Prediction of Turbulence Statistics Behind a Square Cylinder Using Neural Networks and Fuzzy Logic

2003 ◽  
Vol 125 (2) ◽  
pp. 385-387 ◽  
Author(s):  
P. K. Panigrahi ◽  
Manish Dwivedi ◽  
Vinay Khandelwal ◽  
Mihir Sen
Keyword(s):  
Neural Networks ◽  
Fuzzy Logic ◽  
Reynolds Stresses ◽  
Square Cylinder ◽  
Turbulence Statistics ◽  
Mean Velocity ◽  
Logic Modeling ◽  
The Neural Network ◽  
The Mean ◽  
Computationally Intensive

Experiments are carried out behind a square cylinder mounted in the freestream of a wind tunnel, and hot-wire anemometry is used to determine the profiles of the mean and statistical turbulence quantities. Artificial neural networks and fuzzy-logic models successfully predict the statistical quantities like mean velocity profiles and Reynolds stresses. The fuzzy-logic modeling is more convenient to use, is less computationally intensive, and gives a higher correlation coefficient in comparison to the neural network.

Estimation of Site Amplification from Geotechnical Array Data Using Neural Networks

2021 ◽  
Author(s):  
Daniel Roten ◽  
Kim B. Olsen
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Site Response ◽  
Site Amplification ◽  
Data Driven ◽  
Vertical Array ◽  
The Neural Network ◽  
Simplifying Assumptions ◽  
The Mean ◽  
Fully Connected

ABSTRACT We use deep learning to predict surface-to-borehole Fourier amplification functions (AFs) from discretized shear-wave velocity profiles. Specifically, we train a fully connected neural network and a convolutional neural network using mean AFs observed at ∼600 KiK-net vertical array sites. Compared with predictions based on theoretical SH 1D amplifications, the neural network (NN) results in up to 50% reduction of the mean squared log error between predictions and observations at sites not used for training. In the future, NNs may lead to a purely data-driven prediction of site response that is independent of proxies or simplifying assumptions.

Prediction of Confined Coaxial Turbulent Jet Mixing With a Streamwise Differencing Scheme

1991 ◽  
Author(s):  
M. A. R. Sharif ◽  
M. A. Gadalla
Keyword(s):  
Reynolds Stresses ◽  
Low Velocity ◽  
Mean Velocity ◽  
Jet Mixing ◽  
Air Jet ◽  
Air Stream ◽  
Constant Diameter ◽  
Flow Domain ◽  
The Mean ◽  
Secondary Air

Abstract Isothermal turbulent mixing of an axisymmetric primary air jet with a low velocity annular secondary air stream inside a constant diameter cylindrical enclosure is predicted. The flow domain from the inlet to the fully developed downstream locations is considered. The predicted flow field properties include the mean velocity and pressure and the Reynolds stresses. Different velocity and diameter ratios between the primary and the secondary jets have been investigated to characterize the flow in terms of these parameters. A bounded stream-wise differencing scheme is used to minimize numerical diffusion and oscillation errors. Predictions are compared with available experimental data to back up numerical findings.

Hot Wire Measurements at the Exit of a Centrifugal Compressor Impeller

1979 ◽  
Vol 193 (1) ◽  
pp. 341-347
Author(s):  
A. Goulas ◽  
R. C. Baker
Keyword(s):  
Centrifugal Compressor ◽  
Magnetic Tape ◽  
Reynolds Stresses ◽  
Hot Wire ◽  
Mean Velocity ◽  
Isentropic Flow ◽  
Compressor Impeller ◽  
The Mean

Hot wire measurements at the exit of a small centrifugal compressor impeller are reported. Three different hot wire readings were obtained and stored on a magnetic tape for each point by gating the analogue hot wire signal with a pulse which indicated circumferential position. The combination of the three readings yielded the mean velocity and some Reynolds stresses at each point. The measurements show a ‘jet-wake’ profile towards the shroud and ‘isentropic’ flow near the hub.

