Coupled CFD-ANN Procedure for Extending Heat Transfer Correlations Out of Their Range of Validity

2012 ◽  
Author(s):  
Andrea Viano ◽  
Gabriele Ottino ◽  
Luca Ratto ◽  
Giuseppe Spataro
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Experimental Data ◽  
Nusselt Number ◽  
Gas Turbine ◽  
Training Session ◽  
Nonlinear Problems ◽  
Artificial Neural ◽  
Cfd Analyses ◽  
Good Agreement

The heat transfer coefficient and pressure losses are among the main parameters to be evaluated in gas turbine cooling network design. Due to the complexity of these estimates, correlation-based computations are typically used as a result of time-consuming and expensive experimental activities. One of the main problems that the industry has to face is that these correlations, based on non-dimensional experimental data, produce reliable results in a range of validity typically different from that encountered in gas turbine applications. This paper will present preliminary results of an innovative procedure based on CFD analyses and Artificial Neural Networks, able to extend correlation predictions out of their range of validity, without any additional experimental data. Well-known test cases were replicated by building corresponding CAD geometries which were discretized by means of appropriate meshes, resulting from grid-independence studies. CFD analyses, based on the RANS approach, were performed to overlay the computations of the Nusselt number obtained from experimental activities. A preliminary comparison among turbulence models was carried out to find one leading to a good agreement with the experimental data. Then, an optimization method, based on Evolutionary Algorithms, was applied to the CFD analyses in order to find the best set of constant values for the chosen turbulence model, leading to the most accurate prediction of the experimental dataset. The resulting ad hoc CFD model was adopted in order to analyse test case configurations characterized by parameters within and external to the correlation validity field, building a sufficiently wide feeding database. A feed-forward multi-layer neural network was selected among network architectures typically used in engineering applications for prediction analyses. ANNs were chosen because they enable the solution of these complex nonlinear problems by using simple computational operations. The selected Artificial Neural Network was trained by a back-propagation procedure on the CFD results regarding Nusselt number. The validation of the resulting ANN was performed comparing its outputs with experimental data external to the correlation range of validity, which had not been used in the training session. Good agreement has been found. Results are presented and discussed.

scholarly journals Using artificial neural network for predicting heat transfer coefficient during flow boiling in an inclined channel

2020 ◽  
pp. 238-238
Author(s):  
Adel Bouali ◽  
Salah Hanini ◽  
Brahim Mohammedi ◽  
Mouloud Boumahdi
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Experimental Data ◽  
Artificial Neural Network ◽  
Heat Transfer Coefficient ◽  
Transfer Coefficient ◽  
Flow Boiling ◽  
Low Flow ◽  
Inclined Channel ◽  
Artificial Neural

The flow and heat transfer characteristics in a nuclear power plant in the event of a serious accident are simulated by boiling water in an inclined rectangular channel. In this study an artificial neural network model was developed with the aim of predicting heat transfer coefficient (HTC) for flow boiling of water in inclined channel, the network was designed and trained by means of 520 experimental data points that were selected from within the literature. orientation ,mass flux, quality and heat flow which were employed to serve as variables of input of multiple layer perceptron (MLP) neural network, whereas the analogous HTC was selected to be its output. Via the method of trial-and-error, MLP network with 30 neurons in the hidden layer was attained as optimal ANN structure. The fact that is was enabled to predict accurately the HTC. For the training set, the mean relative absolute error (MRAE) is about 0.68 % and the correlation coefficient (R) is about 0.9997. As for the testing and validation set they are respectively about 0.60 % and 0.9998 and about 0.79 % and 0.9996. The comparison of the developed ANN model with experimental data and empirical correlations in vertical channel under the low flow rate and low quality shows a good agreement.

Modelling the Natural Evaporation of the Concentrated Seawater after Desalinized

2015 ◽  
Vol 713-715 ◽  
pp. 2989-2992
Author(s):  
Xue Kui Wang ◽  
Ying Zhou ◽  
Ling Li ◽  
Tian Cheng Gao ◽  
Na Tang
Keyword(s):  
Neural Network ◽  
Experimental Data ◽  
Artificial Neural Network ◽  
Relative Humidity ◽  
Evaporation Rate ◽  
Evaporation Model ◽  
Irradiation Intensity ◽  
Artificial Neural ◽  
Good Agreement

The influence of natural evaporation factors (the irradiation intensity, speed of the wind, temperature of the brine, temperature and relative humidity of the air) on the desalinated seawater evaporation rate was measured experimentally. A natural evaporation model was built by correlating the experimental data using the artificial neural network. This model was well correlated with the influence of natural evaporation factors, and it showed a good agreement of the results and evaporation theory.

