Flow Classification Research Articles

On the Scaling of Transport Phenomena at a Monotonously Changing Hydraulic Conductivity Field

Monotonously stratified porous medium, where the layered medium changes its hydraulic conductivity with depth, is present in various systems like tilled soil and peat formation. In this study, the flow pattern within a monotonously stratified porous medium is explored by deriving a non-dimensional number, Fhp, from the macroscopic Darcian-based flow equation. The derived Fhp theoretically classifies the flow equation to be hyperbolic or parabolic, according to the hydraulic head gradient length scale, and the hydraulic conductivity slope and mean. This flow classification is explored numerically, while its effect on the transport is explored by Lagrangian particle tracking (LPT). The numerical simulations show the transition from hyperbolic to parabolic flow, which manifests in the LPT transition from advective to dispersive transport. This classification is also applied to an interpolation of tilled soil from the literature, showing that, indeed, there is a transition in the transport. These results indicate that in a monotonously stratified porous medium, very low conducting (impervious) formations may still allow unexpected contamination leakage, specifically for the parabolic case. This classification of the Fhp to the flow and transport pattern provides additional insight without solving the flow or transport equation only by knowing the hydraulic conductivity distribution.

RGB Mapping: A Dynamic Approach for Flow Pattern Identification and Classification

Thermal-hydraulic modeling and simulation are heavily dependent on the knowledge of flow regimes, which sort multiphase flow patterns into classifications with static boundaries that display characteristic features of flow patterns and phase distributions. Flow regime maps heretofore have been primarily developed mainly for vertical or horizontal pipe orientations by capturing images of flows at varying conditions and assigning them to different flow regimes based on visualization studies. As such, this method is inherently subjective. Furthermore, classifying two-phase flows into regimes with static boundaries is not physical, as two-phase flows undergo gradual changes as they transition from one regime to another. Therefore, such a static approach is not only unphysical but also inaccurate, resulting in the erroneous analysis of two-phase flows. Additionally, subjective observations at such flow conditions are prone to disagreement and inconsistencies in labeling. These difficulties are exacerbated in inclined two-phase flows. In this study, a novel two-phase flow pattern map employing the dynamic red, green, and blue (RGB) color model is presented. The model contains the embedded information of three normalized time-averaged statistics from signals obtained by a dual-ring impedance meter. A database of impedance meter signals from a 25.4-mm-inner-diameter adiabatic air-water inclinable test facility is used to compute statistics for each experimental flow condition. Multiple statistical measures are evaluated, and a principal component analysis is performed for feature selection. By mapping each parameter to intensities of red, green, and blue, a dynamic RGB flow pattern map can be produced in which groups of similar colors are representative of various flow regimes and the dynamic transition characteristics between the flow regimes are characterized physically. Therefore, this new method makes viable the observation of how flow patterns gradually change with respect to angle, as well as the fuzzification or quantification of the uncertainty surrounding experimental transition boundaries.

ФИНАНСОВОЕ УПРАВЛЕНИЕ ДЕНЕЖНЫМИ ПОТОКАМИ НЕФТЯНОЙ КОМПАНИИ ПАО НК "РОСНЕФТЬ"

The article analyzes the classification of cash flows of an oil company. Cash flow management and assessment. The company's cash flows were assessed using direct and indirect methods. An economic assessment of the effectiveness of operational cash flow management of PJSC NK Rosneft is given. The problems in managing cash flows of an oil company are analyzed.

Accurate and efficient elephant-flow classification based on co-trained models in evolved software-defined networks

Accurate early classification of elephant flows (elephants) is important for network management and resource optimization. Elephant models, mainly based on the byte count of flows, can always achieve high accuracy, but not in a time-efficient manner. The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks (SDNs) to achieve a better resource efficiency. This paper addresses this situation by combining co-training and Reinforcement Learning (RL) to enable a closed-loop classification approach that divides the entire classification process into episodes, each involving two elephant models. One predicts elephants and is retrained by a selection of flows automatically labeled online by the other. RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase. Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%, and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.