A Genetic Algorithm to Improve a Neural Network to Predict a Patient’s Response to Warfarin

1993 ◽  
Vol 32 (01) ◽  
pp. 55-58 ◽  
Author(s):  
M. N. Narayanan ◽  
S. B. Lucas
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Neural Networks ◽  
Predictive Ability ◽  
Predictor Variables ◽  
Optimal Subset ◽  
The Neural Network ◽  
International Normalised Ratio ◽  
The Mean ◽  
Paired T Test

Abstract:The ability of neural networks to predict the international normalised ratio (INR) for patients treated with Warfarin was investigated. Neural networks were obtained by using all the predictor variables in the neural network, or by using a genetic algorithm to select an optimal subset of predictor variables in a neural network. The use of a genetic algorithm gave a marked and significant improvement in the prediction of the INR in two of the three cases investigated. The mean error in these cases, typically, reduced from 1.02 ± 0.29 to 0.28 ± 0.25 (paired t-test, t = −4.71, p <0.001, n = 30). The use of a genetic algorithm with Warfarin data offers a significant enhancement of the predictive ability of a neural network with Warfarin data, identifies significant predictor variables, reduces the size of the neural network and thus the speed at which the reduced network can be trained, and reduces the sensitivity of a network to over-training.

An Explicit Non-Real Time Data Reduction Method of Triple Sensors Hot-Wire Anemometer in Three-Dimensional Flow

1988 ◽  
Vol 110 (2) ◽  
pp. 110-119 ◽  
Author(s):  
Y. T. Chew ◽  
R. L. Simpson
Keyword(s):  
Real Time ◽  
Three Dimensional ◽  
Computation Time ◽  
Reynolds Stresses ◽  
Hot Wire ◽  
Time Data ◽  
Three Dimensional Flow ◽  
Mean Velocity ◽  
Dimensional Flow ◽  
The Mean

An explicit non-real time method of reducing triple sensor hot-wire anenometer data to obtain the three mean velocity components and six Reynolds stresses, as well as their turbulence spectra in three-dimensional flow is proposed. Equations which relate explicitly the mean velocity components and Reynolds stresses in laboratory coordinates to the mean and mean square sensors output voltages in three stages are derived. The method was verified satisfactorily by comparison with single sensor hot-wire anemometer measurements in a zero pressure gradient incompressible turbulent boundary layer flow. It is simple and requires much lesser computation time when compared to other implicit non-real time method.

Hot-Wire Measurements Inside a Centrifugal Fan Impeller

1989 ◽  
Vol 111 (4) ◽  
pp. 363-368 ◽  
Author(s):  
A. Kjo¨rk ◽  
L. Lo¨fdahl
Keyword(s):  
Suction Side ◽  
Design Flow ◽  
Shear Stresses ◽  
Centrifugal Fan ◽  
Reynolds Stresses ◽  
Low Velocity ◽  
Mean Velocity ◽  
Attached Flow ◽  
The Mean ◽  
Velocity Region

Measurements of the three mean velocity components and five of the Reynolds stresses have been carried out in the blade passage of a centrifugal fan impeller. The impeller was of ordinary design, with nine backward curved blades, and all measurements were carried out at the design flow rate. The mean velocity measurements show that the flow can be characterized as an attached flow with almost linearly distributed velocity profiles. However, in a region near the suction side close to the shroud a low velocity region is created. From the turbulence measurements it can be concluded that relatively low values of the turbulent stresses are predominating in the center region of the channel. Closer to the walls higher values of the normal as well as shear stresses are noted.

Separated and Reattached Flow Over Square, Rectangular and Semi-Circular Blocks

2007 ◽  
Author(s):  
M. Agelinchaab ◽  
M. F. Tachie
Keyword(s):  
Boundary Layer ◽  
Cross Sections ◽  
Recirculation Region ◽  
Reynolds Stresses ◽  
Mean Velocity ◽  
Block Height ◽  
Turbulent Statistics ◽  
Image Velocimetry ◽  
The Mean ◽  
Separated And Reattached Flow

A particle image velocimetry is used to study the characteristics of separated and reattached turbulent flow over two-dimensional transverse blocks of square, rectangular and semi-circular cross-sections fixed to the bottom wall of an open channel. The ratio of upstream boundary layer thickness to block height is considerably higher than in prior studies. The results show that the mean and turbulent statistics in the recirculation region and downstream of reattachment are significantly different from the upstream boundary layer. The variation of the Reynolds stresses along the separating streamlines is discussed within the context of vortex stretching, longitudinal strain rate and wall damping. It appears wall damping is a more dominant mechanism in the vicinity of reattachment. The levels of turbulence diffusion and production by the normal stresses are significantly higher than in classical turbulent boundary layers. The bulk of turbulence production occurs in mid-layer and transported into the inner and outer layers. The results also reveal that the curvature of separating streamline, separating bubble beneath it as well as the mean velocity and turbulent quantities depend strongly on block geometry.

