In the current study we consider the problem of accuracy in heat rate estimations from artificial neural network models of heat exchangers used for refrigeration applications. The network configuration is of the feedforward type with a sigmoid activation function and a backpropagation algorithm. Limited experimental measurements from a manufacturer are used to show the capability of the neural network technique in modeling the heat transfer in these systems. Results from this exercise show that a well-trained network correlates the data with errors of the same order as the uncertainty of the measurements. It is also shown that the number and distribution of the training data are linked to the performance of the network when estimating the heat rates under different operating conditions, and that networks trained from few tests may give large errors. A methodology based on the cross-validation technique is presented to find regions where not enough data are available to construct a reliable neural network. The results from three tests show that the proposed methodology gives an upper bound of the estimated error in the heat rates.
{"title":"Analysis of Fin-Tube Evaporator Performance With Limited Experimental Data Using Artificial Neural Networks","authors":"A. Pacheco-Vega, M. Sen, R. McClain","doi":"10.1115/imece2000-1466","DOIUrl":"https://doi.org/10.1115/imece2000-1466","url":null,"abstract":"\u0000 In the current study we consider the problem of accuracy in heat rate estimations from artificial neural network models of heat exchangers used for refrigeration applications. The network configuration is of the feedforward type with a sigmoid activation function and a backpropagation algorithm. Limited experimental measurements from a manufacturer are used to show the capability of the neural network technique in modeling the heat transfer in these systems. Results from this exercise show that a well-trained network correlates the data with errors of the same order as the uncertainty of the measurements. It is also shown that the number and distribution of the training data are linked to the performance of the network when estimating the heat rates under different operating conditions, and that networks trained from few tests may give large errors. A methodology based on the cross-validation technique is presented to find regions where not enough data are available to construct a reliable neural network. The results from three tests show that the proposed methodology gives an upper bound of the estimated error in the heat rates.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122076848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper we present our model-modifier approach as an economical method for the development of accurate manufacturing equipment models. The model modifier method leverages knowledge from one ANN model to another of a similar type, thus reducing the development effort required as compared to starting from scratch. The economy afforded by this knowledge-sharing technique was evaluated on a Chemical Vapor Deposition (CVD) reactor. The results show that the model-modifier approach is a valid method for transferring knowledge between similar ANN models and that significant savings in training data accrue from this approach. In our case, a highly accurate ANN model was developed with a mere one-fifth of the data that would have been required without this approach. Further, we have also shown that an ANN model developed by the model-modifier approach can be easily and reliably utilized for process optimization.
{"title":"An Economical Method for Artificial Neural Network Process Modeling by the Model-Modifier Approach","authors":"S. Bhatikar","doi":"10.1115/imece2000-1471","DOIUrl":"https://doi.org/10.1115/imece2000-1471","url":null,"abstract":"\u0000 In this paper we present our model-modifier approach as an economical method for the development of accurate manufacturing equipment models. The model modifier method leverages knowledge from one ANN model to another of a similar type, thus reducing the development effort required as compared to starting from scratch. The economy afforded by this knowledge-sharing technique was evaluated on a Chemical Vapor Deposition (CVD) reactor. The results show that the model-modifier approach is a valid method for transferring knowledge between similar ANN models and that significant savings in training data accrue from this approach. In our case, a highly accurate ANN model was developed with a mere one-fifth of the data that would have been required without this approach. Further, we have also shown that an ANN model developed by the model-modifier approach can be easily and reliably utilized for process optimization.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129628591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kyeong K. Lee, T. Brown, G. Dagnall, R. Bicknell-Tassius, A. Brown, G. May
This paper presents the systematic characterization of the molecular beam epitaxy (MBE) process to quantitatively model the effects of process conditions on film qualities. A five-layer, undoped AlGaAs and InGaAs single quantum well structure grown on a GaAs substrate is designed and fabricated. Six input factors (time and temperature for oxide removal, substrate temperatures for AlGaAs and InGaAs layer growth, beam equivalent pressure of the As source and quantum well interrupt time) are examined by means of a fractional factorial experiment. Defect density, x-ray diffraction, and photoluminescence are characterized by a static response model developed by training back-propagation neural networks. In addition, two novel approaches for characterizing reflection high-energy electron diffraction (RHEED) signals used in the real-time monitoring of MBE are developed. In the first technique, principal component analysis is used to reduce the dimensionality of the RHEED data set, and the reduced RHEED data set is used to train neural nets to model the process responses. A second technique uses neural nets to model RHEED intensity signals as time series, and matches specific RHEED patterns to ambient process conditions. In each case, the neural process models exhibit good agreement with experimental results.