The effect of regional import shocks on job flows in Japanese manufacturing establishments

Using a dataset of all Japanese manufacturing establishments in 2006 and 2016 and industry- and region-level import shocks, this study examined the impact of the China Shock on the job flows of establishments. Regression analyses found that the industry- level import shock increased the probability of exiting in only a few groups. Interestingly, small establishments in single-unit firms located in non-agglomerated regions adopted a hibernation strategy to cope with the region-level import shock. Furthermore, surviving establishments in this category accelerated both job creation and destruction in response to these two import shocks, thereby amplifying the overall magnitude of job flows. These results became observable owing to the use of establishment-level observations and a detailed classification of job flows.

Radio Frequency-Enabled Cerebral Blood Flow Monitoring and Classification Using Data Augmentation and Machine Learning Techniques

Radio Frequency-Enabled Cerebral Blood Flow Monitoring and Classification Using Data Augmentation and Machine Learning Techniques

Neural network based two-phase flow classification in a vertical narrow rectangular channel

Flow Regimes in a vertical narrow rectangular channel of cross-section 20×1cm2 are investigated up-to wispy annular flow using a dense test matrix and double sensor conductivity probe at section mid point. The data from the probe is used to calculate void fraction, velocity of interfaces, and chord length of various flow structures. A five unit self organizing neural network is used to identify various flow regimes by using single point geometrical data of flow structures. Six separate flow regimes are found to exist. A new flow regime is identified and is called rolling-wispy flow. A discussion on boiling crisis is given regarding this flow regime. The resultant flow regime map is compared with various existing maps.

A Numerical Study on the Flow Field and Classification Performance of an Industrial-Scale Micron Air Classifier under Various Outlet Mass Airflow Rates

In this study, the gas−particle flow field in a real-size industrial-scale micron air classifier manufactured by Phenikaa Group using 3D transient simulations with the FWC-RSM–DPM (Four-Way Coupling-Reynold Stress Model-Discrete Phase Model) in ANSYS Fluent 2022 R2 and with the assistance of High-Performance Computing (HPC) systems is explored. A comparison among three coupling models is carried out, highlighting the significant influence of the interactions between solid and gas phases on the flow field. The complex two-phase flow, characterized by the formation of multiple vortices with different sizes, positions, and rotation directions, is successfully captured on the real-size model of the classifier. Additionally, analyzing the effects of the vortices on the flow field provides a comprehensive understanding of the gas–solid flow field and the classification mechanism. The effect of the outlet mass airflow rate is also investigated. The classifier’s Key Performance Indicators (KPIs: d50, K, η, ΔP) and the constrained condition of the particle size distribution curve of the final product are used to evaluate the classification efficiency. The contributions of this work are as follows: (i) a simulation analysis of a real-size industrial-scale classifier is conducted that highlights its advantages over a lab-scale one; (ii) a comparison is conducted among three coupling models, showing the advancement of four-way coupling in providing accurate results for simulations of interactions between the gas phase and particles; and (iii) the particle size distribution curve performances of a classified product under different simulation models and outlet airflow rates are addressed, from which optimal parameters can be selected in the design and operation processes to achieve the required efficiency of an air classifier.

What Is Relatively Permanent? Flow Regimes of Arizona Streams within the Context of the 2023 Conforming Rule on the Revised Definition of “Waters of the United States”

The classification of stream flow regimes has been a subject of study for over a half century in the fields of hydrology, geomorphology, ecology, and water resources management. But with the most recent Supreme Court decision on jurisdictional Waters of the United States (WOTUS) and the 2023 Conforming Rule, the answer to the question of which waters are relatively permanent has increased in importance and urgency. One state where this question is salient is Arizona, where approximately 95% of its streams are nonperennial. In this study, we use long-term (>30 years) daily discharge records from Arizona to assess semi-natural flow regimes of arid streams within the context of the 2023 Conforming Rule. Using flow percentile distributions, we distinguished flow permanency—ephemeral vs. intermittent vs. perennial—for 70 stream reaches distributed throughout the state. Ephemeral streams had a median flow of 0 cms and a 75th percentile flow permanence less than 25% (i.e., less than 3 months of flow for every 7.5 out of 10 years). On the other end of the spectrum, perennial streams had a 90th percentile flow permanence of 100%. In the middle, intermittent streams had a 75th percentile flow permanence greater than 25% and a 90th percentile flow permanence less than 100%. We also assessed the effect of the recent megadrought (since 1994) on flow permanency. As a result of the megadrought, four perennial streams transitioned to intermittent, four intermittent streams transitioned to ephemeral, and one perennial stream became ephemeral. The flow classification we present here is specific to Arizona streams but could be useful to other arid regions seeking to answer the question of which streams are relatively permanent in a typical year.

