Time-averaged Research Articles

On BASTA for Discrete-Time Queues

We address the issue of Bernoulli/Geometric arrivals see time-averages, BASTA, in discrete-time queues when using a variety of scheduling rules. It is well-known that BASTA/ASTA holds in the framework of a discrete-time process with an imbedded arrival process without regard to scheduling rules. However, when scheduling rules are invoked, it is not always clear whether arrivals see time averages as asserted by the general theory. We resolve this issue so that BASTA/ASTA holds regardless of the scheduling rule at play.

Research on tomato disease image recognition method based on DeiT

Tomatoes, globally cultivated and economically significant, play an essential role in both commerce and diet. However, the frequent occurrence of diseases severely affects both yield and quality, posing substantial challenges to agricultural production worldwide. In China, where tomato cultivation is carried out on a large scale, disease prevention and identification are increasingly critical for enhancing yield, ensuring food safety, and advancing sustainable agricultural practices. As agricultural production scales and the demand for efficient methodologies grows, traditional disease recognition methods no longer meet current needs. The agricultural sector's move towards more modern and scalable production methods necessitates more effective and precise disease recognition technologies to support swift decision-making and timely preventive actions. To address these challenges, this paper proposes a novel tomato disease recognition method that integrates the data-efficient image transformers (DeiT) model with strategies like exponential moving average (EMA) and self-distillation, named EMA-DeiT. By leveraging deep learning technologies, this method significantly improves the accuracy of disease recognition. The enhanced EMA-DeiT model demonstrated exemplary performance, achieving a 99.6 % accuracy rate in identifying ten types of tomato leaf diseases within the PlantVillage public dataset and 98.2 % on the Dataset of Tomato Leaves, which encompasses six disease types. In generalization tests, it achieved 97.1 % accuracy on the PlantDoc dataset and 97.6 % on the Tomato-Village dataset. Utilizing the improved DeiT model, a comprehensive tomato disease recognition system was developed, featuring modules for image collection, disease detection, and information display. This system facilitates an integrated process from image collection to intelligent disease analysis, enabling agricultural workers to promptly understand and respond to disease occurrences. This system holds significant practical value for implementing precision agriculture and enhancing the efficiency of agricultural production.

Gas Transfer Across Air‐Water Interfaces in Inland Waters: From Micro‐Eddies to Super‐Statistics

AbstractIn inland water covering lakes, reservoirs, and ponds, the gas exchange of slightly soluble gases such as carbon dioxide, dimethyl sulfide, methane, or oxygen across a clean and nearly flat air‐water interface is routinely described using a water‐side mean gas transfer velocity , where overline indicates time or ensemble averaging. The micro‐eddy surface renewal model predicts , where is the molecular Schmidt number, is the water kinematic viscosity, and is the waterside mean turbulent kinetic energy dissipation rate at or near the interface. While has been reported across a number of data sets, others report large scatter or variability around this value range. It is shown here that this scatter can be partly explained by high temporal variability in instantaneous around , a mechanism that was not previously considered. As the coefficient of variation in increases, must be adjusted by a multiplier that was derived from a log‐normal model for the probability density function of . Reported variations in with a macro‐scale Reynolds number can also be partly attributed to intermittency effects in . Such intermittency is characterized by the long‐range (i.e., power‐law decay) spatial auto‐correlation function of . That varies with a macro‐scale Reynolds number does not necessarily violate the micro‐eddy model. Instead, it points to a coordination between the macro‐ and micro‐scales arising from the transfer of energy across scales in the energy cascade.

Non-elastic time series fuzzy clustering for efficient analysis of industrial data sets

