PA and (b) to examine two self-reported methods for assessing PA postpartum: self-reported PA volume and intensity through questionnaire vs. asking whether women met PA guidelines of 150 min of moderate-to-vigorous PA per week. A total of 526 women who had given birth within the past 18 months completed an online questionnaire (majority were Canadian or American). Descriptive statistics were used to assess PA during pregnancy and postpartum, and chi-square analyses were run to assess the association between related variables and to evaluate self-report methods. During pregnancy, 27.4% of women reported meeting PA guidelines and 25.3% reported meeting PA guidelines postpartum. No significant relationship between return-to-work status or number of children and meeting PA guidelines was found. Participants ≤12 weeks postpartum were less likely to meet PA guidelines compared to those >12 weeks postpartum. There was a significant relationship between meeting PA guidelines during pregnancy and engagement in PA postpartum. Lastly, there was a significant relationship between a binary measure of meeting PA guidelines (i.e., yes or no) and calculated PA volume and intensity when provided through type, frequency, and duration. This study provides insights into PA patterns of women during pregnancy and postpartum. Findings highlight the need for targeted interventions to support maternal health and well-being, emphasizing the importance of establishing PA habits during pregnancy to assist in maintenance postpartum. Results also suggest that simplified assessment methods may be effective for monitoring women’s PA, potentially making it easier for healthcare providers to track and promote healthy behaviors among new mothers. Full article

Online customer reviews contain rich signals about product value but are difficult to convert into strategy-ready evidence. This study proposes an end-to-end framework that maps review text to the Value Proposition Canvas (VPC) and quantifies alignment between user needs and product performance. Using customer reviews for three Samsung Galaxy Watch generations, an LLM extracts six dimensions (Customer Jobs, Pains, Gains, Feature Gaps, Emotions, Usage Context). Extracted phrases are embedded with a transformer model, clustered via K-means with data-driven k selection, and labeled by an LLM to form an interpretable taxonomy. Subsequently, the analysis derives frequency profiles, a gap density indicator, a context–gap matrix, and a composite Product–Market Fit (PMF) score that balances gain rate, gap rate, and coverage with sensitivity analysis to alternative weights. The findings show predominantly positive affect, with unmet needs concentrated in battery endurance and interaction stability. Productivity- and interaction-centric jobs attain the highest PMF score, while several monitoring-centric jobs are comparatively weaker. Significant cross-generation differences in job composition indicate evolving usage priorities across successive releases. The framework provides a scalable, reproducible path from unstructured VOC to decision support, enabling data-driven prioritization for product and UX management while advancing theory-grounded analysis of customer value. Full article

Objective: This cross-sectional survey aimed to comprehensively gather data on radial forearm free flap (RFFF) utilization and practices in head and neck reconstructive surgery. Methods: An online questionnaire was organized into seven sections: demographics, surgeon experience, harvesting techniques, microsurgical considerations, postoperative care, flap monitoring, and outcomes. It was distributed by email to 216 head and neck reconstructive surgeons who attended the International Federation of Head and Neck Oncologic Societies (IFHNOS) congress in Rome (21–25 June 2023) using the congress mailing list. Responses were collected from 54 surgeons (25% response rate), representing 15 countries across Europe, Asia, the Americas, and Oceania, underscoring the international scope of the survey between 5 February and 25 March 2024. The questionnaire was not formally piloted or validated. Missing data were managed on a per-question basis. Descriptive statistics were used, and 95% confidence intervals (CIs) were calculated for key surgical outcomes to indicate estimate precision. Associations between categorical variables were analyzed using Pearson’s χ² test with Cramér’s V as an effect size, and relationships between continuous variables were examined using Spearman’s rank correlation (ρ) with 95% confidence intervals (CIs). Given the exploratory design and limited sample size, no correction for multiple comparisons was applied, and the risk of both Type I and Type II errors was acknowledged. Results: Variations were observed in harvesting techniques, microsurgical preferences, and postoperative care protocols. Most surgeons initiated flap harvesting concurrently with tumor resection, primarily preserving superficial sensory nerves. Regarding venous outflow, 50% of respondents preferred the cephalic vein, 19% used comitant veins, and 29% utilized both systems when possible. Perioperative antibiotic use was standard practice, though anticoagulant preferences and flap monitoring methods varied. The study achieved a high success rate for RFFF procedures, exceeding 95%, with venous thrombosis identified as the main cause of flap failure. No significant correlations were found between flap failure rate and training method (p = 0.21), specialty (p = 0.37), annual number of RFFF procedures (p = 0.89), surgeon age (p = 0.42), or hospital type (p = 0.48). Effect sizes were small to moderate, indicating weak or negligible associations. Similarly, perioperative factors such as anticoagulant use (p = 0.84), preoperative antibiotics (p = 0.42), surgical instruments (p = 0.61), suture techniques (p = 0.51), and donor vein selection (p = 0.20) showed no statistically significant associations with flap loss. Patient satisfaction assessments were inconsistent, with only 39% of surgeons routinely performing them. Conclusions: The study provides valuable insights into current RFFF practices and outcomes across an international cohort of head and neck surgeons, highlighting patterns and variability in techniques, perioperative care, and monitoring strategies. Full article

