Prediction Of Heat Transfer Coefficient Research Articles

Transfer learning model to predict flow boiling heat transfer coefficient in mini channels with micro pin fins

Flow boiling in mini channels with micro pin fins is a promising heat sink technique to achieve high-efficient aircraft thermal management. The accurate prediction of its heat transfer coefficient is critical for the practical design of two-phase heat exchanger based on mini channels with micro pin fins. Previous investigation shows that heat transfer coefficient prediction accuracy of the machine learning method is generally better than that of conventional empirical correlations. However, the machine learning method cannot guarantee its prediction accuracy among different data domains. To extend the application region of the conventional machine learning method, a transfer learning framework was proposed in present study. First of all, an experimental system was built to acquire test data from different sample domains (i.e., the diamond pin fins with different geometries). Then a conventional machine learning model was developed based on the deep learning method. Furthermore, the developed deep learning model was adjusted with transfer learning process, and the performance of these two kinds of models (i.e., conventional machine learning model and transfer learning model) was comprehensively evaluated. Results showed that the conventional machine learning model had a good prediction accuracy with an overall deviation of 4.11 % but was only confined in the same data domain as training data. Differently, the transfer learning model with 70 % data set from the new domain can achieve appropriate prediction in the new domains with deviation of 4.28 %. The results of this paper demonstrated that the conventional machine learning model can extend into different domains with reasonable prediction accuracy through transfer learning frameworks. It will benefit the practical design of high-efficiency heat exchangers for aircraft thermal management.

Ammonia heat pumps for district heating: thermo-economic optimization analysis based on evaporator geometry for either direct expansion or chilled water configurations

Electric heat pumps are recognized as a key technology for decarbonization and have received increasing policy support in several countries over the last few years. These devices offer a highly efficient form of electric heating and could make an important contribution to the transition to a low carbon future, especially in case of district heating systems. Within this context, this paper presents a thermo-economic optimization analysis of a 7.35 MW electric heat pump system employed for district heating purposes and using ammonia as working fluid. The implemented code gathers multiple sub-models calibrated ad-hoc from real data or manufacturers’ datasheets. The heat exchangers are simulated by considering phenomenological equations and well-known heat transfer coefficient and pressure drop prediction methods. The optimization analysis is performed at constant condensation temperature that guarantees in/out hot water temperatures of 30/62 °C and is focused on the matching between compressor and evaporator heat exchanger, by considering either a direct expansion system or a chilled water heat pump. Both the coefficient of performance (COP) and the set-up costs are considered as optimization performance indicators for the construction of the Pareto front.

Statistical and machine learning applied to the universal consolidated database to predict heat transfer coefficient in flow condensation

Due to complex physical mechanisms, and a general lack of accurate modeling techniques, two phase convective heat transfer is hard to characterize. In this article we show the general effectiveness of statistical and machine learning techniques for predicting two-phase heat transfer coefficients. This is a comprehensive study that: performs a full exploratory data analysis (EDA) to determine the range by which we may apply these models, derives an optimal data set as a basis for predicting heat transfer coefficient, tests all models on an independent data set, and assesses model performance on multiple metrics. We find that random forest methods, gradient boosted machines, and multi-layer perceptrons (deep neural networks) perform the best in predicting heat transfer coefficient on the given data set. We find that in general, the increased data base predicts better due to the inclusion of a few important thermophysical and flow properties.

Artificial neural network-based modeling for the prediction of heat and mass transfer coefficient of the adiabatic liquid desiccant system

This work, based on the data obtained from the literature reported by (Varela et al., 2018), aims to use the artificial neural network approach to predict the heat and mass transfer in a dehumidifier system, using lithium chloride as a liquid desiccant. A neural network model was developed in MATLAB environment based on multilayer perceptron that included an input, hidden, and output layer. The network input parameters are air velocity, air temperature, air humidity ratio, liquid desiccant temperature, liquid flow rate, and liquid desiccant concentration. The network output includes two variables which are the heat transfer coefficient (Kh) and mass transfer coefficient (Km). The performance of the ANN model was evaluated using the statistical parameters between the prediction results and experimental values. The performance regression yields R2 and MSE values of 0.9344 and 9.0032, respectively, for the test data set of heat transfer coefficient (Kh). Moreover, for the mass transfer coefficient (Km), the regression parameter R2 and MSE values for the ANN tests were found to be 0.9657 and 2.0414, respectively. In addition, air velocity, air temperature, solution mass flow rate, and solution concentration are the most influential parameters on the heat and mass transfer between the air and liquid desiccant.