Effectiveness of different tirofiban administration times in patients with no-reflow myocardial infarction during percutaneous coronary intervention.

To compare the effectiveness of different tirofiban administration time windows in patients with no-reflow myocardial infarction (MI) during percutaneous coronary intervention (PCI). This single centre retrospective observational study included patients with no-reflow MI, undergoing PCI at the Hanyang Hospital affiliated to Wuhan University of Science and Technology from March 2020 to May 2023. All patients were administered tirofiban. Patients who received tirofiban with postinterventional thrombolysis in myocardial infarction (TIMI) flow ≥ 1 were grouped as Group-I, and patients who were directly given tirofiban through the guiding catheter without forward blood flow were grouped as Group-II. TIMI blood flow classification, levels of cardiac troponin T (cTnT) and creatine kinase isoenzyme MB (CK-MB), incidence of complications and major adverse cardiovascular events (MACE) in the two groups before and after the treatment were statistically analyzed. A total of 156 patients were included in this study, including 79 patients in Group-I and 77 patients in Group-II. There was no significant difference in the baseline data between the two groups (P>0.05). After treatment, TIMI blood flow classification of the two groups improved and was significantly better in Group-I compared to Group-II (P<0.05). After treatment, levels of Serum cTnT and CK-MB in the two groups decreased, and were significantly lower in Group-I than in Group-II (P<0.05). There was no significant difference in the incidence of complications between Group-I (3.80%) and Group-II (6.49%) (P>0.05). The incidence of MACE in Group-I (3.80%) was lower than that in Group-II (12.99%) (P<0.05). Compared with the direct application of tirofiban, tirofiban given when TIMI Grade≥ 1 for patients with no-reflow MI during PCI can more effectively regulate the blood flow status of target vessels, reduce myocardial injury, and reduce the risk of MACE.

Empirical evaluation of feature selection methods for machine learning based intrusion detection in IoT scenarios

This paper delves into the critical need for enhanced security measures within the Internet of Things (IoT) landscape due to inherent vulnerabilities in IoT devices, rendering them susceptible to various forms of cyber-attacks. The study emphasizes the importance of Intrusion Detection Systems (IDS) for continuous threat monitoring. The objective of this study was to conduct a comprehensive evaluation of feature selection (FS) methods using various machine learning (ML) techniques for classifying traffic flows within datasets containing intrusions in IoT environments. An extensive benchmark analysis of ML techniques and FS methods was performed, assessing feature selection under different approaches including Filter Feature Ranking (FFR), Filter-Feature Subset Selection (FSS), and Wrapper-based Feature Selection (WFS). FS becomes pivotal in handling vast IoT data by reducing irrelevant attributes, addressing the curse of dimensionality, enhancing model interpretability, and optimizing resources in devices with limited capacity. Key findings indicate the outperformance for traffic flows classification of certain tree-based algorithms, such as J48 or PART, against other machine learning techniques (naive Bayes, multi-layer perceptron, logistic, adaptive boosting or k-Nearest Neighbors), showcasing a good balance between performance and execution time. FS methods' advantages and drawbacks are discussed, highlighting the main differences in results obtained among different FS approaches. Filter-feature Subset Selection (FSS) approaches such as CFS could be more suitable than Filter Feature Ranking (FFR), which may select correlated attributes, or than Wrapper-based Feature Selection (WFS) methods, which may tailor attribute subsets for specific ML techniques and have lengthy execution times. In any case, reducing attributes via FS has allowed optimization of classification without compromising accuracy. In this study, F1 score classification results above 0.99, along with a reduction of over 60% in the number of attributes, have been achieved in most experiments conducted across four datasets, both in binary and multiclass modes. This work emphasizes the importance of a balanced attribute selection process, taking into account threat detection capabilities and computational complexity.

