scholarly journals Morphology Decoder to Predict Heterogeneous Rock Permeability with Machine Learning Guided 3D Vision

2021 ◽  
Author(s):  
Omar Alfarisi ◽  
Aikifa Raza ◽  
Djamel Ouzzane ◽  
Mohamed Sassi ◽  
Hongtao Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Network Flow ◽  
Biological Fluids ◽  
Flow Behavior ◽  
Flow Simulation ◽  
Semantic Segmentation ◽  
Integration Model ◽  
3D Vision ◽  
Heterogeneous Rock ◽  
Novel Processes

<p><a></a><a>Permeability has a dominant influence on the flow behavior of a natural fluid, and without proper quantification, biological fluids (Hydrocarbons) and water resources become waste. During the first decades of the 21<sup>st</sup> century, permeability quantification from nano-micro porous media images emerged, aided by 3D pore network flow simulation, primarily using the Lattice Boltzmann simulator. Earth scientists realized that the simulation process holds millions of flow dynamics calculations with accumulated errors and high computing power consumption. Therefore, accuracy and efficiency challenges obstruct planetary exploration. To effic­­­iently, consistently predict permeability with high quality, we propose the Morphology Decoder. It is a parallel and serial flow reconstruction of machine learning-driven semantically segmented heterogeneous rock texture images of 3D X-Ray Micro Computerized Tomography (μCT) and Nuclear Magnetic Resonance (MRI). For 3D vision, we introduce controllable-measurable-volume as new supervised semantic segmentation, in which a unique set of voxel intensity corresponds to grain and pore throat sizes. The morphology decoder demarks and aggregates the morphologies' boundaries in a novel way to quantify permeability. The morphology decoder method consists of five novel processes, which we describe in this paper, these novel processes are (1) Geometrical: 3D Permeability Governing Equation, (2) Machine Learning: Guided 3D Properties Recognition of Rock Morphology, (3) Analytical: 3D Image Properties Integration Model for Permeability, (4) Experimental: MRI Permeability Imager, and (5) Morphology Decoder (the process that integrates the other four novel processes).</a></p>

scholarly journals Morphology Decoder to Predict Heterogeneous Rock Permeability with Machine Learning Guided 3D Vision

2021 ◽  
Author(s):  
Omar Alfarisi ◽  
Aikifa Raza ◽  
Djamel Ouzzane ◽  
Mohamed Sassi ◽  
Hongtao Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Network Flow ◽  
Biological Fluids ◽  
Flow Behavior ◽  
Flow Simulation ◽  
Semantic Segmentation ◽  
Integration Model ◽  
3D Vision ◽  
Heterogeneous Rock ◽  
Novel Processes

<p><a></a><a>Permeability has a dominant influence on the flow behavior of a natural fluid, and without proper quantification, biological fluids (Hydrocarbons) and water resources become waste. During the first decades of the 21<sup>st</sup> century, permeability quantification from nano-micro porous media images emerged, aided by 3D pore network flow simulation, primarily using the Lattice Boltzmann simulator. Earth scientists realized that the simulation process holds millions of flow dynamics calculations with accumulated errors and high computing power consumption. Therefore, accuracy and efficiency challenges obstruct planetary exploration. To effic­­­iently, consistently predict permeability with high quality, we propose the Morphology Decoder. It is a parallel and serial flow reconstruction of machine learning-driven semantically segmented heterogeneous rock texture images of 3D X-Ray Micro Computerized Tomography (μCT) and Nuclear Magnetic Resonance (MRI). For 3D vision, we introduce controllable-measurable-volume as new supervised semantic segmentation, in which a unique set of voxel intensity corresponds to grain and pore throat sizes. The morphology decoder demarks and aggregates the morphologies' boundaries in a novel way to quantify permeability. The morphology decoder method consists of five novel processes, which we describe in this paper, these novel processes are (1) Geometrical: 3D Permeability Governing Equation, (2) Machine Learning: Guided 3D Properties Recognition of Rock Morphology, (3) Analytical: 3D Image Properties Integration Model for Permeability, (4) Experimental: MRI Permeability Imager, and (5) Morphology Decoder (the process that integrates the other four novel processes).</a></p>