Assessment of machine learning models and conventional correlations for predicting heat transfer coefficient of liquid hydrogen during flow boiling

Liquid hydrogen has attracted widespread attention due to industrial decarbonization as an excellent carrier of renewable energy accommodation. Accurate prediction of the heat transfer coefficient during flow boiling has become a fundamental prerequisite for enhancing the design rationality and safety of hydrogen devices since a slight heat leak will trigger flow boiling. A novel selection standard of the dataset for liquid hydrogen flow boiling is proposed, and an experimental database of liquid hydrogen is compiled, including 923 data points from 5 sources. The prediction performance of eight conventional empirical correlations and six machine learning models for liquid hydrogen flow boiling is compared and evaluated. Results show that the prediction ability of machine learning models for the heat transfer coefficient of liquid hydrogen flow boiling is much higher than that of traditional correlations. Fang correlation is recommended to use as the calculation benchmark due to simplicity and reliability if the inlet condition is obscure. The Extra tree model shows the most excellent evaluating performance among the six machine learning models with a mean absolute deviation of 11.44% and a fairly superior R2 of 0.9543. The Froude number Fr exhibits an essential impact according to the feature importance analysis, which may need to be taken into account when developing new empirical correlations. An improved correlation for the flow boiling of liquid hydrogen is provided based on the Fang correlation, achieving a mean absolute deviation of 18.61% and a mean relative deviation of 16.49%. The results of this paper can offer some theoretical guidance for the design of a liquid hydrogen apparatus.

Prediction of subcooled flow boiling local heat transfer coefficient in microchannel heatsinks

This study aims to evaluate previous models that describe the average heat transfer coefficient (HTC) of subcooled flow boiling in microchannels and modify them to predict the local HTC. To enrich and extend the databases of these models, three-dimensional computational fluid dynamics simulations of HFE-7100, ethanol, and water in a wide range of heat fluxes were performed. The effect of the inlet degree of subcooling, or the difference between saturation and bulk fluid temperatures, on thermal performance is first investigated. Following that, the effect of the channel height-to-width ratio is investigated by examining four microchannels of different sections. Although decreasing the inlet subcooling tends to increase the overall HTC, it results in a lower maximum allowable heat flux due to the increase in bubble coverage of the bottom wall. A greater height-to-width ratio may improve the HTC, but there is an optimum ratio beyond which thermal performance degrades. Finally, the predictability of five models describing the average HTC of the subcooled flow boiling region is investigated, and adjusted correlations for accurate local HTC prediction are presented. A new correlation predicting the length of the onset of nucleate boiling is also demonstrated.

Experimental investigation on heat transfer characteristics of upward flow boiling in a vertical narrow rectangular channel

ABSTRACT Heat transfer experiment on water flow boiling was carried out in the upward direction under atmospheric pressure through a narrow rectangular channel heated on one-side by vapor having a gap of 2.75 mm, a width of 250 mm, and length of 1400 mm. The heat transfer coefficient and thermal hydraulic thresholds of the flow boiling in forced convective flow, such as the onset of nucleate boiling (ONB) and onset of fully developed nucleate boiling (OFDB) were investigated. The experiment was performed over a wide range of inlet temperature (75–95°C), mass flow rate (2–9 g/m), and heat fluxes (5–16 kW/m2). The effects of the parameters on the heat transfer coefficients have been discussed in detail. A series of ONB, FDB and heat transfer correlations were evaluated using the experimental data, and most of the correlations did not adequately fit the experimental results. A modified Hong et al. correlation and modified Zhu et al. correlation were used to predict the wall superheat at ONB and FDB respectively. The average errors (AEs) of the two correlations were 2.36% and −0.68%, and the root mean square errors (RMSEs) of them were both 14.00%. A modified Li and Wu correlation was used to predict heat transfer coefficients with the average error (AE) of 1.46% and the root mean square errors (RMSE) of 11.88%.

