Classification Of Flow Patterns Research Articles

A Distributed Conductance Cross-Correlation Method for Measuring Low-Velocity and High Water-Cut Oil-Water Flows

Oil-water two-phase flow widely exists in the oil production and transportation. The flow pattern presents obvious complexity due to non-uniform distribution of water and oil. When the onshore oilfields are on a middle-late stage of oil recovery, the oil production gradually decreases and the water-cut increases. Accurate measurement of oil-water two-phase flow is a significant issue to improve oil production performance and optimize oil reservoir management. Therefore, we propose a distributed four-sector conductance cross-correlation sensor to realize the flow pattern classification and the measurement of flow velocity and water holdup for low-velocity and high water-cut oil-water flows in a 20 mm inner diameter pipe. The geometry dimension is optimized according to the sensitive field distributions of the sensor. In the flow loop test of vertical upward oil-water two-phase flow, the output signals measured from upstream and downstream electrodes are used to figure out three flow patterns combining with a high-speed camera, and the cross-correlation velocity. Then we use the kinematic wave theory to connect the relationship between the cross-correlation flow velocity and the mixture flow velocity considering water- cut and flow patterns. In addition, according to Maxwell equation, we calculate the water holdup. Finally, a drift velocity model based on three flow patterns is constructed to measure the oil phase superficial velocity.

Classification of two-phase flow patterns using multifractal analysis of pressure oscillations

The two-phase flow patterns (air–water) in horizontal square minichannel (3 × 3 mm) has been analysed. The multifractal analysis of pressure drop fluctuations was used for qualitative assessment of two-phase flow patterns. The results of the complexity analysis using the multifractal spectral width (Δh) are presented. The proposed method allows us to identify the following two-phase flow patterns: micro-bubbles flow, micro- and minibubbles flow, micro- and mini-bubbles with confined bubbles flow, slug flow, stratified flow. The obtained results confirm that this type of analysis can be considered as an alternative way of identification of two-phase flow patterns in the minichannel. The work also focuses on the discussion of the occurrence and identification of bubbles churns in slugs and churns.

Machine learning classification of in-tube condensation flow patterns using visualization

Identifying two-phase flow patterns is fundamental to successfully design and subsequently optimize high-precision heat transfer equipment, given that the heat transfer efficiency and pressure gradients occurring in such thermo-hydraulic systems are dependent on the flow structure of the working fluid. This paper shows that with visualization data and artificial neural networks, the flow pattern images of condensation of R-134a refrigerant in inclined smooth tubes can be classified with more than 98% accuracy. The study considers 10 classes of flow pattern images acquired from previous experimental works for a wide range of flow conditions and the full range of tube inclination angles. Although not the focus of this paper, the use of a Principal Component Analysis allowed feature dimensionality reduction, dataset visualization, and decreased associated computational cost when used together with multilayer perceptron neural networks. In addition, the superior two-dimensional spatial learning capability of convolutional neural networks allowed improved image classification and generalization performance. In both cases, the classification was performed sufficiently fast to enable real-time implementation in two-phase flow systems.

Steady compressible radially symmetric flows with nonzero angular velocity in an annulus

In this paper, we investigate steady inviscid compressible flows with radial symmetry in an annulus. The major concerns are transonic flows with or without shocks. One of the main motivations is to elucidate the role played by the angular velocity in the structure of steady inviscid compressible flows. We give a complete classification of flow patterns in terms of boundary conditions at the inner and outer circle. Due to the nonzero angular velocity, many new flow patterns will appear. There exists accelerating or decelerating smooth transonic flows in an annulus satisfying one side boundary conditions at the inner or outer circle with all sonic points being nonexceptional and noncharacteristically degenerate. More importantly, it is found that besides the well-known supersonic-subsonic shock in a divergent nozzle as in the case without angular velocity, there exists a supersonic-supersonic shock solution, where the downstream state may change smoothly from supersonic to subsonic. Furthermore, there exists a supersonic-sonic shock solution where the shock circle and the sonic circle coincide, which is new and interesting.

