Publication Date: 2023
Journal of Building Engineering (23527102)
A solar combisystem is utilized to supply domestic hot water (DHW) and heat a residential building. This system decreases the fossil fuel consumption and, hence, reduces air pollution and global warming. This paper proposes an innovative method to design and optimize a suitable solar combisystem for a residential building with respect to technical and economic parameters. First, different configurations of solar combisystems, including various components, are considered. Then, the optimum design variables are determined for each configuration using Grouped Method of Data Handling (GMDH) type of Artificial Neural Network (ANN) and Non-Dominated Sorting Genetic Algorithm II (NSGA-II). Total Solar Fraction (TSF) and Life Cycle Cost (LCC) are considered as objective functions. Finally, the optimum design is chosen for the solar combisystem. Typically, researchers have focused on evaluating the performance of a single solar combisystem configuration in each study. However, this research takes a different approach by optimizing multiple configurations, resulting in a significant improvement in the total solar fraction by 7.3%. This is the main novelty of this paper. Based on the results, the optimum configuration of the solar combisystem includes 17.91 m2 of evacuated-tube collectors with a tilt angle of 50°. Also, the volume of the hot water tank and heating buffer tank are, respectively, equal to 204 and 500 L (L). In this system, solar energy provides 94% of the required energy for supplying DHW and 23% of the energy for heating the building. Moreover, reduction of annual CO2 emissions is 1806 kg. This paper presents a guideline to design solar systems for the residential buildings considering technical and economic aspects. © 2023 Elsevier Ltd
Publication Date: 2022
Optik (00304026)
This article presents a modified version of the WENO numerical method with an increased order of accuracy over critical points and higher resolution in detecting shock-turbulence interactions. The proposed method is an improved version of the WENO-η-Z scheme. The optimization is based on a new Global Smoothness Indicator definition that produces less numerical error at relative extremum points as an indicator of fluctuations in the flow field. Both 1-D and 2-D benchmark problems are implemented to verify the proposed scheme's accuracy. The convergence of the presented scheme is compared with that of a standard and optimal WENO-η-Z, in the linear wave transfer problem, which shows better convergence for the proposed method. The modified method's capability to detect discontinuity and shocks in the flow-field has been evaluated by solving two shock-tube problems, namely the Lax shock tube problem and Sod's problem. The proposed method's ability to detect fluctuations and disturbances in the flow-field in the presence of shocks has also been assessed in two problems, including the 1-D Shu-Osher shock-disturbance interaction and the 2-D shock-turbulence interaction. Improvements is observed in convergence and reduction in numerical errors in the proposed method compared to the standard WENO and WENO-η-Z method, whilst the capability to detect shocks has not reduced in the modified version. © 2022 Elsevier GmbH
Publication Date: 2022
Energy and Buildings (03787788)
Determination of the optimum setpoint temperature of thermostats in various climates is a problem in air conditioning of residential buildings. In this paper, a new method is developed to optimize the setpoint temperature of thermostats in different climates of Iran. Design variables in the optimization process are heating setpoint, cooling setpoint, thickness, and thermal conductivity of insulations in the building envelopes. The optimization goals are minimizing energy consumption and cost of insulations in addition to maximizing thermal comfort of occupants. Thus, the static payback period (SPP) and the predicted percentage dissatisfied (PPD) indices are selected as objective functions which should be minimized in the optimization process. The methods applied to attain these objectives are numerical modeling by EnergyPlus software, Grouped Method of Data Handling (GMDH) type of Artificial Neural Network (ANN), and Non-Dominated Sorting Genetic Algorithm II (NSGA-II). Therefore in this process, first, EnergyPlus is used to train the neural network. Afterward, the GMDH-type neural network is applied to derive polynomials computing the objective functions from the design variables. Then, Pareto optimal points for the objective functions are obtained through using these polynomials and NSGA-II multi-objective optimization. Finally, the optimum design point is selected for different cities. According to the results, type and thickness of insulation integrated in the building envelopes affect the static payback period and thermal comfort of occupants. For all the climates of Iran, the most appropriate insulation is XPS and the optimum heating setpoint of thermostat is 22 °C. Also, the optimum value for the cooling setpoint pertains to the type of climate, so that this value for Bandar Abbas, Yazd, Tehran, Rasht and, Tabriz is, respectively, equal to 24.5, 24.7, 25.2, 25.3, and 25.6 °C. Moreover, thermal comfort of occupants increases with thickness of insulation, except for Bandar Abbas whose PPD is almost constant. The most value of PPD reduction with insulation thickness is related to Tehran where by increasing the insulation thickness from 1 cm to 5 cm, PPD decreases up to 53%. © 2022 Elsevier B.V.