Research on prediction and analysis of supercritical water heat transfer coefficient based on support vector machine

In order to better perform thermal hydraulic calculation and analysis of supercritical water reactor, based on the experimental data of supercritical water, the model training and predictive analysis of the heat transfer coefficient of supercritical water were carried out by using the support vector machine (SVM) algorithm. The changes in the prediction accuracy of the supercritical water heat transfer coefficient are analyzed by the changes of the regularization penalty parameter C, the slack variable epsilon and the Gaussian kernel function parameter gamma. The predicted value of the SVM model obtained after parameter optimization and the actual experimental test data are analyzed for data verification. The research results show that: the normalization of the data has a great influence on the prediction results. The slack variable has a relatively small influence on the accuracy change range of the predicted heat transfer coefficient. The change of gamma has the greatest impact on the accuracy of the heat transfer coefficient. Compared with the calculation results of traditional empirical formula methods, the trained algorithm model using SVM has smaller average error and standard deviations. Using the SVM trained algorithm model, the heat transfer coefficient of supercritical water can be effectively predicted and analyzed.

Experimental study on condensate heat transfer coefficient of multi-channel cylinder dryer integrated with Bayesian-optimized machine learning prediction

The condensation heat transfer coefficient of the multi-channel dryer, one of the key component of paper-making machine, directly determines the efficiency of heat energy utilization. However, the prediction of condensation heat transfer coefficient remains a challenge because the heat transfer characteristics in multi-channel dryer is a complex fundamental issue involving the thermal behavior of two-phase fluid systems. Herein, we successfully developed the four supervised machine learning models to predict the heat transfer coefficient of a multi-channel cylinder dryer under different working conditions. The multi-channel cylinder dryer experiments under different steam mass flux and cooling water mass flow rates were performed and the measured data is used as the input data for training. Interestingly, the four trained Bayesian-optimized machine models present the excellent capability of prediction for condensation heat transfer coefficient of multi-channel cylinder dryer, where the values of R 2 for tested Bayesian-optimized-based SVR, ANN, linear SVR, and RF are 0.983, 0.997, 0.996, and 0.953, respectively. In addition, the feature importance of descriptors is quantified based on a random forest algorithm. Our study suggests that machine learning models can effectively predict the condensate heat transfer coefficient of two-phase fluid systems, which not only would be beneficial to optimizing the structures and operation parameters of multi-channel cylinder dryer in the industry but to develop a reasonable correlation of heat transfer coefficient in fundamental research.

Prediction of heat transfer coefficients for steam condensation in the presence of air based on ANN method

Artificial neural network (ANN) methods have been gradually used in the field of nuclear reactor thermal-hydraulics as new methods to improve accuracy or fast prediction. This study establishes a back propagation (BP) neural network model based on the ANN methods to predict the steam condensation heat transfer coefficient outside a heat tube in the presence of air. The main factors affecting condensing heat transfer, such as pressure, air mass fraction, subcooling, and tube diameter, are used as input quantities, and the condensation heat transfer coefficient is used as output quantity. A complete set of neural networks for predicting the heat transfer coefficient for steam condensation in the presence of air is established based on the relevant experimental data collected over the world. The results predicted by the ANN model are compared with the experimental data and those of traditional correlation methods. The data from 2276 experiments are distributed within a ±10% error band at the 95% confidence level. This means that the prediction accuracy of the ANN model is higher than that of the traditional experimental correlation. Therefore, the neural network model developed in this study can be used for the prediction of the heat transfer coefficient for steam condensation in the presence of air.