In line with the green transition and digitalization trends, there is an increasing need for effective performance and health monitoring of production machines. Machine operation footprints resonate in quantities such as supply current and pneumatic line pressure, providing valuable insights when analyzed. Examining these low-level signals allows for monitoring and identification of changes in machine operation over time. A critical step in analyzing machine data is clustering, requiring an effective method for calculating time series cluster centroids. This paper introduces Error in Aligned series (ERAL), a novel method that generates a time series centroid that distills the fundamental shape of the underlying datasets. ERAL employs a fuzzy clustering-inspired iterative process for temporal alignment and averaging, avoiding the pathological artifacts often introduced by popular time-warping methods. Our analysis shows that existing methods can create artificial spikes and plateaus, which ERAL mitigates; it remains faithful to the original data shape. Additionally, ERAL offers improvements in computational efficiency and prototype quality. We evaluate the method against Dynamic Time Warping (DTW)-based methods across various datasets, and apply it to a time series clustering task using a real-world industrial dataset. In combination with a fuzzy clustering algorithm, ERAL generates visually convincing clusters. By leveraging fuzzy membership concepts, it achieves robust and adaptable clustering outcomes that reflect real-world data complexity and ambiguity.

Artificial Intelligence-Assisted Daily Quality Control System for the Histologic Diagnosis of Gastrointestinal Endoscopic Biopsies: A 1-Year Experience.

Seegene Medical Foundation, one of the major clinical laboratories in South Korea, developed SeeDP, an artificial intelligence (AI)-based postanalytic daily quality control (QC) system that reassesses all gastrointestinal (GI) endoscopic biopsy (EB) slides for incorrect diagnoses. To review the operational records and clinical impact of SeeDP since its launch in March 2022. Operational records of SeeDP were retrieved for the period of March 1, 2022, to February 28, 2023. Among cases scanned during 40 working days (March 10, 2022, to May 4, 2022), all discordant cases encountered by 2 pathologists were reviewed. Cases of SeeDP-assisted revised diagnoses were collected and compared with cases recognized using conventional methods. Occasional scanner failures and various types of aberrant errors compromised QC coverage, resulting in the scanning of only 67.7% (572 254 of 844 906) of all EB slides submitted and 0.8% of the scanned slides being further excluded from the AI analysis. The AI predictions differed from the pathologists' diagnoses in 42 760 of the 557 672 gastrointestinal EB slides (7.7%) successfully assessed by the AI models; however, a detailed review of discordant slides revealed that true misdiagnosis accounted for only 5.5% (25 of 454) of the disagreements. Compared with conventional error recognition methods, SeeDP detected more misdiagnoses (7 versus 14) within a significantly shorter time (average, 3.6 versus 38.7 days; P < .001), including 1 signet ring cell carcinoma initially diagnosed as gastritis. AI-based daily QC systems are plausible solutions to guarantee high-quality pathologic diagnosis by enabling rapid detection and correction of misdiagnosis.

Fast Convex Optimization via Time Scale and Averaging of the Steepest Descent

In a Hilbert setting, we develop a gradient-based dynamic approach for fast solving convex optimization problems. By applying time scaling, averaging, and perturbation techniques to the continuous steepest descent (SD), we obtain high-resolution ordinary differential equations of the Nesterov and Ravine methods. These dynamics involve asymptotically vanishing viscous damping and Hessian-driven damping (either in explicit or implicit form). Mathematical analysis does not require developing a Lyapunov analysis for inertial systems. We simply exploit classical convergence results for SD and its external perturbation version, then use tools of differential and integral calculus, including Jensen’s inequality. The method is flexible, and by way of illustration, we show how it applies starting from other important dynamics in optimization. We consider the case in which the initial dynamic is the regularized Newton method, then the case in which the starting dynamic is the differential inclusion associated with a convex lower semicontinuous potential, and finally we show that the technique can be naturally extended to the case of a monotone cocoercive operator. Our approach leads to parallel algorithmic results, which we study in the case of fast gradient and proximal algorithms. Our averaging technique shows new links between the Nesterov and Ravine methods. Funding: The research of R.I. Boţ and D.-K. Nguyen was supported by the Austrian Science Fund (FWF), projects W 1260 and P 34922-N.

The benefit of reduced serum phosphate levels depends on patient characteristics: a nationwide prospective cohort study.