A mechanistic model for predicting erosion in churn flow

Sand particles, which are regularly entrained with the production fluids while transporting from a reservoir to surface facilities, can cause severe erosion and erosion-corrosion damage to production equipment. A crucial parameter for studying erosion in multiphase flows is the flow pattern. The particle motion characteristics will change drastically with changes in the flow pattern in the pipelines. In vertical upward gas-liquid flow, a widely accepted flow pattern classification is bubble, slug, churn and annular flow. Churn flow is a chaotic flow pattern consisting of Taylor bubbles and liquid slugs that are distorted in shape and it is characterized by the presence of periodic large interfacial waves called flooding-type waves. Studying the erosion due to sand particles entrained in churn flows is extremely difficult, since the solid particle erosion magnitudes are coupled with several multiphase flow parameters such as phase distribution and velocities. This work focuses on improvement of a multiphase churn flow erosion model for elbows in vertical pipes. In this investigation, a drift-flux model is proposed to better predict the flow behavior and the characteristic initial particle velocity for churn gas-liquid flow. Based on the drift-flux model, the actual gas velocity is calculated as a function of mixture and drift velocities. The coefficients of the drift flux expression under churn flow conditions are determined from datasets available in literature. The model predictions are compared with the erosion data bank at the Erosion/Corrosion Research Center and other published papers. The data comprises effects of pipe diameter, particle characteristics, flow velocities, and liquid viscosity. The results show that the new approach provides much better predictions for the maximum erosion in bends when compared with the original mixture based model and other models available in literature.

Case study on the identification and classification of small-scale flow patterns in flaring active region

Context. We propose a novel methodology to identity flows in the solar atmosphere and classify their velocities as either supersonic, subsonic, or sonic. Aims. The proposed methodology consists of three parts. First, an algorithm is applied to the Solar Dynamics Observatory (SDO) image data to locate and track flows, resulting in the trajectory of each flow over time. Thereafter, the differential emission measure inversion method is applied to six Atmospheric Imaging Assembly (AIA) channels along the trajectory of each flow in order to estimate its background temperature and sound speed. Finally, we classify each flow as supersonic, subsonic, or sonic by performing simultaneous hypothesis tests on whether the velocity bounds of the flow are larger, smaller, or equal to the background sound speed. Methods. The proposed methodology was applied to the SDO image data from the 171 Å spectral line for the date 6 March 2012 from 12:22:00 to 12:35:00 and again for the date 9 March 2012 from 03:00:00 to 03:24:00. Eighteen plasma flows were detected, 11 of which were classified as supersonic, 3 as subsonic, and 3 as sonic at a 70% level of significance. Out of all these cases, 2 flows cannot be strictly ascribed to one of the respective categories as they change from the subsonic state to supersonic and vice versa. We labeled them as a subclass of transonic flows. Results. The proposed methodology provides an automatic and scalable solution to identify small-scale flows and to classify their velocities as either supersonic, subsonic, or sonic. It can be used to characterize the physical properties of the solar atmosphere. Conclusions. We identified and classified small-scale flow patterns in flaring loops. The results show that the flows can be classified into four classes: sub-, super-, trans-sonic, and sonic. The flows occur in the complex structure of the active region magnetic loops. The detected flows from AIA images can be analyzed in combination with the other high-resolution observational data, such as Hi-C 2.1 data, and be used for the development of theories describing the physical conditions responsible for the formation of flow patterns.

Flow pattern classification in water-air vertical flows using a single ultrasonic transducer

Researchers around the world have proposed different methodologies for flow-pattern identification in multiphase flows. The most common explored techniques are based on optical imaging, pressure fluctuation, electrical impedance, radiation absorption, and ultrasound. Because the ultrasonic technique is relatively low in cost, easy to conduct, accurate, nonintrusive, and suitable for opaque systems, this paper presents a flow-pattern classifier based on ultrasound for two-phase vertical water–air flows. It is based on the coefficient of variation of the energy of the pulses reflected by the dispersed phase. Also, it uses just one ultrasonic transducer. An experimental plant was built to develop two-phase water–air flows to evaluate the proposed technique performance by comparing it to the visual analysis of the recorded image of the flow. A total of 102 different experimental points were collected and evaluated. The results have shown that the method was capable to distinguish dispersed bubbles, slug and churn flow patterns.