Experimental study on convective heat transfer characteristics of supercritical carbon dioxide in a vertical tube with low mass flux

A thorough understanding of the heat transfer behaviour of supercritical CO2 (sCO2) flowing through channels is crucial for designing safe and effective heat exchangers in sCO2 power cycles. Existing findings on convective heat transfer of sCO2 in tubes with low mass flux have not reached a consistent conclusion, further research on the complex mechanism of convective heat transfer of sCO2 is still needed. This study experimentally investigates the convective heat transfer of sCO2 as it flows upward and downward in a 10 mm diameter tube with pressures of 7.4–10 MPa, mass fluxes of 100–350 kg/(m2 s), and heat fluxes of 10–130 kW/m2. The effects of heat flux, mass flux, pressure and flow orientation on heat transfer have been examined. The results show that a rising heat flux impairs heat transfer, and a wall temperature valley appears in the heat transfer enhancement (HTE) region at high heat flux for both upward and downward flow. An increase in mass flux enhances heat transfer, while an increase in pressure harms heat transfer. Downward flow exhibits better heat transfer performance compared to upward flow under the same conditions. The mechanism of the wall temperature valley in the HTE region was analyzed based on the velocity, turbulent kinetic energy and thermophysical property distribution of sCO2 in the tube, attributing to the effects of buoyancy and specific heat. In addition, new heat transfer correlations were proposed for the upward and downward sCO2 flow with similar heat transfer behaviour at low mass flux, with over 90 % of the predicted heat transfer coefficients falling within a relative error range of ±20 %.

Prediction of Friction Factor and Heat Transfer Coefficient for Single-Phase Forced Convection Inside Microfin Tubes

Microfin tubes are widely used to enhance heat transfer in heat exchangers in order to reduce volumes, costs and refrigerant charge. Much experimental work has been published for the flow of liquids, while some experimental work is available for the flow of gases for the chemical, refrigeration and air conditioning industry. This work reviews the literature and presents new experimental friction factors for the flow of the superheated vapor of R1234ze(E) in a 5 mm outside diameter microfin tube. The authors have also collected an extensive data bank of heat transfer coefficients (around 648 points from different research laboratories) and friction factors (around 536 points), covering 45 different geometries of inner finned tubes. After comparing the predictions from available correlations with experimental data, the present paper suggests the best performing equations for the calculation of the friction factor and of the Nusselt number during forced convection flow of liquids and gases. The suggested model for friction factor estimates the experimental values with a relative and absolute deviation of −0.3% and 7.9%, respectively, whereas the suggested model for the heat transfer coefficient predicts the experimental data bank with a relative and absolute deviation of −3.3% and 13.9%, respectively. The validity range of the two correlations is extremely wide, covering microfin tubes with diameters from 2.6 mm to 24.4 mm, and Reynolds number from about approximately 1000 to 300,000 for the friction factor, and from 3000 to 1,000,000 for the heat transfer coefficient.

A novel machine-learning schemes to predict heat transfer coefficient during condensation of CO2 in porous media

A novel machine-learning schemes to predict heat transfer coefficient during condensation of CO2 in porous media

Prediction of heat transfer coefficient and pressure drop of R1234yf and R134a flow condensation in horizontal and inclined tubes using machine learning techniques

Machine learning techniques have great potential to predict two-phase flow characteristics instead of classic empirical correlations. In the present study, four different machine learning models, including multi-layer perceptron artificial neural network (ANNMLP), support vector regression (SVR), K nearest neighbors (KNN), and extreme gradient boosting (XGBoost), are employed to predict the heat transfer coefficient (HTC) and the frictional pressure drop (FPD) of R134a and R1234yf condensation flow in horizontal and inclined tubes. The dataset includes 348 points from previous works and the current research. To extend the data, an experimental study is also performed on the condensation of R134a in a horizontal tube for different mass velocities and vapor qualities. The results show that, in the best model, HTC can be estimated by ANNMLP with the mean absolute percentage error (MAPE) of 7.01%. The best prediction of FPD is achieved using XGBoost machine with MAPE of 10.87% on test data. Also, the feature importance procedure is implemented to recognize the most useful features. Based on the results, the mass velocity and the inclination angle are identified as the most influencing parameters on the prediction of HTC and FPD, respectively.