Publication Date: 2020
International Journal of Computational Fluid Dynamics (10618562)(5)
Canard is one of the aerodynamic add-on devices which can reduce drag coefficient of the car. In this paper, different parameters of the canard geometry are determined using a multi-objective optimisation. Design variables are entrance velocity (U), geometrical parameters of canard (L 1, L 2, r, α) and canard angle from horizontal axis (θ). The objective functions include magnitude of drag and lift coefficients that should be minimised and maximised, respectively. First, the neural network is trained by means of a series of ANSYS Fluent-based CFD calculations. A GMDH-type neural network is then applied to derive polynomials that compute the objective functions from input variables. Finally, Pareto optimal points for objective functions are obtained through using these polynomials and NSGA-II multi-objective optimisation. According to the results, the canard’s optimum state is specified as L 1 = 0.37 m, L 2 = 0.18 m, r = 0.09 m, α = 25°, θ = 20° with potential drag reduction of 4.5%. © 2020, © 2020 Informa UK Limited, trading as Taylor & Francis Group.
Publication Date: 2018
International Journal of Engineering, Transactions A: Basics (17281431)(1)
In the present paper, a supersonic wind-tunnel is designed to maintain a flow with Mach number of 3 in a 30cm×30cm test section. An in-house CFD code is developed using the Roe scheme to simulate flow-field and detect location of normal shock in the supersonic wind-tunnel. In the Roe scheme, flow conditions at inner and outer sides of cell faces are determined using an upwind biased algorithm. The in-house CFD code has been parallelized using OpenMp to reduce the computational time. Also, an appropriate equation is derived to predict the optimum number of cores for running the program with different grid sizes. In the design process of the wind-tunnel, firstly geometry of the nozzle is specified by the method of characteristics. The flow in the nozzle and test section is simulated in the next step. Then, design parameters of the diffuser (convergence and divergence angles, area of the throat, and ratio of the exit area to the throat area) are determined by a trial and error method. Finally, an appropriate geometry is selected for the diffuser which satisfies all necessary criteria. © 2018 Materials and Energy Research Center. All rights reserved.
Publication Date: 2017
Neural Computing and Applications (09410643)
In the present study group method of data handling (GMDH) type of artificial neural networks are used to model deviation angle (θ), total pressure loss coefficient (ω), and performance loss coefficient (ξ) in wet steam turbines. These parameters are modeled with respect to four input variables, i.e., stagnation pressure (Pz), stagnation temperature (Tz), back pressure (Pb), and inflow angle (β). The required input and output data to train the neural networks has been taken from numerical simulations. An AUSM–Van Leer hybrid scheme is used to solve two-phase transonic steam flow numerically. Based on results of the paper, GMDH-type neural networks can successfully model and predict deviation angle, total pressure loss coefficient, and performance loss coefficient in wet steam turbines. Absolute fraction of variance (R2) and root-mean-squared error related to total pressure loss coefficient (ω) are equal to 0.992 and 0.002, respectively. Thus GMDH models have enough accuracy for turbomachinery applications. © 2016, The Natural Computing Applications Forum.
Publication Date: 2017
Scientia Iranica (10263098)(2)
In the present paper, an in-house CFD code is developed using Roe scheme to simulate a condensing two-phase flow in blade to blade passage of a steam turbine. Effects of condensation on the flow field of steam turbine rotor tip section are investigated for different outlet pressures. Firstly, comparison is performed between results of wet and dry cases. Then, effects of outlet pressure variations on the flow field are studied. Finally, effects of condensation on different specifications of the flow field (total pressure loss coefficient, entropy generation, and deviation angle) are investigated. Also, the mechanism of flow deviation in the cascade flow field is described. Condensation has a great influence on the behavior of the flow field based on the numerical results of this paper. It changes the out-flow direction, and consequently the flow entering to the next blade deviates from its on-design condition; thus, additional losses are produced. For example, the value of deviation angle reaches 7:62° for wet case and exit Mach number Me = 1:45. Also, there are stagnation pressure loss and entropy generation due to non-equilibrium condensation that reduce the efficiency of the steam turbine. © 2017 Sharif University of Technology.