Flow pattern classification in liquid-gas flows using flow-induced vibration

The multiphase flow is not only the most common flow in nature but also occurs in various major industrial fields. Furthermore, in many industrial plants, the single and multiphase flows generates vibration and noise. In the context of two-phase flows, a specific case of multiphase flow, the flow pattern determination is crucial to their analysis, and despite the recent progress and developments in flow-induced vibration for two-phase flows, it is still considered an open topic. This paper develops a novel algorithm for flow pattern classification using the vibration signal from a vertical pipe conveying a liquid-gas two-phase flow to determine the flow pattern. An experimental apparatus and procedure were developed to perform this investigation. The analysis in the frequency domain showed a distinct frequency band activity for slug and churn flows. The analysis in time domain showed a significant amplitude variation for these flow patterns. Finally, by using the RMS and Pearson correlation coefficient, it was possible to classify the studied cases accurately. The results present a non-intrusive technique to identify the flow pattern in two-phase liquid-gas vertical flows.

Redchyts Numerical simulation of the viscous incompressible flow around of the group of two bodies

Analysis of recent research and publications. It is known that the presence near the body of another body or a solid wall in a flow can significantly change both the overall flow pattern and the aerodynamic characteristics of bodies in a group. Studies of the interaction of bodies in the flow are conducted for a long time. In [6], the results of a study of changes in the overall flow pattern and the form of interaction of vortices behind tandem-arranged circular cylinders are presented. Further, experimental studies of the flow around a group consisting of two cylinders were aimed at classifying flow patterns depending on the position of the group in the flow, the distance between the cylinders and the Reynolds number [1, 2, 9]. A rather complete identification and classification of the pattern of flow was performed in [6, 7]. Studies on the classification and analysis of flow patterns are still being conducted [1]. Studies on the classification of patterns of the flow around group of spheres are currently performed mainly with the help of numerical simulation. In [3, 4, 5], simulation of the flow around spheres on the side-by-side position was performed. In [8], the classification of typical patterns of the flow around two spheres (Re = 300) with considering of different positions of the spheres relative to the flow direction was made. The authors of [8] describe nine typical patterns of the flow around two spheres in analogy with the patterns of the flow of the two cylinders.The purpose of the study. The main goal of this work is study the mutual influence of two bodies in a flow of a viscous incompressible fluid and a change in the flow structure with a change in the position of the bodies in the group relative to the incoming flow. Also, the aim of the work was to study the influence of the mutual arrangement of bodies in a group on the non-stationary and time-averaged aerodynamic characteristics of bodies in a group.Modeling of the flow around groups of cylinders and spheres. Numerical simulation of the flow around the group of cylinders was carried out with the values of the angle θ = 0°, 15°, 30°, 45°, 60°, 75°, 90° and the gap between the cylinders h = 0.2D, 0.4D, 0.6D, 0.8D, 1.0D, 2.0D, 3.0D, 4.0D, 5.0D. The flow parameters was corresponded to the flow around a circular cylinder at Re = 80 and 1.66 105. Eight patterns (regimes) of flow around a group of two cylinders at Re = 80 were found. Regimes 1 and 2 are steady state flows. In regime 1, the drag coefficient is Cx2 <0, and for regime 2, Cx2> 0. Regimes 3-8 are unsteady flows. Regime 8 is an aperiodic change in Cx, Cy. Regimes 3 - 7 are periodic, characterized by different values of the coefficients Cx, Cy, as well as those oscillations of Cx and Cy that occur in phase or antiphase. Simulation of the turbulent flow around a group of two cylinders took place at the tandem and the side-by-side positions at distances between cylinders centers 1.435D and 3.7D.Similarly, in this work, was performed the parametric study of the flow around two spheres for Reynolds number 750 with the distances between the centers of the spheres along axis Δx = 0.0, 1.0, 2.0, 3.0 and Δy = 0.0, 1.0, 2.0, 3.0. The drag and lift coefficients were obtained, as well as the patterns of flow around two spheres were analyzed.Conclusions. Depending on the position of the group relative to the flow, the average drag coefficient of the cylinders and spheres in the group can be both smaller and larger than the drag coefficient of a single body with the same parameters of the free flow. With a tandem arrangement, the second cylinder has a stabilizing effect and with a decrease in the gap of less than three diameters, the flow becomes steady state. For all cases with staggered arrangement of spheres the symmetry restoration of vortex structures is observed. In the case of the tandem arrangement of spheres, the separation of loop-shaped vortex structures is realized as in the case of a flow around a single sphere.