Numerical and experimental analysis of LBE flow and heat transfer characteristics in hexagonal rod bundles

Lead-bismuth eutectic (LBE) is a liquid with low Prandtl number and high thermal conductivity, generally used as the coolant in lead-cooled fast reactors of Generation-IV. The traditional Reynolds analogy is not appropriate for LBE, hence the theoretical analysis and experimental measurements are studied to interpret the flow and heat transfer of LBE in this article. In this way, a single-rod experiment on flow and heat transfer for LBE is performed and a new empirical correlation is fitted. To analyze the behavior of flow and heat transfer for low Prandtl number fluids, and also to validate the turbulence model and new correlation in the hexagonal rod bundles. The CFD simulations are performed for the XJTU 7-rod bundle, the KIT 19-rod bundle, and the ENEA 37-rod bundle based on ANSYS Fluent. The convective heat transfer coefficient of LBE, the temperature of heating wall at different sub-channels, and the temperature non-uniformity are investigated in this paper, to reveal the flow and heat transfer characteristics of LBE. The results show that the deviations of the section averaged Nusselt number are in 25%, and the deviations of the local temperature of LBE and heating wall are in 10℃. Grid spacers will result in a 30% rise in the section averaged Nusselt number, which will improve the turbulent intensity and heat transfer. Temperature non-uniformity is reduced as the Peclet number is increased. The validations demonstrate that the new empirical correlation provides accurate convective heat transfer coefficient predictions for LBE flow and heat transfer in hexagonal rod bundles with a relative derivation in 20%, which can be used as a reference for additional LBE simulations and experiments.

Experimental investigation on high-temperature flow boiling heat transfer characteristics of R245fa in a horizontal circular tube

Working fluid is the lifeblood of thermodynamic cycle for its heat transfer behaviour is critical to the design of heat exchangers in a thermodynamic cycle. Though being widely applied in organic Rankine cycles and high-temperature heat pump cycles, R245fa has witnessed little research on its high-temperature flow boiling heat transfer characteristics by far. In the present study, an experimental investigation on the flow boiling heat transfer characteristics of R245fa with a mass flux of 248–460 kg·m−2∙s−1, a heat flux of 5.5–17.3 kW·m−2 and an evaporating temperature of 80–115 °C in an inner diameter of 10 mm horizontal smooth circular tube is conducted. The influences of vapor quality, mass flux and evaporating temperature on the flow boiling heat transfer coefficient are also analyzed. The comparison of experimental heat transfer coefficients and the predicted heat transfer coefficients based on five heat transfer coefficient correlations is conducted and it is discovered that Gungor and Winterton's correlation with a mean deviation of 20.4% gives the best fit to the experimental data. Thus, based on Gungor and Winterton's work, an improved heat transfer coefficient correlation with a mean deviation of 12.8% is developed, in which 95% of the data points fall within the range of ±30%.

Improved correlation for predicting heat transfer coefficients during condensation inside smooth horizontal tubes

Abstract This study presents an improved correlation for predicting heat transfer coefficients for condensation inside smooth macro-tubes. Experimental data was obtained for various fluids, including R12, R1234yf, R1234ze, R134a, R22, R263fa, R245a, R32, R404a, R410a, and R600a. Tube diameters ranging from 3 to 14 mm, mass fluxes ranging from 50 to 1400 kg/m2s, and a vapor quality between 0 and 1 were used to cover a wide range of operating conditions. This data included 1683 data points from 50 sources. The comparison of the existing experimental data base with different correlations showed that the Cavallini et al. correlation had a mean absolute deviation of 19%, followed by those made by the correlations of Koyama and Yonemoto (24%), Dobson and Chato (24%), Huang et al. (24%), Shah (29%), and Haraguchi et al. (33%). Based on this comparison, an improved correlation modelled after the correlation of Cavallini et al. correlation was proposed. The accuracy and reliability of the proposed correlations were evaluated by comparing the predicted values with independent experimental data from the literature, wherein it was found that this new correlation predicted the databank with a mean absolute error of 17%. It was also found that this new proposed correlation provided better predictions of the heat transfer coefficient than existing correlations. The proposed correlation will improve the design and optimization of heat exchangers used in various industries, as well as provide valuable insights into condensation behavior inside smooth macro-tubes.