Publication Date: 2016
Journal of Mechanical Science and Technology (1738494X)(3)
In this paper, the AUSM-van Leer hybrid scheme is extended to solve the governing equations of two-phase transonic flow in a steam turbine stage. The dominant solver of the computational domain is the non-diffusive AUSM scheme (1993), while a smooth transition from AUSM in regions with large gradients to the diffusive scheme by van Leer (1979) guarantees a robust hybrid scheme throughout the domain. The steam is assumed to obey non-equilibrium thermodynamic model. The effects of condensation on different specifications of the flow field are studied at subsonic/supersonic flow regimes. It is observed that as a result of condensation, the aerothermodymics of the flow field changes. For example in supersonic wet case (Pb = 14.55 kPa), pressure loss coefficient of rotor and total entropy generation are, respectively, 77% and 29% more than those in dry conditions. Also the value of rotor deviation angle reaches 6.27° in wet case and Pb = 14.55 kPa. © 2016, The Korean Society of Mechanical Engineers and Springer-Verlag Berlin Heidelberg.
Publication Date: 2016
Scientia Iranica (10263098)(6)
Petroleum Refinery Wastewaters (PRW) contain water-soluble hydrocarbons which cannot be separated by physical methods. In recent years, there have been enormous approaches to treat PRW. The most outstanding methods involve biological, photocatalytic, electro-and photo-Fenton, etc. Using microbial fuel cell is a new method to treat PRW. In this paper, PRW treatment in MFC was studied using oxygen and permanganate as cathodic electron acceptors. Also, effects of temperature and external resistance on MFC performance and PRW treatment were investigated. Finally, an electrochemical model was fitted on empirical polarization curves to evaluate activation, ohmic, and mass transfer losses. Maximum power production was 0.03 W/m2 at 33°C using oxygen as cathodic electron acceptor. Also, COD removal eficiency was 49.27% during 44 h. To enhance power production of the MFC, potassium permanganate was used as cathodic electron acceptor. At the temperature of 33°C and 0.2 g/L of permanganate concentration, the maximum power density was 0.95 W/m2 and COD removal eficiency was 78% during 44 h. © 2016 Sharif University of Technology. All rights reserved.
Publication Date: 2016
Journal Of Applied Fluid Mechanics (17353572)(2)
In the present paper, the hybrid AUSM-van Leer scheme is extended to solve the governing equations of twophase condensing flows. The method of moments with the classical homogeneous nucleation theory is used to model the non-equilibrium condensation phenomenon. Firstly, the hybrid method is validated using two test cases (i.e. Laval nozzle and rotor-tip cascade) and the results are compared with the MacCormack method. Then the hybrid method is used to solve two other problems (i.e. wavy channel and VKI stage). Based on the numerical results of the paper, the hybrid AUSM-van Leer scheme is an accurate method to simulate twophase transonic flows with nucleation. If the super cooling degree reaches to its maximum value, the nonequilibrium condensation begins and wetness fraction increases suddenly. Also across a shock the wetness fraction decreases due to evaporation of the droplets.
Publication Date: 2015
Applied Thermal Engineering (13594311)
In this paper, effects of turbine blade roughness and steam condensation on deviation angle and performance losses of the wet stages are investigated. The steam is assumed to obey non-equilibrium thermodynamic model, in which abrupt formation of liquid droplets produces condensation shocks. An AUSM-van Leer hybrid scheme is used to solve two-phase turbulent transonic steam flow around turbine rotor tip sections. The dominant solver of the computational domain is taken to be the AUSM scheme (1993) that in regions with large gradients smoothly switches to van Leer scheme (1979). This guarantees a robust hybrid scheme throughout the domain. It is observed that as a result of condensation, the aerothermodymics of the flow field changes. For example for a supersonic wet case with exit isentropic Mach number Me,is = 1.45, the deviation angle and total pressure loss coefficient change by 65% and 200%, respectively, when compared with dry case. It is also observed that losses due to surface roughness in subsonic regions are much larger than those in supersonic regions. Hence, as a practical guideline for maintenance sequences, cleaning of subsonic parts of steam turbines should be considered first. © 2015 Published by Elsevier Ltd.