In cohort studies of hyperphosphatemic hemodialysis patients, reduced serum phosphate levels have been linked to a lower mortality risk. To investigate whether this benefit is influenced by patient characteristics, we calculated the number needed to be exposed (NNE), stratified by patient characteristics. In this 9-year prospective cohort study using the nationwide Japanese registry, we enrolled 78256 hemodialysis patients aged 18years or older. We investigated the relationship between time-averaged (TA) phosphate levels and mortality due to cardiovascular disease (CVD) using Cox proportional models. We estimated the 1-year NNE for CVD death in patients with baseline serum phosphate levels ≥6.0mg/dL and exposure to TA phosphate levels decreasing to 3.5-<5.0mg/dL using mixed-effects Poisson models. The hazard ratio of CVD mortality decreased linearly with lower serum TA phosphate levels in those with prior atherosclerotic CVD (ACVD) or diabetic nephropathy (DN) but plateaued with serum phosphate <5.0mg/dL in those without. The hazard ratios (95% confidence interval) for phosphate ≥7.0mg/dL compared with 3.5-<3.9mg/dL were 1.58 (1.38-1.81) in those with prior ACVD, 1.91 (1.68-2.17) in those without, 1.87 (1.63-2.16) in those with DN and 1.65 (1.46-1.87) in those without. However, the NNE for one more person to benefit (NNEB) for CVD death was lower in patients with a history of ACVD than in those without (61 vs 118). Patients with DN had lower NNEB than those without (69 vs 113). In patients with TA albumin ≥3.8g/dL, older patients had lower NNEB, while patients with TA albumin <3.45g/dL showed no benefit in some groups, including the elderly. The benefit of intensive phosphate management may be pronounced in patients with a history of ACVD or DN. A comprehensive approach that considers both age and nutritional status may be necessary when managing serum phosphate levels.

Calibration and Release of Magnetograms/Dopplergrams Obtained at the Mt. Wilson 150-Foot Tower Telescope (MWO)

The Mt. Wilson Observatory archive of observations of solar disk magnetic fields, Doppler velocities, and spectral line intensities is a resource for studying the Sun’s state from 1967 to 2013. Instrument changes/upgrades during this time must be considered when interpreting this record. Portions of this record have been previously released. This publication documents the data record in order to allow its independent interpretation. The archive is available through two directory trees which can be accessed at http://sha.stanford.edu/mwo/msm.html. The calibration of the observations is impacted by the solar surface convective flows, which produce offsets for both differential rotation and meridional circulation functions. The effects of these offsets have been reduced in this and other publications by temporal averaging.

Electron Dissipation and Electromagnetic Work

AbstractWith the increase in technical capabilities of computer simulation in recent years, it has become feasible to quantify the degradation of fluid scale plasma and electromagnetic energies in favor of increases of internal energies. While it is understood that electromagnetic energy can be exchanged with fluid scale velocities, it is the pressure strain interaction that exchanges energy between fluid motions and internal energy. Here using simulations of both turbulence and reconnection we show that for electrons, the pressure strain and electromagnetic work are closely related and are frequently comparable when appropriate time and spatial averaging is applied. Otherwise, the instantaneous spatial averaged pressure strain and electromagnetic work are nearly equal for slowly evolving systems, like the reconnection case, while they differ significantly in rapidly evolving systems, like the turbulence case. This clarifies the relationship between these two quantities, which are each frequently used as measures of dissipation.

AVA-YOLO: image-based multiscale feature fusion enhanced perception model for snow avalanche detection

Abstract The global climate change has led to frequent occurrences of snow avalanche disasters. However, the significant variations in scale and shape during the avalanche process, and complex background imagery pose significant challenges to automated detection efforts. There is an urgent need to combine advanced deep learning technology to research automatic detection and recognition of avalanches in the field. In this paper, a novel deep learning model based on YOLOv8 improved multi-scale detection called AVA-YOLO is proposed to solve this problem. In AVA-YOLO, a key component, AKA (AKConv Combined Attention) module was designed and developed. This module combines the deformable convolutional properties of AKConv with the state-of-the-art self-attention module Exponential Moving Average, aiming to better perceive the feature map information of different shaped avalanches and to enhance the global relevance, thus improving the utilization of the information. Secondly, a new multi-scale sensing network structure was designed by increasing the number of detection heads to four and introducing the AKA module into the key positions of the network, while the association between model layers was newly designed to enhance the fusion of shallow and deep information to improve the detection accuracy. Experimental results demonstrated the effectiveness of AVA-YOLO, achieving 95.7% mAP50 and 75.6% mAP50:95 detection accuracies, as well as an F1 score of 0.92. Finally, a number of experiments were conducted to demonstrate the superior performance of the proposed model in comparison to other versions of YOLO, which will further exploit the potential of webcams as an underutilized technical capability in snow avalanche intelligence and portable monitoring.