{"title":"Neural Network Modeling of Molecular Beam Epitaxy","authors":"Kyeong K. Lee, T. Brown, G. Dagnall, R. Bicknell-Tassius, A. Brown, G. May","doi":"10.1115/imece2000-1470","DOIUrl":"https://doi.org/10.1115/imece2000-1470","url":null,"abstract":"\u0000 This paper presents the systematic characterization of the molecular beam epitaxy (MBE) process to quantitatively model the effects of process conditions on film qualities. A five-layer, undoped AlGaAs and InGaAs single quantum well structure grown on a GaAs substrate is designed and fabricated. Six input factors (time and temperature for oxide removal, substrate temperatures for AlGaAs and InGaAs layer growth, beam equivalent pressure of the As source and quantum well interrupt time) are examined by means of a fractional factorial experiment. Defect density, x-ray diffraction, and photoluminescence are characterized by a static response model developed by training back-propagation neural networks. In addition, two novel approaches for characterizing reflection high-energy electron diffraction (RHEED) signals used in the real-time monitoring of MBE are developed. In the first technique, principal component analysis is used to reduce the dimensionality of the RHEED data set, and the reduced RHEED data set is used to train neural nets to model the process responses. A second technique uses neural nets to model RHEED intensity signals as time series, and matches specific RHEED patterns to ambient process conditions. In each case, the neural process models exhibit good agreement with experimental results.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117354831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The quality of thermoplastic composites depends on the thermal history during processing. Therefore it is important to determine the temperature distribution in the composite during the fabrication process. The objective of this investigation was to develop a comprehensive thermal model of the thermoplastic filament winding process. The model was developed in two parts to calculate the temperature profiles in the towpreg and the composite substrate. A finite element heat transfer analysis for the composite-mandrel assembly was formulated in the polar coordinate system, which facilitates the description of the geometry and the boundary conditions. A four-node ‘sector element’ is used to describe the domain of interest. Sector elements were selected to give a better representation of the curved boundary shape which should improve accuracy with fewer elements compared to a finite element solution in the Cartesian-coordinate system. The second thermal analysis was a Cartesian coordinate, finite element model of the towpreg as it enters the nippoint. The results show that the calculated temperature distribution in the composite substrate compared well with temperature data measured during winding and consolidation. The analysis also agreed with the experimental observation that the melt region is formed on the surface of the incoming towpreg in the nippoint and not on the substrate.
{"title":"Thermal Modeling for the Consolidation Process of Thermoplastic Composite Filament Winding","authors":"A. Loos, X. Song","doi":"10.1115/imece2000-1494","DOIUrl":"https://doi.org/10.1115/imece2000-1494","url":null,"abstract":"\u0000 The quality of thermoplastic composites depends on the thermal history during processing. Therefore it is important to determine the temperature distribution in the composite during the fabrication process. The objective of this investigation was to develop a comprehensive thermal model of the thermoplastic filament winding process. The model was developed in two parts to calculate the temperature profiles in the towpreg and the composite substrate. A finite element heat transfer analysis for the composite-mandrel assembly was formulated in the polar coordinate system, which facilitates the description of the geometry and the boundary conditions. A four-node ‘sector element’ is used to describe the domain of interest. Sector elements were selected to give a better representation of the curved boundary shape which should improve accuracy with fewer elements compared to a finite element solution in the Cartesian-coordinate system. The second thermal analysis was a Cartesian coordinate, finite element model of the towpreg as it enters the nippoint. The results show that the calculated temperature distribution in the composite substrate compared well with temperature data measured during winding and consolidation. The analysis also agreed with the experimental observation that the melt region is formed on the surface of the incoming towpreg in the nippoint and not on the substrate.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115649320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Naphthalene sublimation measurements are made in a rotating dimpled square coolant flow passage with radially-outward flow. The coolant flow passage represents a typical internal cooling channel of a turbine blade. The dimples are in the form of hemispherical depressions and are arranged in staggered rows. In the present study, only the leading and trailing surfaces are dimpled. Measurements are made at a Reynolds number of 7,000 and 21,000 and for Rotation number of 0.2. The measurements indicate that dimples enhance surface mass transfer by a factor of about two compared to a smooth surface. With rotation, the trailing wall mass transfer is increased to nearly twice that of the leading wall mass transfer. Peak mass transfer occurs immediately downstream of the dimples, while the minimum mass transfer occurs in the dimple region itself. Higher mass transfer is also observed along the lateral edges of the dimple. The locations of the Sherwood number peaks suggest the existence of streamwise vortical structures generated from the leading and lateral edges of the dimples.