Modeling flow pattern transitions in electrical submersible pump under gassy flow conditions

Gas entrainment is frequently encountered in electrical submersible pumps (ESP). When occurring, ESP suffers from moderate to severe performance degradation accompanied by unstable operations, depending on the inlet gas volumetric fraction (GVF) and pump rotational speed. The resulted pressure surging and instabilities may cause pump vibrations and short run-life. For a better design of ESP-based production system, the mechanistic model is needed to predict its performance under gas-liquid flow conditions. Similar to multiphase pipe flow, the flow pattern identification and classification inside a rotating ESP is important. In this paper, a new mechanistic model is proposed, which can map different flow patterns inside ESP with gas presence.Experimental results reveal that the boosting pressure of ESP degrades as GVF increases. With ESP's H-Q performance curves obtained, the flow patterns including dispersed bubble flow, bubbly flow, intermittent flow and segregated flow can be identified by deflection points on H-Q curves (Gamboa and Prado, 2008). The transition boundaries between different flow patterns are mapped. Starting from the free body diagram on a stable bubble in the rotating flow field, the transition boundaries of dispersed bubble flow to bubbly flow and bubbly flow to intermittent flow are formulated. Based on the combined momentum equation, the transition criterion of intermittent flow to segregated flow is derived. The flow pattern map predicted by mechanistic model agrees well with that detected from the experimental H-Q curves of ESP under gassy flow conditions.

Flow patterns classification of dense-phase pneumatic conveying of pulverized coal in the industrial vertical pipeline

Dense-phase pneumatic conveying of pulverized coal experiments with superficial gas velocity ranged from 0.59 m s−1 to 2.43 m s−1 were carried out on the self-established 50 mm I.D. industrial pipe facility. The fluctuations of pressure, solid velocity, solid concentration and the average cross-section concentration were analyzed in terms of relative standard deviation (RSD), powder spectrum density analysis (PSD) and wavelet analysis. Further, a novel classification model of flow patterns for dense-phase pneumatic conveying of pulverized coal was proposed. The model was based on the relationships among superficial gas velocity, particle velocity and solid concentration. K-means method, std method, k-medoids and CART method were compared firstly and then the k-means method and CART method were introduced to classify the flow patterns. The bagged trees method and 5-fold cross-validation were used to train the classification model, whose accuracy was 95.2% and 91.9%. It meant that the new classification model was able to classify the flow patterns in the pneumatic conveying of pulverized coal. In the end, the change of flow patterns and transmission of flow patterns were explained by the wavelet analysis method from the aspects of different signal scales and power spectrum analysis. The new classification model will be of great help for the industry to have a better understanding on the classification and control of flow patterns.

Flow Pattern Classification in Lymphatic Malformations by Indocyanine Green Lymphography.

Lymphatic malformation is a congenital lymphatic disorder. Although a few lymphangiographic and lymphoscintigraphic studies of lymphatic malformation exist, its lymphatic flow has not been fully assessed, and a classification system has not yet been established. However, indocyanine green lymphography has been developed to safely provide a fine assessment of lymph flow in the treatment of lymphedema. In addition, indocyanine green lymphography has been shown to be helpful in detecting the lymphatic malformation inflow for the treatment of refractory microcystic type lymphatic malformation using the venous anastomosis technique. Therefore, the authors aimed to reveal the in vivo lymph flow around the lymphatic malformation using indocyanine green lymphography, and to design a classification system according to the observed patterns. Indocyanine green lymphography was performed in 20 sequential pediatric patients with lymphatic malformation (aged 11 months to 10 years). Most of the cases were intractable, with microcystic or mixed-type lymphatic malformation. All patients successfully completed lymphography with clear observations. The flow patterns were classified into four types: type 1 had a strong detectable inflow; type 2 had multiple small observable inflows; type 3 had a superficial lymph flow over the lesion; and type 4 had a flow around the lymphatic malformation, without any connections to the lesion. The proposed classification system may aid in the further development of surgical treatments for lymphatic malformation.

An adaptive electrical resistance tomography sensor with flow pattern recognition capability

The all traditional electrical resistance tomography (ERT) sensors have a static structure, which cannot satisfy the intelligent requirements for adaptive optimization to ERT sensors that is subject to flow pattern changes during the real-time detection of two-phase flow. In view of this problem, an adaptive ERT sensor with a dynamic structure is proposed. The electrodes of the ERT sensor are arranged in an array structure, the flow pattern recognition technique is introduced into the ERT sensor design and accordingly an ERT flow pattern recognition method based on signal sparsity is proposed. This method uses the sparse representation of the signal to express the sampling voltage of the ERT system as a sparse combination and find its sparse solution to achieve the classification of different flow patterns. With the introduction of flow identification information, the sensor has an intelligent function of adaptively and dynamically adapting the sensor structure according to the real-time flow pattern change. The experimental results show that the sensor can automatically identify four typical flow patterns: core flow, bubble flow, laminar flow and circulation flow with recognition rates of 91%, 93%, 90% and 88% respectively. For different flow patterns, the dynamically optimized sensor can significantly improve the quality of ERT image reconstruction.