Virtual Private Network Flow Detection in Wireless Sensor Networks using Machine Learning Techniques

2021 ◽  
Vol 10 ◽  
Author(s):  
S. Phani Praveen ◽  
T. Bala Murali Krishna ◽  
Sunil K. Chawla ◽  
CH Anuradha
Keyword(s):  
Machine Learning ◽  
Traffic Flow ◽  
Network Flow ◽  
Virtual Private Network ◽  
Machine Learning Techniques ◽  
Protocol Security ◽  
Learning Techniques ◽  
Private Network ◽  
Hyper Text Transfer Protocol ◽  
Transfer Protocol

Background: Every organization generally uses a VPN service individually to leather the actual communication. Such communication is actually not allowed by organization monitoring network. But these institutes are not in a position to spend huge amount of funds on secure sockets layer to monitor traffic over their computer networks. Objective: Our work suggests simple technique to block or detect annoying VPN clients inside the network activities. This method does not requires the network to decrypt or even decode any network communication. Method: The proposed solution selects two machine learning techniques Feature Tree and K-means as classifiction techniques which work on time related features. First, the DNS mapping with the ordinary characteristic of the transmission control protocol / internet protocol computer network stack is identified and it is not to be considered as a normal traiffic flow if the domain name information is not available. The process also examines non-standard utilization of hyper text transfer protocol security and also conceal such communication from hyper text transfer protocol security dependent filters in firewall to detect as anomaly in largely. Results: we define the trafic flow as normal trafic flow and VPN traffic flow. These two flows are characterized by taking two machine learning techniques Feature Tree and K-means. We have executed each experment 4 times. As a result, eight types of regular traffics and eight types of VPN traffics were represented. Conclusion: Once trafic flow is identified, it is classified and studied by machine learning techniques. Using time related features, the traffic flow is defined as normal flow or VPN traffic flow.

scholarly journals Semantic segmentation of PolSAR image data using advanced deep learning model

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Rajat Garg ◽  
Anil Kumar ◽  
Nikunj Bansal ◽  
Manish Prateek ◽  
Shashi Kumar
Keyword(s):  
Machine Learning ◽  
Remote Sensing ◽  
Deep Learning ◽  
Urban Area ◽  
Urban Areas ◽  
Learning Algorithms ◽  
Semantic Segmentation ◽  
Learning Model ◽  
Machine Learning Algorithms ◽  
Deep Learning Model

AbstractUrban area mapping is an important application of remote sensing which aims at both estimation and change in land cover under the urban area. A major challenge being faced while analyzing Synthetic Aperture Radar (SAR) based remote sensing data is that there is a lot of similarity between highly vegetated urban areas and oriented urban targets with that of actual vegetation. This similarity between some urban areas and vegetation leads to misclassification of the urban area into forest cover. The present work is a precursor study for the dual-frequency L and S-band NASA-ISRO Synthetic Aperture Radar (NISAR) mission and aims at minimizing the misclassification of such highly vegetated and oriented urban targets into vegetation class with the help of deep learning. In this study, three machine learning algorithms Random Forest (RF), K-Nearest Neighbour (KNN), and Support Vector Machine (SVM) have been implemented along with a deep learning model DeepLabv3+ for semantic segmentation of Polarimetric SAR (PolSAR) data. It is a general perception that a large dataset is required for the successful implementation of any deep learning model but in the field of SAR based remote sensing, a major issue is the unavailability of a large benchmark labeled dataset for the implementation of deep learning algorithms from scratch. In current work, it has been shown that a pre-trained deep learning model DeepLabv3+ outperforms the machine learning algorithms for land use and land cover (LULC) classification task even with a small dataset using transfer learning. The highest pixel accuracy of 87.78% and overall pixel accuracy of 85.65% have been achieved with DeepLabv3+ and Random Forest performs best among the machine learning algorithms with overall pixel accuracy of 77.91% while SVM and KNN trail with an overall accuracy of 77.01% and 76.47% respectively. The highest precision of 0.9228 is recorded for the urban class for semantic segmentation task with DeepLabv3+ while machine learning algorithms SVM and RF gave comparable results with a precision of 0.8977 and 0.8958 respectively.