{"title":"Mass/Heat Transfer in Rotating Dimpled Turbine-Blade Coolant Passages","authors":"S. Acharya, Fuguo Zhou","doi":"10.1115/imece2000-1460","DOIUrl":"https://doi.org/10.1115/imece2000-1460","url":null,"abstract":"\u0000 Naphthalene sublimation measurements are made in a rotating dimpled square coolant flow passage with radially-outward flow. The coolant flow passage represents a typical internal cooling channel of a turbine blade. The dimples are in the form of hemispherical depressions and are arranged in staggered rows. In the present study, only the leading and trailing surfaces are dimpled. Measurements are made at a Reynolds number of 7,000 and 21,000 and for Rotation number of 0.2. The measurements indicate that dimples enhance surface mass transfer by a factor of about two compared to a smooth surface. With rotation, the trailing wall mass transfer is increased to nearly twice that of the leading wall mass transfer. Peak mass transfer occurs immediately downstream of the dimples, while the minimum mass transfer occurs in the dimple region itself. Higher mass transfer is also observed along the lateral edges of the dimple. The locations of the Sherwood number peaks suggest the existence of streamwise vortical structures generated from the leading and lateral edges of the dimples.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126899185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Laser ablation is becoming increasingly important in the fields of micromaching, thin film formation, and bioengineering applications. In laser ablation, the ablation rates and feature quality strongly depend on the size of the breakdown region in the material. This region is characterized by a high density of free electrons, which absorb a large fraction of energy from the laser pulse that results in material vaporization in solids or liquids. For nanosecond- and picosecond pulses, the breakdown region tends to form near the beam focus and then expand back along the beam path toward the laser; this phenomenon is called moving breakdown. For femtosecond pulses, however, breakdown begins up the beam path and then propagates toward the focal point. A moving breakdown model presented by Docchio et al. (1988a) successfully explains and predicts the time-dependent breakdown region in the nanosecond regime, however it does not adequately describe propagation of the breakdown region at pico- and femtosecond time scales. In the present work, a modified moving breakdown model is proposed that includes the pulse propagation and small spatial extent of ultrafast laser pulses. This revised model shows that pulse propagation becomes significant for pulsewidths less than 10 picoseconds. The new model characterizes the pulse behavior as it interacts with a material within the focal volume in both solids and liquids. The model may also be useful in estimating the time- and space-resolved electron density in the interaction volume, the breakdown threshold of a material, shielding effectiveness, energy deposition, and the temperature increase in the material.