Flow induced form: Shape optimization of bridge piers using CFD analyses and adjoint method

This study investigates the optimization of cylindrical bridge pier geometry using a relatively new technique involving Computational Fluid Dynamics (CFD) and adjoint shape optimization methods. Initially, Reynolds Averaged Navier-Stokes (RANS) equations were used in CFD models, integrated with an adjoint algorithm to reshape the pier and reduce bed shear stress. This approach is hereinafter called the “Flow induced Form”, FiF, aiming to minimize shear stress on the bed while maintaining the pier's cross-sectional and projected areas in the flow direction. Subsequently, Large Eddy Simulation (LES) transient analyses assessed the performance of the optimized geometry. Results show a reduction of approximately 5% in the time average of maximum shear stresses on the bed and a reduction of around 50% in instantaneous maximum values compared to the original pier design. These improvements were achieved with minimal changes to the cross-sectional area and projected area of the original geometry (∼2%), considering that the load-bearing capacity of the bridge pier.

Focus-SLAM: A visual Monocular SLAM

Abstract This research presents a novel simultaneous localization and mapping algorithm, called Focus-SLAM, which simulates human navigation strategies by synthesizing a optical saliency model, SalNavNet, in a traditional single-shot SLAM paradigm. SalNavNet introduces an innovative design that incorporates a correlation module and an adaptive Exponential Moving Average (EMA) component, effectively addressing the prevalent center bias issue found in contemporary saliency models. As a result, the system enhances target fixation by accentuating salient features more effectively. Extensive experimental evaluations, conducted across a spectrum of indoor environments exhibiting diverse luminosity conditions. These findings attest to the efficacy of Focus-SLAM in enhancing both the accuracy and efficiency of SLAM operations under challenging real-world conditions.

Surfacing of Latent Time Memories Supports the Representational Basis of Timing Behavior in Mice

Animals can adapt their reward expectancy to changes in delays to reward availability. When temporal relations are altered, associative models of interval timing predict that the original time memory is lost due to the updating of the underlying associative weights, whereas the representational models render the preservation of the original time memory (as previously demonstrated in the extinction of conditioned fear). The current study presents the critical test of these theoretical accounts by training mice with two different intervals in a consecutive fashion (short → long or long → short) and then testing timing behaviors during extinction where neither temporal relation is in effect. Mice that were trained with the long interval first clustered their anticipatory responses around the average of two intervals (indirect higher-order manifestation of two memories in the form of temporal averaging), whereas mice trained with the short interval first clustered their responses either around the short or long interval (direct manifestation of memory representations by their independent indexing). We assert that the original memory representation formed during training with the long interval “metrically affords” the integration of subsequent experiences with a shorter interval, allowing their co-activation during extinction. The original memory representation formed during training with the short interval would not metrically afford such integration and thus result in the formation of a new (mutually exclusive) time memory representation, which does not afford their co-activation during extinction. Our results provide strong support for the representational account of interval timing. We provide a new theoretical account of these findings based on the “metric affordances” of the original memory representation formed during training with the original intervals.

Research on airfoil surface flow structure testing based on Tuft velocity measurement method

Fluorescent tuft visualization technology, as a practical and convenient technique, has attracted widespread attention. In this study, a novel fluorescent tuft velocity measurement method is developed based on existing technologies. This method involves the recognition of tuft deflection angles, image transformation, block processing, temporal averaging, and post-visualization processing. Additionally, quantitative measurement of flow field velocity is achieved through tuft calibration. The study begins by comparing tufts of different materials, examining both fluorescent and deflection characteristics to determine optimal tuft parameters. The impact of tuft length on deflection characteristics is also investigated. Finally, the cotton tuft with a length of 7 mm and a diameter of about 0.1 mm was obtained as the best tuft sought in this experiment. Subsequently, at Reynolds number Re = 1.7 × 105, airfoil surface flow field visualization and quantitative analysis are conducted, exploring the relationship between airfoil surface flow structure and aerodynamics at different angles of attack. Finally, a comparative study is conducted between tuft velocity results and oil flow visualization experiments, as well as particle image velocimetry experiments, confirming the feasibility of the fluorescent tuft velocity measurement method.