Publication Date: 2012
Journal Of Applied Fluid Mechanics (17353572)(3)
In this paper vorticity confinement parameters are successfully developed for compressible flows. The first new confinement parameter is proportional to spectral radii of the flux Jacobian matrix. Therefore, the confinement parameter implicitly contains the local conditions of the flow field. This new method is named as lambda vorticity confinement method. In order to gain confidence in the applicability of vorticity confinement, it would be ideal to completely eliminate constant coefficients from confinement parameters. Because these constant coefficients should be determined for every problem by trial and error and it takes a long time. In the next part of this paper, a suitable relation is introduced for the vorticity confinement parameter that doesn't need any constant coefficient. This new method is named as adaptive vorticity confinement method. Then the capability of these new methods is compared with the other vorticity confinement methods for solving shock-vortex interaction and three dimensional moving vortex problems.
Publication Date: 2011
Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering (20413025)(8)
The conventional vorticity confinement methods have a constant confinement parameter that should be determined for every problem by trial and error. In this article, vorticity confinement parameters are successfully developed for compressible flows. The first new confinement parameter is proportional to spectral radii of the flux Jacobian matrix. Therefore, the confinement parameter implicitly contains the grid size and the other local fluid properties. In order to gain confidence in the applicability of vorticity confinement, it would be ideal to completely eliminate such constant parameters. In the next part of this article, a suitable relation is introduced for the vorticity confinement parameter that does not need any constant coefficient. The scalar dissipation scheme (SCDS-1) and matrix dissipation scheme (MADS-1) are two common artificial dissipation schemes that have been used for several years. Two new artificial dissipation schemes are introduced by using the QUICK scheme in this article (SCDS-2, MADS-2). The capabilities of these four artificial dissipation schemes are compared for channel flow problem. Then, the new confinement parameters and artificial dissipation schemes are used for solving moving vortex in a uniform flow and supersonic shear layer problems. The methods have been shown to be very effective at treating shock waves and vortex dominant flows. © Authors 2011.
Publication Date: 2011
International Review of Mechanical Engineering (19708734)(1)
The SCalar Dissipation Scheme (SCDS-1) and MAtrix Dissipation Scheme (MADS-1) are two common artificial dissipation schemes that have been used for several years. Two new artificial dissipation schemes are introduced by using the QUICK scheme in this paper (SCDS-2, MADS-2). The capability of these four artificial dissipation schemes is compared for two different problems. First for the channel flow problem and then for the moving vortex problem. The results of two problems show that the accuracy of these new artificial dissipation schemes (SCDS-2, MADS-2) are almost equal to two other schemes (SCDS-1, MADS-1). The implementation of the boundary conditions is more convenient in the new schemes. Also the new artificial dissipation schemes don't need any sensor. © 2011 Praise Worthy Prize S.r.l. - All rights reserved.
Publication Date: 2021
Communications in Algebra (00927872)49pp. 3837-3849
We study the class of virtually homo-uniserial modules and rings as a nontrivial generalization of homo-uniserial modules and rings. An R-module M is virtually homo-uniserial if, for any finitely generated submodules (Formula presented.) the factor modules (Formula presented.) and (Formula presented.) are virtually simple and isomorphic (an R-module M is virtually simple if, (Formula presented.) and (Formula presented.) for every nonzero submodule N of M). Also, an R-module M is called virtually homo-serial if it is a direct sum of virtually homo-uniserial modules. We obtain that every left R-module is virtually homo-serial if and only if R is an Artinian principal ideal ring. Also, it is shown that over a commutative ring R, every finitely generated R-module is virtually homo-serial if and only if R is a finite direct product of almost maximal uniserial rings and principal ideal domains with zero Jacobson radical. Finally, we obtain some structure theorems for commutative (Noetherian) rings whose every proper ideal is virtually (homo-)serial. © 2021 Taylor & Francis Group, LLC.