{"title":"Modeling of Moving Breakdown by Femtosecond Laser Pulses in Dielectrics","authors":"C. Fan, J. Longtin","doi":"10.1115/imece2000-1476","DOIUrl":"https://doi.org/10.1115/imece2000-1476","url":null,"abstract":"\u0000 Laser ablation is becoming increasingly important in the fields of micromaching, thin film formation, and bioengineering applications. In laser ablation, the ablation rates and feature quality strongly depend on the size of the breakdown region in the material. This region is characterized by a high density of free electrons, which absorb a large fraction of energy from the laser pulse that results in material vaporization in solids or liquids. For nanosecond- and picosecond pulses, the breakdown region tends to form near the beam focus and then expand back along the beam path toward the laser; this phenomenon is called moving breakdown. For femtosecond pulses, however, breakdown begins up the beam path and then propagates toward the focal point. A moving breakdown model presented by Docchio et al. (1988a) successfully explains and predicts the time-dependent breakdown region in the nanosecond regime, however it does not adequately describe propagation of the breakdown region at pico- and femtosecond time scales. In the present work, a modified moving breakdown model is proposed that includes the pulse propagation and small spatial extent of ultrafast laser pulses. This revised model shows that pulse propagation becomes significant for pulsewidths less than 10 picoseconds. The new model characterizes the pulse behavior as it interacts with a material within the focal volume in both solids and liquids. The model may also be useful in estimating the time- and space-resolved electron density in the interaction volume, the breakdown threshold of a material, shielding effectiveness, energy deposition, and the temperature increase in the material.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126986415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
It is now known the generally it can be demonstrated that artificial neural network (ANN), particularly the fully-connected feedforward configuration with backward propagation error-correction routine, can be a rather effective and accurate tool to correlate performance data of thermal devices such as heat exchangers (Sen and Yang, 2000; Kalogirou, 1999). Good examples are the recent demonstrations for the compact fin-tube heat exchangers (Diaz et al., 1999a; Yang et al., 2000; Pacheco-Vega et al., 1999) including those with complex geometries and also two-phase evaporators (Pacheco-Vega et al., 2000) as well as the dynamic modeling of such heat exchangers and their adaptive control (Diaz et al., 1999b; Diaz et al., 2000). Unfortunately, despite such successes, there are still implementation issues of the ANN analysis which lead to uncertainties in its applications and the achieved results. The present paper discusses such issues and the current practices in dealing with them. Those that will be discussed include the number of hidden layers, the number of nodes in each hidden layer, the range within which the input-output data are normalized, the initial assignment of weights and biases, the selection of training data sets, and the training rate. As will be shown, the specific choices are by no means trivial, and yet are rather important in achieving good ANN results in any given application. Since there are no general sound theoretical basis for such choices at the present time, past experience and numerical experimentation are often the best guides. However, many of these choices and issues relating to them involve optimization. As a result. Some of the existing optimization algorithms may prove to be useful and highly desirable in this regard. The current on-going research to provide some rational basis in these issues will also be discussed. Finally, it will also be mentioned that successfully implemented ANNs have many additional uses in practice. Examples include parameter sensitivity analysis, training, design of new experiments, and clustering of data sets.
现在我们知道，一般可以证明，人工神经网络(ANN)，特别是具有反向传播纠错程序的全连接前馈配置，可以是一种相当有效和准确的工具，用于关联热交换器等热设备的性能数据(Sen和Yang, 2000;Kalogirou, 1999)。最近对紧凑型翅片管换热器的演示就是很好的例子(Diaz et al.， 1999a;Yang et al.， 2000;Pacheco-Vega et al.， 1999)，包括那些具有复杂几何形状和两相蒸发器的热交换器(Pacheco-Vega et al.， 2000)，以及此类热交换器及其自适应控制的动态建模(Diaz et al.， 1999b;Diaz et al.， 2000)。不幸的是，尽管取得了这样的成功，但人工神经网络分析仍然存在实施问题，导致其应用和取得的结果存在不确定性。本文讨论了这些问题以及目前处理这些问题的做法。将讨论的内容包括隐藏层的数量、每个隐藏层的节点数量、输入输出数据归一化的范围、权重和偏差的初始分配、训练数据集的选择以及训练率。正如将显示的那样，特定的选择绝不是微不足道的，但是对于在任何给定的应用程序中获得良好的ANN结果是相当重要的。由于目前这种选择还没有普遍可靠的理论依据，过去的经验和数值实验往往是最好的指导。然而，许多这些选择和与之相关的问题都涉及到优化。因此。在这方面，一些现有的优化算法可能被证明是有用的和非常可取的。本文还将对目前正在进行的研究提供一些合理的依据。最后，还将提到成功实施的人工神经网络在实践中有许多其他用途。例子包括参数敏感性分析、训练、新实验设计和数据集聚类。
{"title":"Implementation Issues in Artificial Neural Network Based Thermal Analysis","authors":"K. T. Yang","doi":"10.1115/imece2000-1465","DOIUrl":"https://doi.org/10.1115/imece2000-1465","url":null,"abstract":"\u0000 It is now known the generally it can be demonstrated that artificial neural network (ANN), particularly the fully-connected feedforward configuration with backward propagation error-correction routine, can be a rather effective and accurate tool to correlate performance data of thermal devices such as heat exchangers (Sen and Yang, 2000; Kalogirou, 1999). Good examples are the recent demonstrations for the compact fin-tube heat exchangers (Diaz et al., 1999a; Yang et al., 2000; Pacheco-Vega et al., 1999) including those with complex geometries and also two-phase evaporators (Pacheco-Vega et al., 2000) as well as the dynamic modeling of such heat exchangers and their adaptive control (Diaz et al., 1999b; Diaz et al., 2000). Unfortunately, despite such successes, there are still implementation issues of the ANN analysis which lead to uncertainties in its applications and the achieved results. The present paper discusses such issues and the current practices in dealing with them. Those that will be discussed include the number of hidden layers, the number of nodes in each hidden layer, the range within which the input-output