Computational Modelling of Tip Heat Transfer to a Super-Scale Model of an Unshrouded Gas Turbine Blade

2008 ◽  
Author(s):  
Brian M. T. Tang ◽  
Pepe Palafox ◽  
David R. H. Gillespie ◽  
Martin L. G. Oldfield ◽  
Brian C. Y. Cheong
Keyword(s):  
Heat Transfer ◽  
Experimental Data ◽  
Nusselt Number ◽  
Gas Turbine ◽  
Turbine Blade ◽  
Computational Study ◽  
Turbulence Models ◽  
Scale Model ◽  
Tip Leakage Flow ◽  
Blade Tip

Control of over-tip leakage flow between turbine blade tips and the stationary shroud is one of the major challenges facing gas turbine designers today. The flow imposes large thermal loads on unshrouded high pressure turbine blades and is significantly detrimental to turbine blade life. This paper presents results from a computational study performed to investigate the detailed blade tip heat transfer on a sharp-edged, flat tip HP turbine blade. The tip gap is engine representative at 1.5% of the blade chord. Nusselt number distributions on the blade tip surface have been obtained from steady flow simulations and are compared to experimental data carried out in a super-scale cascade, which allows detailed flow and heat transfer measurements in stationary and engine representative conditions. Fully structured, multiblock hexahedral meshes were used in the simulations, performed in the commercial solver Fluent. Seven industry-standard turbulence models, and a number of different tip gridding strategies are compared, varying in complexity from the one-equation Spalart-Allmaras model to a seven-equation Reynolds Stress model. Of the turbulence models examined, the standard k-ω model gave the closest agreement to the experimental data. The discrepancy in Nusselt number observed was just 5%. However, the size of the separation on the pressure side rim was underpredicted, causing the position of reattachment to occur too close to the edge. Other turbulence models tested typically underpredicted Nusselt numbers by around 35%, although locating the position of peak heat flux correctly. The effect of the blade to casing motion was also simulated successfully, qualitatively producing the same changes in secondary flow features as were previously observed experimentally, with associated changes in heat transfer to the blade tip.

scholarly journals Numerical Investigation of the Use of a New Nano-Particle in Microchannel

2021 ◽  
Author(s):  
Fuat Kaya
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Artificial Neural Network ◽  
Boron Nitride ◽  
Nano Particle ◽  
Flow And Heat Transfer ◽  
Comparison Research ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Good Agreement

Abstract The purpose of this paper is to study the effects of the use of Boron nitride (BN) as nano-particle on pressure drop and heat transfer in a microchannel. The governing equations for the fluid flow were solved by using Fluent CFD code and artificial neural network (ANN). Computational results acquired from Fluent CFD code and artificial neural network (ANN) for alumina (Al2O3) as nano-particle were compared with numerical values obtained in the literature for validation. On the basis of a water-cooled (only water, water+alumina and water+boron nitride) smooth microchannel were designed, and then the corresponding laminar flow and heat transfer were studied numerically. Results derived from the numerical tests (NT) and artificial neural network (ANN) show good agreement with the values mentioned in the literature and these results also show by the comparison research which was conducted considering the heat transfer and pressure loss parameters between BN and widely used alumina that BN is more convenient nano-particle.

Modeling of mixed convection in an enclosure using multiple regression, artificial neural network, and adaptive neuro-fuzzy interface system models

2020 ◽  
Vol 234 (15) ◽  
pp. 3078-3093
Author(s):  
Ece Aylı
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Artificial Neural Network ◽  
Nusselt Number ◽  
Forced Convection ◽  
Number Distribution ◽  
Neuro Fuzzy ◽  
Interface System ◽  
Nusselt Number Distribution ◽  
Artificial Neural