Experimental and simulation study of flow field characteristics of a disturbing rotary centrifugal air classifier

In the present study, a new disturbing rotary centrifugal classifier was designed and a method was proposed to characterize its pressure signals. The models for evaluating the classification effects were established based on multi-zones. It was found that the flow field, processing parameters, and material properties significantly affected on the particle force and classification performance. The pressure with high amplitude and energy analyzed using Hilbert-Huang Transform (HHT) concentrated within 200–400 Hz, and the screening cage facilitated energy transfer to lower frequency bands. Correlation analysis demonstrated the correlation coefficients (A5 and A6) between adjacent measurement points increased with frequency, further enhanced by the screening cage. The flow field of simulation was characterized by a low-pressure, high-velocity gradient at the center and a high-pressure, low-velocity gradient at the sidewalls and the gas velocity and flow for experiment achieved the maximum at z = 0.50 m, while the magnitude of pressure change was significant at the inlet end and discharge end, where the velocity gradient and pressure drop had the same trend and small errors for experiments and simulation. The flow field affected the competing relationship between centrifugal and drag forces on the particles with different kinematic behavior during the classification process. The errors between the predicted values of the theoretical, simulated, and experimental cut particle size models and the true values were all within 10 %, based on which the multi-zone classification efficiency model applicable to the disturbing centrifugal classifier was established, and it was found that the models all had high prediction accuracy and suitability. The article delves into the correlation between pressure signals, flow field characteristics and particle behavior, aiming to provide an important reference for the internal flow field and classification performance of air classifiers.

Incremental federated learning for traffic flow classification in heterogeneous data scenarios

This paper explores the comparative analysis of federated learning (FL) and centralized learning (CL) models in the context of multi-class traffic flow classification for network applications, a timely study in the context of increasing privacy preservation concerns. Unlike existing literature that often omits detailed class-wise performance evaluation, and consistent data handling and feature selection approaches, our study rectifies these gaps by implementing a feed-forward neural network and assessing FL performance under both independent and identically distributed (IID) and non-independent and identically distributed (non-IID) conditions, with a particular focus on incremental training. In our cross-silo experimental setup involving five clients per round, FL models exhibit notable adaptability. Under IID conditions, the accuracy of the FL model peaked at 96.65%, demonstrating its robustness. Moreover, despite the challenges presented by non-IID environments, our FL models demonstrated significant resilience, adapting incrementally over rounds to optimize performance; in most scenarios, our FL models performed comparably to the idealistic CL model regarding multiple well-established metrics. Through a comprehensive traffic flow classification use case, this work (i) contributes to a better understanding of the capabilities and limitations of FL, offering valuable insights for the real-world deployment of FL, and (ii) provides a novel, large, carefully curated traffic flow dataset for the research community.

Acupuncture combined with gonadotropin-releasing hormone agonists improves endometrial receptivity and pregnancy outcome in patients with recurrent implantation failure of in vitro fertilization-embryo transfer.

Gonadotropin-releasing hormone agonists (GnRHa), combined with other auxiliary treatments, can improve pregnancy outcomes in in vitro fertilization-embryo transfer (IVF-ET). This research investigated the effect of acupuncture combined with GnRHa in patients with recurrent implantation failure (RIF) of IVF-ET. A total of 164 patients who intended to undergo frozen-thawed embryo transfer after RIF of IVF-ET were selected for experiments and then divided into the control (received conventional hormone replacement therapy (HRT) for endometrial preparation) and study groups (received a combination of acupuncture, GnRHa, and HRT for endometrial preparation) (n = 82). Endometrial thickness (EMT), endometrial morphological classification, submucosal uterine blood flow classification, clinical pregnancy rate, embryo implantation rate, and early abortion rate for each transfer cycle were compared between the two groups. EMT of the study group was higher than that of the control group 1 day before transfer. There were more patients with linear endometrium (A + B type) in the study group on the day of endometrial transformation than in the control group. The number of patients with type I submucosal uterine blood flow in the study group was decreased and the number of patients with type III was increased compared with the control group on the day of endometrial transformation. The clinical pregnancy rate and embryo implantation rate of the study group were higher than those of the control group. Acupuncture combined with GnRHa improves the endometrial receptivity of patients with RIF of IVF-ET, thereby increasing clinical pregnancy rates and improving pregnancy outcomes.