data are normalized, the initial assignment of weights and biases, the selection of training data sets, and the training rate. As will be shown, the specific choices are by no means trivial, and yet are rather important in achieving good ANN results in any given application. Since there are no general sound theoretical basis for such choices at the present time, past experience and numerical experimentation are often the best guides. However, many of these choices and issues relating to them involve optimization. As a result. Some of the existing optimization algorithms may prove to be useful and highly desirable in this regard. The current on-going research to provide some rational basis in these issues will also be discussed. Finally, it will also be mentioned that successfully implemented ANNs have many additional uses in practice. Examples include parameter sensitivity analysis, training, design of new experiments, and clustering of data sets.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125152651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The steady flow and temperature fields around a film cooled inlet guide vane are determined numerically by a CFD method. In particular the outer surface temperatures and heat transfer coefficient distributions are calculated. Static pressure distributions are also presented. The film cooling is achieved by 10 rows of film cooling holes. The computed results are compared with experimental data. The governing equations are solved by a 3D finite-volume Navier-Stokes solver. The low Reynolds number version of the k-ω turbulence model by Wilcox is implemented to enable calculations of turbulent flow cases. A realizability constraint is applied to reduce the generation of unphysical turbulent kinetic energy, particularly close to the leading edge. To handle the film cooling process a special procedure is used. An injection model was implemented in the computer code. This injection model adds the mass flow rate passing through the film cooling holes to the main flow as source terms in the equations for mass, momentum, energy and turbulent kinetic energy. The grid used in the calculations is block-structured, and the total number of grid points is around 250,000. In a related investigation for the same vane geometry but considering pure convective heat transfer, the authors have investigated the importance of the wall thermal boundary condition. Based on this a conjugate heat transfer approach was applied in this paper. The conjugate heat transfer condition means that the heat transfer coefficient distribution is prescribed on the inner surface of the vane and also the wall thickness and thermal conductivity of the vane material are prescribed. The vane outer surface temperature is then found as part of the numerical solution. Some essential parameters in the injection model were varied and the calculated results for the vane outer surface temperature were found to compare favourably with measurements. The static pressure distribution on the vane surface agrees well with experiments. The Mach number distribution provides information of the flow field.
{"title":"Heat Transfer on a Film Cooled Inlet Guide Vane","authors":"Marie-Louise Holmer, L. Eriksson, B. Sundén","doi":"10.1115/imece2000-1459","DOIUrl":"https://doi.org/10.1115/imece2000-1459","url":null,"abstract":"\u0000 The steady flow and temperature fields around a film cooled inlet guide vane are determined numerically by a CFD method. In particular the outer surface temperatures and heat transfer coefficient distributions are calculated. Static pressure distributions are also presented. The film cooling is achieved by 10 rows of film cooling holes. The computed results are compared with experimental data.\u0000 The governing equations are solved by a 3D finite-volume Navier-Stokes solver. The low Reynolds number version of the k-ω turbulence model by Wilcox is implemented to enable calculations of turbulent flow cases. A realizability constraint is applied to reduce the generation of unphysical turbulent kinetic energy, particularly close to the leading edge.\u0000 To handle the film cooling process a special procedure is used. An injection model was implemented in the computer code. This injection model adds the mass flow rate passing through the film cooling holes to the main flow as source terms in the equations for mass, momentum, energy and turbulent kinetic energy.\u0000 The grid used in the calculations is block-structured, and the total number of grid points is around 250,000.\u0000 In a related investigation for the same vane geometry but considering pure convective heat transfer, the authors have investigated the importance of the wall thermal boundary condition. Based on this a conjugate heat transfer approach was applied in this paper. The conjugate heat transfer condition means that the heat transfer coefficient distribution is prescribed on the inner surface of the vane and also the wall thickness and thermal conductivity of the vane material are prescribed. The vane outer surface temperature is then found as part of the numerical solution.\u0000 Some essential parameters in the injection model were varied and the calculated results for the vane outer surface temperature were found to compare favourably with measurements. The static pressure distribution on the vane surface agrees well with experiments. The Mach number distribution provides information of the flow field.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130044501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A 3D simulation of the thermal plasma spraying process is reported. In particular, the effect of the radial injection of a carrier gas is taken into account for a dilute spray. The thermal history of powder particles of different sizes is predicted. It is shown that introduction of a carrier gas can lead to a significant modification of the plasma jet, and can have an effect on the thermal histories of the injected particles. The study is motivated by the processing of non-traditional materials, specifically nanostructured ceramics.