In this study, the heat transfer characteristics of laminar combined forced convection through a horizontal duct are obtained with the help of the numerical methods. The effect of the geometrical parameters of the cavity and Reynolds number on the heat transfer is investigated. New heat transfer correlation for hydrodynamically fully developed, laminar combined forced convection through a horizontal duct is proposed with an average error of 6.98% and R2 of 0.8625. The obtained correlation results are compared with the artificial neural network and adaptive neuro-fuzzy interface system models. Due to the obtained results, good agreement is identified between the numerical results and predicted adaptive neuro-fuzzy interface system results. In conclusion, it is seen that adaptive neuro-fuzzy interface system can predict the Nusselt number distribution with a higher accuracy than the developed correlation and the artificial neural network model. The developed adaptive neuro-fuzzy interface system model predicts the Nusselt number with 1.07% mean average percentage error and 0.9983 R2 value. The effect of the different training algorithms and their ability to predict Nusselt number distribution are examined. According to the results, the Bayesian regulation algorithm gives the best approach with a 2.235% error. According to the examination that is performed in this study, the adaptive neuro-fuzzy interface system is a powerful, robust tool that can be used with confidence for predicting the thermal performance.

Computational Modeling of Tip Heat Transfer to a Superscale Model of an Unshrouded Gas Turbine Blade

2010 ◽  
Vol 132 (3) ◽  
Author(s):  
Brian M. T. Tang ◽  
Pepe Palafox ◽  
Brian C. Y. Cheong ◽  
Martin L. G. Oldfield ◽  
David R. H. Gillespie
Keyword(s):  
Heat Transfer ◽  
Experimental Data ◽  
Nusselt Number ◽  
Gas Turbine ◽  
Turbine Blade ◽  
Computational Study ◽  
Turbine Blades ◽  
Turbulence Models ◽  
Tip Leakage Flow ◽  
Blade Tip

Control of over-tip leakage flow between turbine blade tips and the stationary shroud is one of the major challenges facing gas turbine designers today. The flow imposes large thermal loads on unshrouded high pressure (HP) turbine blades and is significantly detrimental to turbine blade life. This paper presents results from a computational study performed to investigate the detailed blade tip heat transfer on a sharp-edged, flat tip HP turbine blade. The tip gap is engine representative at 1.5% of the blade chord. Nusselt number distributions on the blade tip surface have been obtained from steady flow simulations and are compared with experimental data carried out in a superscale cascade, which allows detailed flow and heat transfer measurements in stationary and engine representative conditions. Fully structured, multiblock hexahedral meshes were used in the simulations performed in the commercial solver FLUENT. Seven industry-standard turbulence models and a number of different tip gridding strategies are compared, varying in complexity from the one-equation Spalart–Allmaras model to a seven-equation Reynolds stress model. Of the turbulence models examined, the standard k-ω model gave the closest agreement to the experimental data. The discrepancy in Nusselt number observed was just 5%. However, the size of the separation on the pressure side rim was underpredicted, causing the position of reattachment to occur too close to the edge. Other turbulence models tested typically underpredicted Nusselt numbers by around 35%, although locating the position of peak heat flux correctly. The effect of the blade to casing motion was also simulated successfully, qualitatively producing the same changes in secondary flow features as were previously observed experimentally, with associated changes in heat transfer with the blade tip.

Prediction of Post-Dryout Heat Transfer in Vertical Annular Channels Using Artificial Neural Network Method

2008 ◽  
Author(s):  
D. W. Zhao ◽  
G. H. Su ◽  
S. Z. Qiu ◽  
W. X. Tian
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Artificial Neural Network ◽  
Nusselt Number ◽  
Annular Channel ◽  
Experimental Investigations ◽  
Gap Size ◽  
Annular Channels ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Experimental investigations on post-dryout heat transfer in 10×8.1, 10×7 and 10×6mm annular test sections have been carried out under low-pressure and low mass flow rate conditions. An Artificial Neural Network (ANN) was trained successfully based on the experimental data for predicting the average post-dryout Nusselt number. Based on the ANN, the effects of gap size, pressure, steam Reynolds number, Reg, inlet quality, xi, Prandtl number, (Prg)W, and the ratio of heat flux of inner-tube to that of outer-tube, qi/qo, on post-dryout heat transfer were analyzed, respectively. In present study, Nusselt number in annular channels with big gap size is larger than that in annular channels with small gap size. Nusselt number increases significantly in 1.5mm and 2.0mm annular channels while it is almost constant in 0.95mm annular channel with increasing pressure or qi/qo. Nusselt number increases with Reg in case of 0.95mm and 1.5mm gap sizes. However, Nusselt number in 2.0mm annular channel firstly increases and then decreases with increasing Reg. Nusselt number decreases with increasing inlet quality under all three annular channels condition. Nusselt number decreases significantly with increasing (Prg)W when (Prg)W is less than 1.5. The changes of Nusselt number in 1.5mm or 2.0mm annular channels are larger than that in 0.95mm annular channel.