A core challenge in urban green space (UGS) management lies in precisely identifying public demand heterogeneity toward landscape elements. Grounded in Expectation Confirmation Theory (ECT), this study aims to systematically identify the key landscape elements shaping public satisfaction and elucidate their driving mechanisms to inform UGS planning. Using 107 UGS in central Beijing as case studies, this study first retrieved 712,969 social media data (SMD) from multiple online platforms. A landscape element lexicon derived from these data was then integrated with the Bidirectional Encoder Representations from Transformers (BERT) model to assess public attention and satisfaction toward the natural, cultural, and artificial attributes of UGS, achieving an accuracy of 84.4%. Finally, spatial variations and the effects of different landscape elements on public satisfaction were analyzed using GIS-based visualization, K-means clustering, and multiple linear regression. Key findings reveal the following: (1) satisfaction follows a “core-periphery” gradient, peaking in heritage-rich City Wall Parks (>0.63) and plunging in green belts due to imbalanced element configurations (~0.04); (2) naturally dominant green spaces contribute most to satisfaction, while a nonlinear relationship exists between element dominance and satisfaction: strong features enhance perception, balanced patterns mask issues; (3) regression analysis confirms natural elements (vegetation β = 0.280, water β = 0.173) as core satisfaction drivers, whereas artificial facilities (e.g., service infrastructure β = 0.112, p > 0.05) exhibit a high frequency but low satisfaction paradox. These insights culminate in a practical implementation framework for policymakers: first, establish a data-driven monitoring system to flag high-frequency, low-satisfaction facilities; second, prioritize budgeting for enhancing natural elements and contextualizing cultural elements; and finally, implement site-specific optimization based on primary UGS functions to counteract green space homogenization in high-density cities. Full article

This article presents the development and implementation of an IoT-enabled, off-grid solar power supply prototype designed to power a range of electrical devices. The developed system comprises a Photovoltaic panel, a Maximum Power Point Tracking (MPPT) charger, a 2.5 kWh/24 V high-performance LiFePO4 battery bank with a Battery Management System, an embedded controller with IoT connectivity, and DC/DC and DC/AC converters. The PV panel serves as the primary energy source, with the MPPT controller optimizing battery charging, while the DC/DC and DC/AC converters supply power to the connected electrical devices. The article includes a case study of a developed platform for powering an information and advertising system. The system features a predictive energy management algorithm, which optimizes the appliance operation based on daily solar irradiance forecasts and real-time battery State-of-Charge monitoring. The IoT-enabled controller obtains solar irradiance forecasts from an online meteorological service via API calls and uses these data to estimate energy availability for the next day. Using this prediction, the system schedules and prioritizes the operations of connected electrical devices dynamically to optimize the performance and prevent critical battery discharge. The IoT-based controller is equipped with both Wi-Fi and an LTE modem, enabling communication with online services via wireless or cellular networks. Full article

Floating offshore wind turbines (FOWTs) face complex aero-hydro-servo-elastic interactions that challenge conventional modeling, monitoring, and control. This review critically examines how artificial intelligence (AI) is being applied across four domains—design and surrogate modeling, structural health monitoring, control and operations, and digital twins—with explicit attention to uncertainty and reliability. Using PRISMA 2020 (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), a Scopus search identified 412 records; after filtering for articles, conference papers, and open access, 115 studies were analyzed. We organize the literature into a taxonomy covering classical supervised learning, deep neural surrogates, physics-informed and hybrid models, reinforcement learning, digital twins with online learning, and uncertainty-aware approaches. Neural surrogates accelerate coupled simulations; probabilistic encoders improve structural health monitoring; model predictive control and trust-region reinforcement learning enhance adaptive control; and digital twins integrate reduced-order physics with data-driven calibration for lifecycle management. The corpus reveals progress but also recurring limitations: simulation-heavy validation, inconsistent metrics, and insufficient field-scale evidence. We conclude with a bias-aware synthesis and propose priorities for future work, including shared benchmarks, safe RL with stability guarantees, twin-in-the-loop testing, and uncertainty-to-decision standards that connect model outputs to certification and operational risk. Full article