scholarly journals Understanding Rooftop PV Panel Semantic Segmentation of Satellite and Aerial Images for Better Using Machine Learning

2021 ◽  
pp. 100057
Author(s):  
Peiran Li ◽  
Haoran Zhang ◽  
Zhiling Guo ◽  
Suxing Lyu ◽  
Jinyu Chen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Semantic Segmentation ◽  
Aerial Images ◽  
Pv Panel

scholarly journals Internal face finishing for a cooling channel using a fluid containing free abrasive grains

2021 ◽  
Author(s):  
Mitsugu Yamaguchi ◽  
Tatsuaki Furumoto ◽  
Shuuji Inagaki ◽  
Masao Tsuji ◽  
Yoshiki Ochiai ◽  
...  
Keyword(s):  
High Speed ◽  
Flow Behavior ◽  
Flow Simulation ◽  
Cooling Channel ◽  
H13 Steel ◽  
Channel Diameter ◽  
Abrasive Grains ◽  
Low Viscosity ◽  
Short Period ◽  
Internal Pressures

AbstractIn die-casting and injection molding, a conformal cooling channel is applied inside the dies and molds to reduce the cycle time. When the internal face of the channel is rough, both cooling performance and tool life are negatively affected. Many methods for finishing the internal face of such channels have been proposed. However, the effects of the channel diameter on the flow of a low-viscosity finishing media and its finishing characteristics for H13 steel have not yet been reported in the literature. This study addresses these deficiencies through the following: the fluid flow in a channel was computationally simulated; the flow behavior of abrasive grains was observed using a high-speed camera; and the internal face of the channel was finished using the flow of a fluid containing abrasive grains. The flow velocity of the fluid with the abrasive grains increases as the channel diameter decreases, and the velocity gradient is low throughout the channel. This enables reduction in the surface roughness for a short period and ensures uniform finishing in the central region of the channel; however, over polishing occurs owing to the centrifugal force generated in the entrance region, which causes the form accuracy of the channel to partially deteriorate. The outcomes of this study demonstrate that the observational finding for the finishing process is consistent with the flow simulation results. The flow simulation can be instrumental in designing channel diameters and internal pressures to ensure efficient and uniform finishing for such channels.

scholarly journals Long-term streamflow forecasting for the Cascade Reservoir System of Han River using SWAT with CFS output

2018 ◽  
Vol 50 (2) ◽  
pp. 655-671
Author(s):  
Tian Liu ◽  
Yuanfang Chen ◽  
Binquan Li ◽  
Yiming Hu ◽  
Hui Qiu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Assessment Tool ◽  
Flow Simulation ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Streamflow Forecasting ◽  
Distribution Models ◽  
Han River ◽  
Reservoir System

Abstract Due to the large uncertainties of long-term precipitation prediction and reservoir operation, it is difficult to forecast long-term streamflow for large basins with cascade reservoirs. In this paper, a framework coupling the original Climate Forecasting System (CFS) precipitation with the Soil and Water Assessment Tool (SWAT) was proposed to forecast the nine-month streamflow for the Cascade Reservoir System of Han River (CRSHR) including Shiquan, Ankang and Danjiangkou reservoirs. First, CFS precipitation was tested against the observation and post-processed through two machine learning algorithms, random forest and support vector regression. Results showed the correlation coefficients between the monthly areal CFS precipitation (post-processed) and observation were 0.91–0.96, confirming that CFS precipitation post-processing using machine learning was not affected by the extended forecast period. Additionally, two precipitation spatio-temporal distribution models, original CFS and similar historical observation, were adopted to disaggregate the processed monthly areal CFS precipitation to daily subbasin-scale precipitation. Based on the reservoir restoring flow, the regional SWAT was calibrated for CRSHR. The Nash–Sutcliffe efficiencies for three reservoirs flow simulation were 0.86, 0.88 and 0.84, respectively, meeting the accuracy requirement. The experimental forecast showed that for three reservoirs, long-term streamflow forecast with similar historical observed distribution was more accurate than that with original CFS.