High-Performance Flow Classification of Big Data Using Hybrid CPU-GPU Clusters of Cloud Environments

High-Performance Flow Classification of Big Data Using Hybrid CPU-GPU Clusters of Cloud Environments

BPNN‐based flow classification and admission control for software defined IIoT

AbstractFlow admission control (FAC) aims to efficiently manage the service requests while maximizing the network utilization. With multiple connection requests, access delay or even service interruption may occur. This paper proposes a novel FAC approach to reduce the contention between the end nodes and ensure high utilization of the networking resources for software defined IIoT. First, incoming flows are classified into different priorities using back propagation neural network based on selected features representing the current network status. Second, with the designed flow admission policies, bandwidth and buffer size are estimated with stochastic network calculus model. Finally, the thresholds of the proposed FAC scheme are dynamically decided based on the above two parameters. Various flows are admitted or rejected via the proposed FAC to maintain real time processing. Unlike traditional FAC schemes rely on static priority systems, the proposed scheme leverages machine learning technique for dynamic flow prioritization and the stochastic network calculus model for precise estimation. Computer simulation reveals that the proposed scheme accurately classifies the flows, and substantially decreases the transmission delay and improves the network utilization compared to the existing FAC schemes. This highlights the superiority of the proposed scheme meeting the demands of software defined IIoT.

Dual concentric echo sign of ultrasound in primary hyperparathyroidism: The clinical and histopathologic features and differentiation from lymph nodes

ContextUltrasound (US) is the most economical and widely used method for detecting lesions in parathyroid regions. Identifying typically parathyroid adenomas as hypoechoic nodules with clear margins. However, 10 % of lesions exhibit atypical features, such as the dual concentric sign, and the cognition of them still needs to be improved. ObjectiveTo promote understanding of clinical and histopathological features for parathyroid lesions with the dual concentric echo sign and to investigate its pathogenesis and methods for distinguishing from cervical lymph nodes to improve US diagnostic accuracy. MethodsRetrospectively, patients were categorized into three groups: Group 1, with 36 patients showing parathyroid lesions with dual concentric echo signs; Group 2, with 40 patients displaying classic hypoechoic parathyroid lesions; and Group 3, comprising 36 patients with identified lymph nodes, which were all examined from January 2018 to December 2019. The clinical data on demographics, clinical symptoms, serum levels, histopathologic findings, and US image characteristics were thoroughly reviewed. ResultsAccording to the clinical data, no significant differences in demographics or lesion sizes were observed in Group 1 and Group 2 (p > 0.05). No significant variances were noted in biochemical markers, including PTH, T-25OHD, and ALP. However, a notable difference was identified in adjusted serum calcium levels, which were significantly lower in Group 1 compared to Group 2 (p < 0.05). Additionally, the proportion of asymptomatic patients was significantly higher in Group 1 compared to Group 2 (p < 0.05). Pathological examination revealed that all lesions with dual concentric echo signs were parathyroid adenomas. The isoechoic central region predominantly corresponded to areas of loose edema, while the hypoechoic peripheral layer was primarily associated with chief and/or oncocytic cells. By comparing the ultrasonography of Groups 1 and 3, the parathyroid lesions with dual concentric echo signs exhibited significant distinctions from lymph nodes in size, blood flow classification, vascular distribution, and anatomical location (p < 0.05). ConclusionThe parathyroid lesions with dual concentric echo signs in US corresponded to specific histopathological manifestations and relatively mild clinical features in the patients, this finding may increase the likelihood of incidental detection of parathyroid lesions by US. Attention to the details of size, location, and blood flow, especially, may aid US physicians in differentiating parathyroid adenomas from cervical lymph nodes.