Reducing Hospital Length of Stay and Hepatic Artery Thrombosis Rates for Children Receiving a Liver Transplant: A Single-Center Experience From 2000 to 2021.

Pediatric liver transplantation is a very resource-intensive therapy. This study aimed to identify the changes made between two epochs of management and analyze their influence on length of stay (LOS). Data from a single center were obtained from the liver transplant and Pediatric Intensive Care Unit (PICU) databases for 336 transplants (282 children) performed between 2000 and 2021. Transplants were analyzed in two epochs, before and after July 2012, representing a change in postoperative anticoagulation management. Differences in graft recipient demographics and perioperative management factors were compared between epochs. Multivariate regression was performed to identify the complications that correlated most strongly with hospital LOS. There was a difference in hospital LOS between Epoch 1 (Median = 31.7 days) and Epoch 2 (Median = 26.3 days) (p < 0.001), but not in PICU LOS (E1 Median = 7.3 days, E2 Median = 7.4 days; p = 0.792). Epoch 2 saw increased use of split grafts (60.6% of total), decreased pediatric end-stage liver disease (PELD) score at transplant (Average = 16.7; p < 0.001), decreased invasive ventilation time (Average = 4.48 days; p < 0.001), and decreased hepatic artery thrombosis (HAT) rates (E1 = 14.4%, E2 = 4.3%; p < 0.001) without an associated increase in bleeding rates. Hospital LOS has reduced in Epoch 2 due to refinements in intraoperative and postoperative management. There is increased emphasis on early extubation and increased use of noninvasive ventilatory techniques in Epoch 2. Split grafts have effectively expanded our graft donor pool and reduced transplant waitlist times.

Offline constrained reinforcement learning for batch-to-batch optimization of cobalt oxalate synthesis process

The cobalt oxalate synthesis, a batch process, plays a crucial role in the refinement of cobalt metal. The mean particle size of cobalt oxalate is a critical indicator that reflects product quality. However, excessive ammonium oxalate solution flow can heighten waste disposal costs in the production process. To address these issues, we propose a novel offline reinforcement learning (RL) algorithm that guarantees compliance with constraints in the cobalt oxalate synthesis process, utilizing exclusively static datasets. This method employs cost critic networks to assess costs, transforming the constrained optimization problem into an unconstrained one by introducing Lagrangian multipliers. We use exponential moving average (EMA) to optimize the update of proportional integral derivative (PID) control multipliers, reduce overshoot and oscillation in the control process, and thus improve the overall stability of the system. Furthermore, to optimize algorithm performance, a deep residual network (DResNet) is integrated into the policy network. Experimental results indicate that the algorithm’s optimization policy performs significantly better under constraints.

Heating mode transitions in capacitively coupled CF4 plasmas at low pressure

Abstract Capacitively coupled plasmas operated in CF4 at low pressure are frequently used for dielectric plasma etching. For such applications the generation of different ion and neutral radical species by energy dependent electron impact ionization and dissociation of the neutral background gas is important. These processes are largely determined by the space and time dependent electron energy distribution function and, thus, by the electron power absorption dynamics. In this work and based on a particle-in-cell/Monte Carlo collision model, we show that the electron heating mode in such plasmas is sensitive to changes of the gap at a constant pressure of 3 Pa. At a gap of 1.5 cm, the dominant mode is found to be a hybrid combination of the Drift-Ambipolar (DA) and the α-mode. As the gap is increased to 2 cm and 2.5 cm, the bulk power absorption and ambipolar power absorption decreases, and the DA mode decays. When the gap reaches 3 cm, the α-mode becomes more prominent, and at a gap of 3.75 cm the α-mode is dominant. These mode transitions are caused by a change of the electronegativity and are found to affect the discharge characteristics. The presence of the DA-mode leads to significant positive electron power absorption inside the bulk region and negative power absorption within the sheaths on time average, as electrons are accelerated from the bulk towards the collapsed sheath. The heating mode transitions result in a change from negative to positive total electron power absorption within the sheaths as the gap increases. When accounting for secondary electron emission, the transition of the heating mode can occur at shorter gaps due to the enhanced plasma density and decreased electronegativity.