Prediction of effective heat transfer coefficients for vapour condensation inside horizontal tubes in stratified phase flow

In modern condensers of air conditioning systems, heat pumps, evaporators of seawater desalination systems, and heaters of power plants, the process of vapour condensation is carried out mainly inside the horizontal tubes and channels. Heat transfer processes occurring in condensers have a significant effect on the overall energy efficiency of the mentioned systems. In this paper, the experimental investigation of heat transfer during condensation of freons R22, R406a, and R407c in the plain smooth tube with d = 17 mm were carried out with the following parameters: ts = 35–40°C, G = 10–100 kg/(m2s), x = 0.8–0.1, q = 5–50 kW/m2, ΔT = 4–14 K. The unique measurements of circumferential heat fluxes and heat transfer coefficients were carried out with the thick wall method during different condensation modes. It can be inferred that with the increase of the heat flux, at the top part of the tube the thickness of the condensate film increases, which leads to the decrease in heat transfer. At the bottom of the tube, the increase in the heat flux enhances heat transfer coefficient, that is characteristic of the turbulent liquid flow in the tube. The obtained results allowed improving the prediction of effective heat transfer coefficients for vapour condensation, which takes into account the influence of condensate flow in the lower part of the tube on the heat transfer. This method generalises with sufficient accuracy (error ± 30%) the experimental data on condensation of freons R22, R134a, R123, R125, R32, R410a, propane, isobutene, propylene, dimethyl ether, carbon dioxide, and methane under stratified flow conditions. Using this method for designing heat exchangers, which utilise such types of fluids, will increase the efficiency of thermal energy systems.

Experimental Measurement and Theoretical Prediction of Bubble Growth and Convection Heat Transfer Coefficient in Direct Contact Heat Transfer

The measurement of the two-phase contact area is very important to determine the heat transfer coefficient in the process of direct contact heat transfer, but the direct measurement of the two-phase contact area is a difficult problem. The experiments are carried out utilizing a cylindrical Perspex tube of 100 cm in total height and 15 cm inner diameter. The active column height throughout the experiments is taken to be equal to 50 cm. Liquid Therminol®66 with four different initial temperatures (50 °C, 60 °C, 70 °C and 80 °C) is used as a continuous phase, while liquid R245fa at a constant temperature of 23 °C is used as a dispersed phase. In this paper, the empirical correlations between bubble growth and local convection heat transfer coefficient are obtained through modeling and measurement, and its correctness is verified by experiments. The results show that the bubble diameter is positively correlated with continuous phase temperature, flow rate ratio, and height, but the local convection heat transfer coefficient is negatively correlated with continuous phase temperature, flow rate ratio, and height. At the same time, it is found that the maximum error between the actual bubble diameter and the theoretical bubble diameter is 7%, and the error between the heat flux calculated by the local convection heat transfer coefficient and the actual heat flux is within 10%. This study provides theoretical guidance for an in-depth understanding of the direct contact heat transfer process and the development of high-efficiency waste heat recovery systems.

Zeotropic mixtures study in plate heat exchangers and ORC systems

Based on the known scientific literature, there are no correlations in the scientific literature to predict the two-phase heat transfer coefficients of Hydrofluoroethers pure fluids and zeotropic mixtures used in plate heat exchangers. This work has allowed the development of evaporation and condensation correlations for such fluids. Correlations based on dimensionless numbers and working fluids critical properties gave promising results, with a prediction of the two-phase heat transfer coefficient around 15% for heat source and 50% for cooling source which led respectively to a prediction of 4% and 6% for the global thermal power exchanged. The behavior of the organic Rankine cycle heat sources and the working fluids used within was also been studied. For this purpose, the influence of the heat sources flow rates variation on the performance of the hot and cold exchangers was analyzed. Results obtained showed a similar behavior and performance variation of the heat exchangers both when using pure fluids and zeotropic mixtures. Zeotropic mixtures working fluids in ORC system are often supposed in the literature to significantly increase the performance of this thermodynamic cycle due to better heat transfer at the hot source. Nevertheless, the great majority of these studies are numerical and do not take into account the interactions and system effects existing within the thermodynamic cycle when using a mixture instead of a pure fluid. This work has allowed to highlight this complexity and to provide a better understanding of the use of zeotropic mixtures.