Enabling locality-sensitive machine learning towards low predictive overhead in flow classification

The expansion of computer networks and the proliferation of their applications have heightened the necessity for accurate network traffic identification to enhance resource allocation and mitigate congestion, emphasizing flow classification’s pivotal role. Traditional methods, notably deep packet inspection, have encountered challenges due to the rise in encrypted traffic, diminishing their efficacy. In response, machine learning approaches, including decision trees, have been adopted for their ability to classify flows without necessitating access to the payload, thereby enhancing classification efficiency. Nonetheless, these machine learning-based models introduce considerable overhead in category determination. To address this issue, this paper introduces a locality-sensitive flow classification (LSFC) method, leveraging locality-sensitive hashing to efficiently partition and group traffic, thereby expediting the classification process and minimizing overhead. LSFC adeptly balances classification accuracy and processing delay, catering to the specific demands of various scenarios. Comprehensive evaluations demonstrate that LSFC significantly outperforms traditional machine learning models in inference speed by over 50% and improves upon deep packet inspection techniques in handling encrypted traffic, achieving a reduction in classification delay exceeding 15% without compromising classification accuracy.

Numerical Simulation Study of a Pusher Feed Classifier Based on RNG-DPM Method

The classifier is an essential tool for the development of contemporary engineering technology. The application of classifiers is to categorize mixed-sized particles into multi-stage uniform particle sizes. In current studies, the particles in the classifier obtain their initial velocity when feeding. The classification effect is impacted by the inability to precisely control the initial state of the particles. To solve this problem, a pusher feed classifier was designed in this study, and a numerical simulation was performed to investigate its flow field characteristics and classification performance using the RNG-DPM method. A pusher is utilized to achieve particle feeding without initial velocity and to precisely control the initial state of the particles in the classification flow field. A newly developed two-way air inlet structure is designed to provide a superimposed flow field and enable the five-stage classification. Our results show that this pusher feed classifier has the best classification effect when the vertical airflow velocity is 10 m/s and the horizontal airflow velocity is 3 m/s. Meanwhile, the classification size ratio (CSR) from outlet 1 to outlet 5 was 1.24, 0.55, 0.45, 0.39, and 0.15, respectively.

Applying self-supervised learning to network intrusion detection for network flows with graph neural network

Graph Neural Networks (GNNs) have garnered intensive attention for Network Intrusion Detection System (NIDS) due to their suitability for representing the network traffic flows. However, most present GNN-based methods for NIDS are supervised or semi-supervised. Network flows need to be manually annotated as supervisory labels, a process that is time-consuming or even impossible, making NIDS difficult to adapt to potentially complex attacks, especially in large-scale real-world scenarios. The existing GNN-based self-supervised methods focus on the binary classification of network flow as benign or not, and thus fail to reveal the types of attack in practice. This paper studies the application of GNNs to identify the specific types of network flows in an unsupervised manner. We first design an encoder to obtain graph embedding, that introduces the graph attention mechanism and considers the edge information as the only essential factor. Then, a self-supervised method based on graph contrastive learning is proposed. The method samples center nodes, and for each center node, generates subgraph by it and its direct neighbor nodes, and corresponding contrastive subgraph from the interpolated graph, and finally constructs positive and negative samples from subgraphs. Furthermore, a structured contrastive loss function based on edge features and graph local topology is introduced. To the best of our knowledge, it is the first GNN-based self-supervised method for the multiclass classification of network flows in NIDS. Detailed experiments conducted on four real-world datasets (NF-Bot-IoT, NF-Bot-IoT-v2, NF-CSE-CIC-IDS2018, and NF-CSE-CIC-IDS2018-v2) systematically compare our model with the state-of-the-art supervised and self-supervised models, illustrating the considerable potential of our method. Our code is accessible through https://github.com/renj-xu/NEGSC.