Scaling of Favorable Pressure Gradient Turbulent Boundary Layers With Eventual Relaminarization

2005 ◽  
Author(s):  
Rau´l Bayoa´n Cal ◽  
Xia Wang ◽  
Luciano Castillo
Keyword(s):  
Boundary Layer ◽  
Pressure Gradient ◽  
Boundary Layers ◽  
Reynolds Shear Stress ◽  
Velocity Profiles ◽  
Turbulent Boundary Layers ◽  
Reynolds Stresses ◽  
Mean Velocity ◽  
Favorable Pressure Gradient ◽  
The Mean

Applying similarity analysis to the RANS equations of motion for a pressure gradient turbulent boundary layer, Castillo and George [1] obtained the scalings for the mean deficit velocity and the Reynolds stresses. Following this analysis, Castillo and George studied favorable pressure gradient (FPG) turbulent boundary layers. They were able to obtain a single curve for FPG flows when scaling the mean deficit velocity profiles. In this study, FPG turbulent boundary layers are analyzed as well as relaminarized boundary layers subjected to an even stronger FPG. It is found that the mean deficit velocity profiles diminish when scaled using the Castillo and George [1] scaling, U∞, and the Zagarola and Smits [2] scaling, U∞δ*/δ. In addition, Reynolds stress data has been analyzed and it is found that the relaminarized boundary layer data decreases drastically in all components of the Reynolds stresses. Furthermore, it will be shown that the shape of the profile for the wall-normal and Reynolds shear stress components change drastically given the relaminarized state. Therefore, the mean velocity deficit profiles as well as Reynolds stresses are found to be necessary in order to understand not only FPG flows, but also relaminarized boundary layers.

scholarly journals The control to aggregates of pumping stations using a regulator based on a neural network with fuzzy logic

2019 ◽  
Vol 102 ◽  
pp. 03007
Author(s):  
Vladlen Kuznetsov ◽  
Sergey Dyadun ◽  
Valentin Esilevsky
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Fuzzy Logic ◽  
Simulation Modeling ◽  
Neural Controller ◽  
The Neural Network ◽  
Pumping Stations ◽  
Fuzzy Neural ◽  
Pumping Units ◽  
Theory Of Fuzzy Sets

A pumping station control system is considered using a controller based on a fuzzy logic neural network. The simulation of the classical and fuzzy regulators. The possibility of the implementation of the controller in the form of an adaptive multilayer neural network is shown. The use of the theory of fuzzy sets in combination with the theory of neural networks to create a fuzzy-neural regulator to control pumping units provides a promising approach. Simulation modeling and real operation have shown that fuzzy-logic regulators have a number of advantages over classical regulators, which allow the use of form and limitations. Using the neural network model allows you to add the properties of adaptability and learning. The fuzzy-neural controller for controlling pumping units is promising in terms of efficiency and safety by controlling pumping stations.

Neural Networks for Content-Based Image Retrieval

2011 ◽  
pp. 252-272 ◽  
Author(s):  
Brijesh Verma ◽  
Siddhivinayak Kulkarni
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Fuzzy Logic ◽  
Image Retrieval ◽  
Content Based Image Retrieval ◽  
The Neural Network ◽  
Benchmark Database ◽  
Overall Performance ◽  
Critical Literature

This chapter introduces neural networks for Content-Based Image Retrieval (CBIR) systems. It presents a critical literature review of both the traditional and neural network based techniques that are used in retrieving the images based on their content. It shows how neural networks and fuzzy logic can be used in interpretation of queries, feature extraction and classification of features by describing a detailed research methodology. It investigates a neural network based technique in conjunction with fuzzy logic to improve the overall performance of the CBIR systems. The results of the investigation on a benchmark database with a comparative analysis are presented in this chapter. The methodologies and results presented in this chapter will allow researchers to improve and compare their methods and it will also allow system developers to understand and implement the neural network and fuzzy logic based techniques for content based image retrieval.

Sign in / Sign up

Export Citation Format

Share Document