scholarly journals Flow Simulation Using Local Grid Refinements to Model Laminated Reservoirs

2018 ◽  
Vol 73 ◽  
pp. 5 ◽  
Author(s):  
Manuel Gomes Correia ◽  
Célio Maschio ◽  
Denis José Schiozer
Keyword(s):  
Simulation Model ◽  
Dynamic Properties ◽  
Flow Behavior ◽  
Flow Simulation ◽  
Coarse Grid ◽  
Thin Layers ◽  
Static Properties ◽  
Dynamic Representation ◽  
Medium Flow ◽  
Local Grid

Super-giant carbonate fields, such as Ghawar, in Saudi Arabia, and Lula, at the Brazilian pre-salt, show highly heterogeneous behavior that is linked to high permeability intervals in thin layers. This article applies Local Grid Refinements (LGR) integrated with upscaling procedures to improve the representation of highly laminated reservoirs in flow simulation by preserving the static properties and dynamic trends from geological model. This work was developed in five main steps: (1) define a conventional coarse grid, (2) define LGR in the conventional coarse grid according to super-k and well locations, (3) apply an upscaling procedure for all scenarios, (4) define LGR directly in the simulation model, without integrate geological trends in LGR and (5) compare the dynamic response for all cases. To check results and compare upscaling matches, was used the benchmark model UNISIM-II-R, a refined model based on a combination of Brazilian Pre-salt and Ghawar field information. The main results show that the upscaling of geological models for coarse grid with LGR in highly permeable thin layers provides a close dynamic representation of geological characterization compared to conventional coarse grid and LGR only near-wells. Pseudo-relative permeability curves should be considered for (a) conventional coarse grid or (b) LGR scenarios under dual-medium flow simulations as the upscaling of discrete fracture networks and dual-medium flow models presents several limitations. The conventional approach of LGR directly in simulation model, presents worse results than LGR integrated with upscaling procedures as the extrapolation of dynamic properties to the coarse block mismatch the dynamic behavior from geological characterization. This work suggests further improvements for results for upscaling procedures that mask the flow behavior in highly laminated reservoirs.

Computation of Flow Through a Three-Dimensional Periodic Array of Porous Structures by a Parallel Immersed-Boundary Method

2014 ◽  
Vol 136 (4) ◽  
Author(s):  
Zhenglun Alan Wei ◽  
Zhongquan Charlie Zheng ◽  
Xiaofan Yang
Keyword(s):  
Porous Media ◽  
Parallel Implementation ◽  
Flow Behavior ◽  
Three Dimensional ◽  
Flow Simulation ◽  
Periodic Array ◽  
Immersed Boundary ◽  
Navier Stokes Equation ◽  
Flow Through ◽  
Ib Method

A parallel implementation of an immersed-boundary (IB) method is presented for low Reynolds number flow simulations in a representative elementary volume (REV) of porous media that are composed of a periodic array of regularly arranged structures. The material of the structure in the REV can be solid (impermeable) or microporous (permeable). Flows both outside and inside the microporous media are computed simultaneously by using an IB method to solve a combination of the Navier–Stokes equation (outside the microporous medium) and the Zwikker–Kosten equation (inside the microporous medium). The numerical simulation is firstly validated using flow through the REVs of impermeable structures, including square rods, circular rods, cubes, and spheres. The resultant pressure gradient over the REVs is compared with analytical solutions of the Ergun equation or Darcy–Forchheimer law. The good agreements demonstrate the validity of the numerical method to simulate the macroscopic flow behavior in porous media. In addition, with the assistance of a scientific parallel computational library, PETSc, good parallel performances are achieved. Finally, the IB method is extended to simulate species transport by coupling with the REV flow simulation. The species sorption behaviors in an REV with impermeable/solid and permeable/microporous materials are then studied.

Sign in / Sign up

Export Citation Format

Share Document