Omniphobic membranes have gained extensive attention for mitigating membrane wetting in robust membrane separation owing to the super-repulsion toward water and oil. In this study, a Teflon/PAI composite membrane with omniphobic characteristics was prepared by a vacuum-assisted dip-coating strategy on the PAI hollow fiber membrane. A series of characterizations on morphological structure, surface chemical composition, wettability, permeability, mechanical properties, and stability were systematically investigated for pristine PAI and Teflon/PAI composite membranes. Subsequently, the experiment was conducted to explore the oil–gas separation performance of membranes, with standard transformer oil containing dissolved gas as the feed. The results showed that the Teflon AF2400 functional layer was modified, and C-F covalent bonds were introduced on the composite membrane surface. The Teflon/PAI composite membrane exhibited excellent contact angles of 156.3 ± 1.8° and 123.0 ± 2.5° toward DI water and mineral insulating oil, respectively, indicating omniphobicity. After modification, the membrane tensile stress at break increased by 23.0% and the mechanical performance of the composite membrane was significantly improved. In addition, the Teflon/PAI composite membrane presented satisfactory thermal and ultrasonic stability. Compared to the previous membranes, the Teflon/PAI composite membrane presented a thinner Teflon AF2400 separation layer. Furthermore, the omniphobic membrane demonstrated anti-wetting performance by reaching the dynamic equilibrium within 2 h for the dissolved gases separated from the insulating oil. This suggests an omniphobic membrane as a promising alternative for oil–gas separation in monitoring the operating condition of oil-filled electrical equipment online. Full article

The determination of the elemental composition of minerals at mining enterprises is important at all stages of mineral processing. An evaluation of metrological characteristics achieved through the online analysis of lump, ore, charge feed, cake and slag materials on a conveyor belt is presented. Each implementation of the online XRF analysis at mining enterprises was preceded by laboratory studies, the development of measurement methods and the calibration of a specific XRF analyzer using standard reference samples for a specific concentration range of the monitored elements. In this work, typical application areas for monitoring the concentration of elements in rocks on conveyor belts are presented, as well as those solutions that made it possible to achieve the required measurement accuracy with an X-ray fluorescence analyzer in an online mode. Full article

To address challenges in holistic real-time condition monitoring of conventional wind tunnels—caused by large structural dimensions and complex parameter monitoring—this study proposes a wind tunnel condition monitoring surrogate model (POD-BPNN) integrating Proper Orthogonal Decomposition (POD) for data dimensionality reduction with Back Propagation Neural Networks (BPNNs). By implementing POD-based order reduction, the computational load for neural network training is significantly reduced while maintaining predictive accuracy through reduced-order data utilization. When applied to reconstruct stress/displacement fields in a wind tunnel test section and the flow field in its fan section, the POD-BPNN model demonstrated prediction errors below 5% when validated against finite element and computational fluid dynamics simulations, with three orders of magnitude improvement in computational efficiency. This methodology satisfies precision and real-time requirements for structural/fluid field monitoring in wind tunnels. When deployed with an existing health management system, online monitoring and predictive maintenance of the digital twin for the wind tunnel will be achievable. Full article

The ultra-high-frequency (UHF) detection method is highly accurate and has a fault localization function. At present, most gas-insulated switchgear (GIS) installations are equipped with online UHF monitoring devices to detect partial discharges. In order to ensure the accuracy of the detection results, UHF sensors need to be verified regularly. UHF sensors used for online monitoring are usually installed at the handhole of the GIS and cannot be removed. Measuring the laboratory verification indexes (e.g., equivalent height, dynamic range, etc.) of the sensors directly is very difficult. However, it is easier to measure S₁₁ of the sensor for verification and S₂₁ between it and the neighboring sensors by injecting power signals. Accordingly, this paper proposes a degradation identification method for GIS UHF sensors using a cross-comparison of S-parameters. When sensor sensitivity decreases, S₁₁ increases while S₂₁ decreases, both serving as effective indicators of performance degradation. In this study, the equivalent S-parameter network and the variation mechanisms of S₁₁ and S₂₁ during sensor verification were first analyzed. Normal and typically degraded sensor models were then constructed and coupled in different GIS structures for electromagnetic simulation. The simulation and on-site verification results show that S₁₁ is mainly affected by the sensor’s intrinsic performance and installation conditions at the inspection port, whereas S₂₁ is predominantly influenced by sensor performance and the propagation characteristics of the GIS structure. Through cross-comparison of S₁₁ and S₂₁ at corresponding positions across three phases, sensor aging or failure can be effectively identified, enabling rapid on-site verification without removing the sensors. The proposed method was successfully validated on actual GIS equipment at the China Southern Power Grid Research Institute. It exhibits high accuracy, efficiency, and strong engineering applicability, enabling the early detection of degraded sensors and providing valuable support for condition assessment and maintenance decision-making in GIS online monitoring systems. Full article