{"title":"Simulation of Thermal Plasma Spraying of Partially Molten Ceramics: Effect of Carrier Gas on Particle Deposition and Phase Change Phenomena","authors":"I. Ahmed, T. Bergman","doi":"10.1115/1.1338117","DOIUrl":"https://doi.org/10.1115/1.1338117","url":null,"abstract":"\u0000 A 3D simulation of the thermal plasma spraying process is reported. In particular, the effect of the radial injection of a carrier gas is taken into account for a dilute spray. The thermal history of powder particles of different sizes is predicted. It is shown that introduction of a carrier gas can lead to a significant modification of the plasma jet, and can have an effect on the thermal histories of the injected particles. The study is motivated by the processing of non-traditional materials, specifically nanostructured ceramics.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"443 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133625275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
It is well known that the dendritic microstructure of alloys is a consequence of morphological instability of the solidification process, which is a result of the coupling of heat and mass transfer with the composition-dependent phase equilibrium condition mediated by the surface energy. There have been many numerical simulations of dendritic solidification. However, many successful simulations of dendritic growth have used non-discrete front tracking method such as artificial source method or phase field method, with demonstrably first order accuracy. Many also found it necessary to continuously inject random noise during simulation. The continuous injection of random noise raises the suspicion that the numerical schemes used may be overly dissipative. The noise is apparently capable of creating nonuniform solidification, but not sufficient to ensure growth with a clear dendritic pattern. In the present study, to rule out the numerical diffusivity as a cause of the damping of dendritic perturbations, artificial perturbations are either not used, or injected only as initial conditions. Under the unstable solidification mode, the initial perturbation triggers the onset of interface instability. Computations were performed for both sub-cooled pure material as well as directional solidification of alloys. The successful simulation of dendritic solidification without the intentional injection of random noise provided evidence that the present method has less numerical diffusion than many existing front tracking methods.
{"title":"Numerical Simulation of Dendritic Solidification","authors":"J. Jung, M. M. Chen","doi":"10.1115/imece2000-1481","DOIUrl":"https://doi.org/10.1115/imece2000-1481","url":null,"abstract":"\u0000 It is well known that the dendritic microstructure of alloys is a consequence of morphological instability of the solidification process, which is a result of the coupling of heat and mass transfer with the composition-dependent phase equilibrium condition mediated by the surface energy. There have been many numerical simulations of dendritic solidification. However, many successful simulations of dendritic growth have used non-discrete front tracking method such as artificial source method or phase field method, with demonstrably first order accuracy. Many also found it necessary to continuously inject random noise during simulation. The continuous injection of random noise raises the suspicion that the numerical schemes used may be overly dissipative. The noise is apparently capable of creating nonuniform solidification, but not sufficient to ensure growth with a clear dendritic pattern. In the present study, to rule out the numerical diffusivity as a cause of the damping of dendritic perturbations, artificial perturbations are either not used, or injected only as initial conditions. Under the unstable solidification mode, the initial perturbation triggers the onset of interface instability. Computations were performed for both sub-cooled pure material as well as directional solidification of alloys. The successful simulation of dendritic solidification without the intentional injection of random noise provided evidence that the present method has less numerical diffusion than many existing front tracking methods.","PeriodicalId":306962,"journal":{"name":"Heat Transfer: Volume 3","volume":"216 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122369749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：