Publication Date: 2020
Journal of Algebra (00218693)549pp. 365-385
We study the class of virtually uniserial modules and rings as a nontrivial generalization of uniserial modules and rings. An R-module M is virtually uniserial if for every finitely generated submodule 0≠K⊆M, K/Rad(K) is virtually simple. Also, an R-module M is called virtually serial if it is a direct sum of virtually uniserial modules and a left virtually uniserial (resp., left virtually serial) ring is a ring which is virtually uniserial (resp., serial) as a left R-module. We give some useful properties of virtually (uni)serial modules and rings. In particular, it is shown that every left virtually uniserial module is uniform and Bézout. Also, we show that if R is a left virtually serial ring, then R/J(R)≅∏i=1 tMn(Di) where t,n1,…,nt∈N and each Di is a principal left ideal domain. As a consequence, we obtain that a ring R is left virtually serial with J(R)=0 if and only if R≅∏i=1 tMn(Di) where t,n1,…,nt∈N and each Di is a principal left ideal domain with J(Di)=0. Also, several classes of rings for which every virtually uniserial module (resp., ring) is uniserial are given. Noetherian left virtually uniserial rings are characterized. Finally, we obtain some structure theorems for (commutative) rings over which every (finitely generated) module is virtually serial. © 2020 Elsevier Inc.
Publication Date: 2025
International Journal of Energy Research (1099114X)2025(1)
This article introduces a novel optimization approach grounded in fuzzy logic, which transforms the multi-objective optimization problem into a single-objective one. Instead of providing a Pareto front, this method delivers a final optimal point based on predefined design priorities. The proposed methodology is applied to optimize illuminance, heating, and cooling setpoints in an office building across six cities with diverse climates to assess its performance under various conditions. The multi-objective optimization of these setpoints represents a novel contribution to smart building design. Compared to the NSGA-II method, the newly introduced approach exhibits simplicity and achieves a 50% reduction in computational time. The method leverages user experiences in formulating fuzzy rules, yielding more optimal solutions compared to the NSGA-II. The proposed method combines neural network, fuzzy logic, and genetic algorithm to create an efficient and intelligent framework for fast and accurate multi-objective optimization in energy-related problems. Copyright © 2025 Hamed Bagheri-Esfeh and Mohammad Reza Setayandeh. International Journal of Energy Research published by John Wiley & Sons Ltd.
Publication Date: 2021
Iranian Journal Of Fuzzy Systems (17350654)18(5)pp. 181-198
A novel strategy to design optimization is expressed using the fuzzy preference function concept. This method efficiently uses the designer’s experiences by preference functions and it is also able to transform a constrained multi-objective optimization problem into an unconstrained single-objective optimization problem. These two issues are the most important features of the proposed method which using them, you can achieve a more practical solution in less time. To implement the proposed method, two design optimizations of an unmanned aerial vehicle are considered which are: deterministic and non-deterministic optimizations. The optimization problem in this paper is a constrained multi-objective problem that with attention to the ability of genetic algorithm, this algorithm is selected as the optimizer. Uncertainties are considered and the Monte Carlo simulation (MCS) method is used for uncertainties modeling. The obtained results show a good performance of this technique in achieving optimal and robust solutions. © 2021, University of Sistan and Baluchestan. All rights reserved.
Publication Date: 2021
Journal of Aerospace Engineering (08931321)34(4)
Despite the many advantages of the design optimization technique, this method is costly for real engineering problems. This cost will increase sharply for issues with a multidisciplinary and uncertain nature and more than one objective function. In this paper, the metamodel concept has been used to overcome this problem. Because of the ability of neural networks to approximate the behavior of complex engineering systems, this tool has been used to create a surrogate model. Because multidisciplinary design optimization and robust design optimization methods have been used in this study and according to the high cost of the multidisciplinary analysis module, a surrogate model of this module has been made to reduce the imposed costs. To show the capability of the considered approach, robust multidisciplinary design optimization of an unmanned aerial vehicle (UAV) has been done. Take-off weight and cruise drag are the considered objective functions in this study, and the nondominated sorting genetic algorithm (NSGA-I) has been used for minimization of them. The optimization results show that the use of the metamodeling concept has reduced computational costs by 94.1%. © 2021 American Society of Civil Engineers.