HL‐2A's ELM cycle simulations by integrating BOUT++'s drift MHD and transport code

AbstractA new integrating model has been developed to couple tokamak edge multiscale magnetohydrodynamic (MHD) events and transport simulations, such as edge‐localized mode (ELM) cycles. As a proof of principle, we first start from a set of three‐field two‐fluid model equations, which includes the pressure, current, and vorticity. The equations are separated into the slowly evolving part of the axisymmetric component by taking a time average of the axisymmetric component. The time‐averaged fluxes, which are quadratic in fluctuating quantities, act as driven terms for the time‐averaged axisymmetric quantities that determine the plasma transport, and therefore the large‐scale evolution of the plasma profiles. Then the HL‐2A's ELM cycles are simulated using the model. Good agreements of ELM size and pedestal recovery time have been achieved for the solutions obtained from the coupled simulation compared with experiment. For one ELM cycle simulation, the coupled code can achieve a speedup of a factor of up to 30 over standalone code.

Remaining useful life prediction for a cracked rotor system via moving feature fusion based deep learning approach

This study proposes a novel approach for predicting the remaining useful life (RUL) of a cracked rotor system by using a moving fusion gated recurrent unit (MFGRU) model. Firstly, 12 multi-domain features were extracted from the raw collected vibration signals, which were then fused and distributed adaptively using bidirectional exponential moving average (EMA) and multi-head attention (MHA). Then a bidirectional gated recurrent unit network combined with moving feature fusion method was proposed to capture the long-term dependence relationship in the monitoring data of the cracked rotor for RUL prediction. Finally, the model’s performance was validated through accelerated life experiments, with the measured values of root mean square error (RMSE) and mean absolute error (MAE) below 3.17 and 2.60, respectively. This study offers a dynamic RUL prediction method with certain significance and valuable reference for designing deep learning models.

Gendered financial & nutritional benefits from access to pay-as-you-go LPG for cooking in an informal settlement in Nairobi, Kenya

This study investigates the association between adoption of pay-as-you-go (PAYG) liquefied petroleum gas (LPG), an emerging alternative to full cylinder LPG, and women's economic empowerment in an informal settlement in Nairobi, Kenya. From December 2021-January 2022, 293 customers of a PAYG LPG company (PayGo Energy) were surveyed on their cooking patterns, financial savings and shifts in dietary behaviors following uptake of the technology. Among PayGo Energy customers that previously cooked only with polluting fuels (N = 78; 27 % of customers), daily cooking time was reduced by an average of 42 min/day; 82 % of PayGo Energy customers that previously cooked with full cylinder LPG (N = 216; 73 % of customers) also decreased their cooking time (average of 20 min/day) when switching to PAYG LPG. The majority (58 %; N = 70) of female household heads took on additional employment after switching to PAYG LPG, compared with 36 % (N = 55) of females living in male-headed households. Among female household heads, the proportion of informal sector workers earning wages on an irregular (71 %) or daily basis (61 %) that took on new income-generating activities after transitioning to PAYG LPG was over 20 % higher than those earning monthly salaries (39 %). Increased dietary diversity and consumption of protein-rich foods (legumes, meat, fish) from cooking with PAYG LPG was reported by 15 % of female household heads compared with 5 % of those living in male-headed households. While nearly three-quarters (73 %) of PayGo Energy customers would recommend the service to others because of the added convenience it provides, only one-third (29 %) reported associated health benefits as a key reason for promoting use of PAYG LPG to community members. Female household heads were more likely than non-household heads to be socioeconomically empowered when adopting PAYG LPG, illustrating that women's agency may influence the associated benefits of clean energy transitions. Nonetheless, the time savings reported by nearly all women who switched to PAYG LPG for cooking suggests that promoting the increased convenience of cooking with PAYG LPG may be useful for accelerating its adoption..