Classification of Blood Flow Patterns in Cerebral Aneurysms.

We present a Cerebral Aneurysm Vortex Classification (CAVOCLA) that allows to classify blood flow in cerebral aneurysms. Medical studies assume a strong relation between the progression and rupture of aneurysms and flow patterns. To understand how flow patterns impact the vessel morphology, they are manually classified according to predefined classes. However, manual classifications are time-consuming and exhibit a high inter-observer variability. In contrast, our approach is more objective and faster than manual methods. The classification of integral lines, representing steady or unsteady blood flow, is based on a mapping of the aneurysm surface to a hemisphere by calculating polar-based coordinates. The lines are clustered and for each cluster a representative is calculated. Then, the polar-based coordinates are transformed to the representative as basis for the classification. Classes are based on the flow complexity. The classification results are presented by a detail-on-demand approach using a visual transition from the representative over an enclosing surface to the associated lines. Based on seven representative datasets, we conduct an informal interview with five domain experts to evaluate the system. They confirmed that CAVOCLA allows for a robust classification of intra-aneurysmal flow patterns. The detail-on-demand visualization enables an efficient exploration and interpretation of flow patterns.

Exploration of blood flow patterns in cerebral aneurysms during the cardiac cycle

This paper presents a method for clustering time-dependent blood flow data, represented by path lines, in cerebral aneurysms using a reliable similarity measure combined with a clustering technique. Such aneurysms bear the risk of rupture, whereas their treatment also carries considerable risks for the patient. Medical researchers emphasize the importance of investigating aberrant blood flow patterns for the patient-specific rupture risk assessment and treatment analysis. Therefore, occurring flow patterns are manually extracted and classified according to predefined criteria. The manual extraction is time-consuming for larger studies and affected by visual clutter, which complicates the subsequent classification of flow patterns. In contrast, our method allows an automatic and reliable clustering of intra-aneurysmal flow patterns that facilitates their classification. We introduce a similarity measure that groups spatio-temporally adjacent flow patterns. We combine our similarity measure with a commonly used clustering technique and applied it to five representative datasets. The clustering results are presented by 2D and 3D visualizations and were qualitatively compared and evaluated by four domain experts. Moreover, we qualitatively evaluated our similarity measure.

The Research on Flow Imaging and Flow Profile Interpretation Method in Shale Gas Horizontal Well Multiphase Flow

Multiphase flow state is complex in shale gas horizontal well. Using foreign array probe tools to accurately monitor fluid production location and flowrate is very important, but at present domestic flow imaging and flow profile interpretation technology is still immature. In this paper, based on analysis the structure and characteristics of Schlumberger FSI (Flow Scanning Imager), firstly, the multiphase flow in shale gas horizontal wells was numerically simulated by Fluent software, and the velocity and pressure distributions were obtained. Compared with the typical horizontal well gas-water two-phase flow pattern classification, the two methods are in good agreement. Secondly, using the methods of gridding and simple inverse distance interpolation, the calculation of holdup and average velocity are carried out. After that, considering the deviation and flow pattern, the calculation method of key parameters was deduced such as the apparent velocity, phase distribution coefficient and drift velocity. What’s more, Flow profile interpretation model is established in horizontal air-water two-phase. In the end, the field test data show that three-dimensional flow imaging effect and interpretation precision is good for stratified flow.

Effects of water and gas injection and viscosity on volumetric fraction, pressure gradient and phase inversion in upward-vertical three-phase pipe flow

The natural and artificial presence of water and gas is common in the offshore oil production scenario. In the last years, several studies on two-phase flows of viscous oils proved that the water and gas addiction is very efficient in the frictional and gravitational pressure-gradient reduction. In recent works, the oil-water-gas three-phase flow showed promising results in the total pressure-gradient reduction. However, the works on three-phase flow are not as conclusive as those on two-phase flow. In order to better understand the phenomenology of the upward-vertical three-phase flow, new experimental data are presented in a wide range of flow rates of gas, water and oil with three different viscosities (100 mPa s, 220 mPa s and 325 mPa s). The gains obtained by gas (air) and water injection were quantified by pressure-gradient reduction factors. A new flow-pattern classification in three-phase flows is proposed as a function of water and gas superficial velocities. Furthermore, analyzes were performed about the volumetric fractions, the slip between the liquid phases and the applicability of the drift-flux model in three-phase flow. The results indicate that the total pressure gradient in three-phase flow can be considerably lower than in single-phase and two-phase flows; therefore, a profound understanding of oil-water-gas three-phase flow is in order if one intends to improve the oil production in vertical wellbores.