Publication Date: 2020
Soft Computing (14327643)24(16)pp. 12429-12448
A new strategy for solving multidisciplinary design optimization problems is presented in this paper. The main idea of this approach is based on the use of designer experiences and attention to his/her preferences during design optimization which is implemented using a concept called the fuzzy preference function. Two important advantages of this approach are: (1) using the experiences of expert people during optimization and (2) transforming a constrained multiobjective design optimization problem into an unconstrained single-objective design optimization problem. The multidisciplinary design optimization of an unmanned aerial vehicle (UAV) is considered to show the performance of the proposed methodology. The optimization problem in this case study is a constrained two-objective problem (minimization of takeoff weight and drag of the cruise phase), and the genetic algorithm (GA) is utilized as the optimizer. Performance, weight, aerodynamics, center of gravity, trim and dynamic stability are the considered modules in the multidisciplinary analysis that are modeled using empirical and semiempirical equations. The optimization results show that the proposed strategy has been able to offer an optimal design where has higher performance relative to other methods from the point of view of objective functions, low computational cost and simplicity of implementation. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.
Publication Date: 2018
Chinese Journal of Aeronautics (10009361)31(12)pp. 2248-2259
This paper presents a Fuzzy Preference Function-based Robust Multidisciplinary Design Optimization (FPF-RMDO) methodology. This method is an effective approach to multidisciplinary systems, which can be used to designer experiences during the design optimization process by fuzzy preference functions. In this study, two optimizations are done for Predator MQ-1 Unmanned Aerial Vehicle (UAV): (A) deterministic optimization and (B) robust optimization. In both problems, minimization of takeoff weight and drag is considered as objective functions, which have been optimized using Non-dominated Sorting Genetic Algorithm (NSGA). In the robust design optimization, cruise altitude and velocity are considered as uncertainties that are modeled by the Monte Carlo Simulation (MCS) method. Aerodynamics, stability and control, mass properties, performance, and center of gravity are used for multidisciplinary analysis. Robust design optimization results show 46% and 42% robustness improvement for takeoff weight and cruise drag relative to optimal design respectively. © 2018 Chinese Society of Aeronautics and Astronautics
Publication Date: 2018
Journal of Aerospace Technology and Management (21759146)10
In this study, an efficient methodology is proposed for robust design optimization by using preference function and fuzzy logic concepts. In this method, the experience of experts is used as an important source of information during the design optimization process. The case study in this research is wing design optimization of Boeing 747. Optimization problem has two objective functions (wing weight and wing drag) so that they are transformed into new forms of objective functions based on fuzzy preference functions. Design constraints include transformation of fuel tank volume and lift coefficient into new constraints based on fuzzy preference function. The considered uncertainties are cruise velocity and altitude, which Monte Carlo simulation method is used for modeling them. The non-dominated sorting genetic algorithm is used as the optimization algorithm that can generate set of solutions as Pareto frontier. Ultimate distance concept is used for selecting the best solution among Pareto frontier. The results of the probabilistic analysis show that the obtained configuration is less sensitive to uncertainties. © 2018, Journal of Aerospace Technology and Management. All rights reserved.
Publication Date: 2021
International Journal of Engineering, Transactions B: Applications (1728144X)34(8)pp. 1974-1981
Neural networks are powerful tools for evaluating the thermophysical characteristics of nanofluids to reduce the cost and time of experiments. Dynamic viscosity is an important property in nanofluids that usually needs to be accurately computed in heat transfer and nanofluid flow problems. In this paper, the rheological properties of nanofluid phase change material containing mesoporous silica nanoparticles are predicted by the artificial neural networks (ANNs) method based on the experimental database reported in literature. Experimental inputs include nanoparticle mass fractions (0-5 wt.%), temperatures (35-55℃), shear rates (10-200 s-1), targets include dynamic viscosities and shear stresses. A multilayer perceptron feedforward neural network with Levenberg-Marquardt back-propagation training algorithm is utilized to predict rheological properties. The optimal network architecture consists of 22 neurons in the hidden layer based on the minimum mean square error (MSE). The results showed that the developed ANN has an MSE of 6.67×10-4 and 6.55×10-3 for the training and test dataset, respectively. The predicted dynamic viscosity and shear stress also have the maximum relative error of 6.26 and 0.418%, respectively. © 2021 Materials and Energy Research Center. All rights reserved.