scholarly journals Modeling of overall heat transfer coefficient of a concentric double pipe heat exchanger with limited experimental data by using curve fitting and ANN combination

2019 ◽  
Vol 23 (6 Part A) ◽  
pp. 3579-3590 ◽  
Author(s):  
Necati Kocyigit ◽  
Huseyin Bulgurcu
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Experimental Data ◽  
Artificial Neural Network ◽  
Heat Transfer Coefficient ◽  
Heat Exchanger ◽  
Transfer Coefficient ◽  
Curve Fitting ◽  
Overall Heat Transfer Coefficient ◽  
Artificial Neural

The modeling accuracy of artificial neural networks (ANN) was evaluated by using limited heat exchanger data acquired experimentally. The artificial neural networks were used for predicting the overall heat transfer coefficient of a concentric double pipe heat exchanger where oil flowed inside the inner tube while the water flowed in the outer tube. In the cases of parallel and counter flows, the experimental data were collected by testing heat exchanger in wide range of operating conditions. Curve fitting and artificial neural network combination was used for the estimation of the overall heat transfer coefficient to compensate the experimental errors in the data. The curve fitting was used to detect the trend and generate data points between the experimentally collected points. The artificial neural network was trained better from the generated data set. The feed forward type artificial neural network was trained by using the Levenberg-Marquardt algorithm. Two backpropagation network type artificial neural network algorithms were also used, and their performance were compared with the estimation of the Levenberg-Marquardt algorithm. The average estimation error between the predictions and the experimental data were in the range of 1.31e?4 to 4.35e?2%. The study confirmed that curve fitting and artificial neural network combination could be used effectively to estimate the overall heat transfer coefficient of heat exchanger.

Heat Flow in the Gas Turbine

1948 ◽  
Vol 159 (1) ◽  
pp. 245-254 ◽  
Author(s):  
A. G. Smith
Keyword(s):  
Heat Transfer ◽  
Experimental Data ◽  
Nusselt Number ◽  
Heat Flow ◽  
Gas Turbine ◽  
Internal Flow ◽  
Considerable Effect ◽  
Air Cooling ◽  
Transfer Coefficients ◽  
Liquid Cooling

In this lecture experimental data are presented on turbine-blade heat transfer, which is an important factor in the design of high-temperature turbines. Blade stagger has a considerable effect on blade Nusselt number and on the rate of variation of Nusselt number with Reynolds number. For nozzleblades, theoretical and experimental values of the Nusselt number agree fairly well. A method is proposed for the approximate estimation of Nusselt number by Keynolds analogy. The experimental data are used in the calculation of heat flow in blades cooled at the root, and in internally air-cooled blades. Other methods of cooling have been proposed, including partial admission of cold air or liquid through the nozzle ring; the theoretically advantageous “sweat” cooling by exuding coolant air through a porous skin, or the injection of coolant air through slits; root cooling might include modifications, such as effective increase of blade conductivity by tubes of liquid embedded in the blade or reduction, by an insulating layer, of the amount of heat conducted to the blades; internal liquid cooling by forced or convective circulation is a possible alternative to internal air-cooling. Some of these methods have been successfully used-for example, internal air-cooling of hollow blades in German jet engines; internal liquid-cooling in an experimental German turbine; and a certain degree of “root” cooling in British aircraft turbines. Theoretical analysis of the problem of turbine cooling has been rendered difficult, in the past, by absence of data on blade heat-transfer coefficients. For root-cooled blades of heat-resisting alloy the cooled region is small, but it would be satisfactorily large for nozzle blades of high-conductivity material. Internal air-cooling needs only small quantities of coolant, provided a high internal blade conductance is achieved by subdividing the internal flow passage. The effectiveness of root-cooling diminishes with increase of scale; on the other hand, the relative cooling air quantity necessary to cool a blade internally diminishes with increase of scale. The heat exchanger is a bulky and expensive component of most “plant” gas-turbine systems, and with the aim of size reduction “regenerative” type exchangers have been investigated for some time as an alternative to the bulky recuperative type. The lecture deals with the advantage in bulk of the regenerator, and gives some relevant data on heat-transfer in laminar flow

Sign in / Sign up

Export Citation Format

Share Document