To enable real-time monitoring of particle mass flow rate in petrochemical pneumatic conveying systems, thereby facilitating process control optimization and improving energy efficiency, an online measurement system for petrochemical particle mass flow based on a non-intrusive electrostatic coupled capacitance sensor is developed. The measurement system determines particle flow velocity by analyzing electrostatic signals using a cross-correlation method, and calculates particle concentration by applying a pre-calibration that correlates capacitance signals with concentration values. These two parameters are then combined to calculate the real-time particle mass flow rate. The performance of the developed system is evaluated under different pipe diameters and particle concentration ranges, in both lab-scale and pilot-scale pneumatic conveying rigs. The obtained results show that the measurement system achieved a maximum relative error of 5.5% for mass flow measurements in the lab-scale 50 mm pneumatic conveying pipeline when the particle concentration range was between 2.04 kg/m³ and 6.43 kg/m³. As for the pilot-scale 100 mm pneumatic conveying, the maximum relative error of the particle concentration measurement was 3.6% when the particle concentration range was 30.98~68.87 kg/m³. These results demonstrate that the developed system has strong adaptability and reliability, highlighting its broad potential for industrial applications. Full article

Transformer oil ageing alters key physicochemical properties, notably the refractive index (RI), due to physical, particulate, and chemical changes. As a result, refractometric fibre-optic sensors have gained attention for enabling real-time monitoring and overcoming the limitations of traditional offline diagnostics. This study explores the use of a Fabry–Pérot phase-modulated fibre optic sensor (FISO FRI RI Sensor) for in-situ ageing detection in four industrial transformer oils: natural ester, synthetic ester, Nytro Bio 300X (vegetable-based), and Polaris GX (mineral-based). The oils were thermally aged under controlled conditions following degassing and drying. The sensor performance was evaluated using key metrics, including repeatability, thermal response, settling time, and linearity. Results show high repeatability (with standard deviations below 7 × 10⁻⁵ RIU and repeatability coefficients under 2 × 10⁻⁴ RIU), stable thermal response (~0.0004 RIU/°C), and strong thermal linearity (R² > 0.99) across all samples. Natural ester and Nytro Bio 300X exhibited the most stable and consistent sensor responses, while synthetic ester and mineral oils showed greater variability due to temperature-induced RI shifts. These findings demonstrate the reliability and precision of this Fabry–Pérot phase-modulated sensor for online transformer oil condition monitoring, with strong potential for integration into smart grid diagnostics. Full article

Adaptive forest and fire management in parks and protected areas is becoming increasingly complex as climate change alters the frequency and intensity of disturbances (wildfires, pest and disease outbreaks, etc.), while park visitation and the number of people living adjacent to publicly managed lands continues to increase. Evidence-based, climate-adaptive forest and fire management practices are critical for the responsible stewardship of public resources and require the continued availability of long-term ecological monitoring data. The US National Park Service has been collecting long-term fire monitoring plot data since 1998, and has continued to add monitoring plots, but these data are housed in databases with limited access and minimal analytic capabilities. To improve the availability and decision support capabilities of this monitoring dataset, we created the Trends in Forest Fuels Dashboard (TFFD), which provides an implementation framework from data collection to web visualization. This easy-to-use and updatable tool incorporates data from multiple years, plot types, and locations. We demonstrate our approach at Rocky Mountain National Park using the ArcGIS Online (AGOL) software platform, which hosts TFFD and allows for efficient data visualizations and analyses customized for the end user. Adopting interactive, web-hosted tools such as TFFD allows the National Park Service to more readily leverage insights from long-term forest monitoring data to support decision making and resource allocation in the context of environmental change. Our approach translates to other data-to-decision workflows where customized visualizations are often the final steps in a pipeline designed to increase the utility and value of collected data and allow easier integration into reporting and decision making. This work provides a template for similar efforts by offering a roadmap for addressing data availability, cleaning, storage, and interactivity that may be adapted or scaled to meet a variety of organizational and management use cases. Full article