Publication Date: 2021
International Journal of Energy Research (1099114X)45(10)pp. 15092-15109
The weak thermal conductance of a phase change material (PCM) can be intensified by dispersing nanostructured materials called nano-PCM. Accurate thermal conductivity (TC) prediction of nano-PCM is essential to evaluate heat transport during phase change processes, namely, melting and solidification. The present study develops an artificial neural network (ANN) to forecast the TC of n-octadecane as a PCM with dispersed oxide nanoparticles. A total of 122 experimental datasets from existing literature with a wide range of temperatures (5-60°C), nanoparticles (CuO, Al2O3, TiO2, and mesoporous SiO2), nanoparticle mass fractions (0.5-12 wt%) are compiled to train a multi-layered feed-forward ANN with Levenberg-Marquardt back-propagation algorithm. An optimal architecture of the neural network is acquired by varying the number of network hidden layers, the number of neurons in each layer, and the transfer function of layers. The minimum mean square error (MSE) of 1.3512 × 10−5 is obtained for the best developed ANN. Results show that average absolute deviation (AAD) of 0.002458, mean absolute percentage error (MAPE) of 0.8260%, and correlation coefficient (R) of 0.999964948 are achieved for training data. Moreover, MAPE, AAD, and R values are, respectively, 0.9478, 0.002167, and 0.9999715861 for testing data. The maximum percentage errors of ANN computed values are 2.31%, and 0.812% for liquid and solid phases, respectively. This indicates that the ANN model accurately predicts the enhanced TC of nano-PCM across various oxide nanoparticles, temperatures, and nanoparticle loadings. © 2021 John Wiley & Sons Ltd.
Publication Date: 2020
Journal of Energy Storage (2352152X)32
In this paper, a multi-objective optimization was performed to achieve the minimum cooling and heating loads in a residential building integrated with phase change material (PCM). The methods applied to fulfill this objective were numerical modeling by EnergyPlus, Grouped Method of Data Handling (GMDH) type of Artificial Neural Network (ANN) and Non-Dominated Sorting Genetic Algorithm II (NSGA-II). In this study, design variables were melting temperature and thickness of PCM, thermal resistance of exterior walls, internal gain and infiltration rate. Additionally, objective functions included annual cooling and heating loads of the building that should be minimized. Therefore, first, EnergyPlus software was used to calculate the values of objective functions and to train the neural network. Afterward, the GMDH-type neural network was applied to derive polynomials computing the objective functions from input variables. Then, Pareto optimal points for objective functions were obtained through using these polynomials and NSGA-II multi-objective optimization. Finally, the optimum point was determined by a financial analysis. According to the obtained results, the thickness and melting temperature of the optimum PCM layer in the residential building in Tehran were equal to 0.032 m and 24.58°C, respectively. © 2020 Elsevier Ltd
Publication Date: 2020
Journal of Energy Storage (2352152X)30
In this paper, heat transfer during the melting process of n-octadecane as a phase change material (PCM) is experimentally studied. This study is followed by an artificial neural network (ANN) to predict the melting characteristics of PCM. Experiments are performed in a rectangular enclosure subjected to a uniform heat flux in one vertical side. Melting heat transfer is characterized by observing the solid-liquid interface and recording the temperature distribution in the enclosure. Experimental results indicate that heat transfer during the melting process is dominated by natural convection. A multilayered perceptron feed-forward neural network trained by the Levenberg-Marquardt algorithm is used to predict the Nusselt number and the melted volume fraction. Rayleigh, Fourier and Stefan numbers are set as input parameters of the network. The optimal structure of the ANN to predict the Nusselt number show high accuracy in estimating the heat transfer characteristics during melting by achieving the mean square error and the correlation coefficient of 4.42 × 10−6 and 0.999, respectively. Based on the proposed ANN, the majority of the data falls within ±6.23% and ±6.54% of the Nusselt number and the melted volume fraction, respectively. © 2020 Elsevier Ltd