An artificial intelligence based improved classification of two-phase flow patterns with feature extracted from acquired images

Flow pattern recognition is necessary to select design equations for finding operating details of the process and to perform computational simulations. Visual image processing can be used to automate the interpretation of patterns in two-phase flow. In this paper, an attempt has been made to improve the classification accuracy of the flow pattern of gas/ liquid two- phase flow using fuzzy logic and Support Vector Machine (SVM) with Principal Component Analysis (PCA). The videos of six different types of flow patterns namely, annular flow, bubble flow, churn flow, plug flow, slug flow and stratified flow are recorded for a period and converted to 2D images for processing. The textural and shape features extracted using image processing are applied as inputs to various classification schemes namely fuzzy logic, SVM and SVM with PCA in order to identify the type of flow pattern. The results obtained are compared and it is observed that SVM with features reduced using PCA gives the better classification accuracy and computationally less intensive than other two existing schemes. This study results cover industrial application needs including oil and gas and any other gas-liquid two-phase flows.

A Unified Model of Oil/Water Two-Phase Flow in the Horizontal Wellbore

Summary In this article, a more-general flow-pattern classification of oil/water two-phase flow in the horizontal wellbore is proposed first according to the theoretical analysis and previous research achievements, on the basis of which a simplification is then performed through reasonable incorporation, and the ultimate flow patterns considered for modeling are reduced to two categories containing only six standard patterns. By use of the classical two-fluid and homogeneous modeling methodologies stemming from oil/water two-phase flow in conventional pipes, combined with the simplified classification, a mechanistic model is developed to predict the flow characteristics including the flow patterns and pressure losses for oil/water two-phase variable-mass flow in the horizontal wellbore. Model implementation is performed on the basis of the universal principle that a system will stabilize to the equilibrium state of minimum energy. Overall performance of the mechanistic model is then validated against the new data sets measured upon a large-scale experimental apparatus at the China University of Petroleum (CUP), which is designed and constructed to simulate the gas/oil/water multiphase flow in horizontal wellbores with wall mass transfer. Results show that the model developed in this paper can not only properly predict the flow patterns of oil/water two-phase flow in the horizontal wellbore, but also has high prediction accuracy for the pressure drops. Compared with the new experimental data for oil/water two-phase variable-mass flow that covers a series of input water-volumetric fractions ranging from 10 to 90%, the highest absolute average percentage error of the new unified model is 12% and the whole error is 9.2%, which demonstrates an acceptable performance. Investigations conducted in this study further enrich and develop the theory of hydrodynamic calculation for oil/water flow in the horizontal wellbore with wall influx.

Extension of Traffic Flow Pattern Dynamic Classification by a Macroscopic Model Using Multivariate Clustering

In this paper, we evaluate the performance of a dynamic approach to classifying flow patterns reconstructed by a switching-mode macroscopic flow model considering a multivariate clustering method. To remove noise and tolerate a wide scatter of traffic data, filters are applied before the overall modeling process. Filtered data are dynamically and simultaneously input to the density estimation and traffic flow modeling processes. A modified cell transmission model simulates traffic flow to explicitly account for flow condition transitions considering wave propagations throughout a freeway test stretch. We use flow dynamics specific to each of the cells to determine the mode of prevailing traffic conditions. Flow dynamics are then reconstructed by neural methods. By using two methods in part, i.e., dynamic classification and nonhierarchical clustering, classification of flow patterns over the fundamental diagram is obtained by considering traffic density as a pattern indicator. The fundamental diagram of speed-flow is dynamically updated to specify the current corresponding flow pattern. The dynamic classification approach returned promising results in capturing sudden changes on test stretch flow patterns as well as performance compared to multivariate clustering. The dynamic methods applied here are open to use in practice within intelligent management strategies, including incident detection and control and variable speed management.