DSpace Arşivi :: by Yazar "Çolak, Andaç Batur" değerine göre listeleniyor

Yazar "Çolak, Andaç Batur" seçeneğine göre listele

Listeleniyor 1 - 20 / 49

Adjoined ISPH method and artificial intelligence for thermal radiation on double diffusion inside a porous L-shaped cavity with fins
(Emerald Publishing, 2024) Elshehabey, Hillal M.; Çolak, Andaç Batur; Aly, Abdelraheem
Purpose: The purpose of this study is to adapt the incompressible smoothed particle hydrodynamics (ISPH) method with artificial intelligence to manage the physical problem of double diffusion inside a porous L-shaped cavity including two fins. Design/methodology/approach: The ISPH method solves the nondimensional governing equations of a physical model. The ISPH simulations are attained at different Frank–Kamenetskii number, Darcy number, coupled Soret/Dufour numbers, coupled Cattaneo–Christov heat/mass fluxes, thermal radiation parameter and nanoparticle parameter. An artificial neural network (ANN) is developed using a total of 243 data sets. The data set is optimized as 171 of the data sets were used for training the model, 36 for validation and 36 for the testing phase. The network model was trained using the Levenberg–Marquardt training algorithm. Findings: The resulting simulations show how thermal radiation declines the temperature distribution and changes the contour of a heat capacity ratio. The temperature distribution is improved, and the velocity field is decreased by 36.77% when the coupled heat Cattaneo–Christov heat/mass fluxes are increased from 0 to 0.8. The temperature distribution is supported, and the concentration distribution is declined by an increase in Soret–Dufour numbers. A rise in Soret–Dufour numbers corresponds to a decreasing velocity field. The Frank–Kamenetskii number is useful for enhancing the velocity field and temperature distribution. A reduction in Darcy number causes a high porous struggle, which reduces nanofluid velocity and improves temperature and concentration distribution. An increase in nanoparticle concentration causes a high fluid suspension viscosity, which reduces the suspension’s velocity. With the help of the ANN, the obtained model accurately predicts the values of the Nusselt and Sherwood numbers. Originality/value: A novel integration between the ISPH method and the ANN is adapted to handle the heat and mass transfer within a new L-shaped geometry with fins in the presence of several physical effects
Analyzing activation energy and binary chemical reaction effects with artificial intelligence approach in axisymmetric flow of third grade nanofluid subject to soret and dufour effects
(Begell House Inc, 2023) Shafiq, Anum; Çolak, Andaç Batur; Sindhu, Tabassum Naz
The use of nanotechnology has led to the design of many modern and more cost-effective implementation, such as solar power generation, the redevelopment of heat exchangers, and the modernization of the medical and pharmaceutical industries. In this study, the combined effects of activation energy with binary chemical reactant in a steady magnetohydrodynamic mixed convective third-grade nanofluid flow by radially radiative stretching plate has been analyzed with an artificial intelligence approach. Heat transfer analysis was conducted with heat generation, Joule heating, and Soret and Dufour effects. By incorporating appropriate transformations, the initial nonlinear coupled partial differential equations expressing the fluid model were formed as a comparable nonlinear ordinary differential equations system. Three different artificial neural network models were proposed in order to predict the skin friction, Nusselt number, and Sherwood number values of the fluid model by the Shooting Runge-Kutta Fehlberg 4, technique using the data set created by taking various values of the relevant parameters. It is worthy of noting that the average deviation values for each output parameter remained less than 5%. Furthermore it is also observed that mean square error values for skin friction coefficient, local Nusselt number, and local Sherwood number values were attained as 3.63 x 10(-3), 4.03 x 10(-4), and 8.62 x 10(-3), respectively. The obtained results show that artificial neural networks are an engineering tool that can be used with high accuracy to estimate the combined effects of activation energy and binary chemical reaction in a fixed magnetohydrodynamic mixed convective third-grade nanofluid flow with a radial radiative stretched plate.
Application of artificial intelligence techniques for heat exchanger predictions in food industry
(Elsevier, 2024) Öztuna Taner, Öznur; Mercan, Hatice; Çolak, Andaç Batur; Radulovic, Jovana; Taner, Tolga; Dalkılıç, Ahmet Selim
Heat exchangers (HEXs) are deployed in diverse engineering applications, such as cooling and refrigeration systems; power plants; and automotive, chemical, textile, and food industries. Understanding the principles and fluid-to-fluid heat exchange geometry can be complex. Researchers usually apply the first and second laws of thermodynamics to conduct numerical, analytical, and experimental techniques on HEXs. Experimental approaches tend to be costlier due to setup expenses, while theoretical and numerical analyses rely heavily on assumptions and complex equations. To address these challenges, artificial intelligence (AI) models have emerged as a promising solution for modeling, optimization, and performance estimation of thermal systems employing HEXs. In the last 30 years, AI-based approaches have gained widespread adoption in thermal analysis of HEXs, building upon past research. Three main types of thermal analysis have been reported: single-phase flow, two-phase flow, and machine learning-based physical property evaluation. AI approaches have proven effective in estimating crucial HEX parameters like pressure drop (?P), heat transfer coefficient (h), friction factor (f), and Nusselt number (Nu). They have also demonstrated success in assessing phase change characteristics during fluid boiling and condensation processes, as well as identifying two-phase flows. Despite these advancements, it is emphasized that more work remains to fully harness AI’s potential for thermal analysis of HEXs. As AI gains traction, it presents itself as a valuable technology for enhancing the study of HEXs with satisfactory results.
Artificial neural network-based cooling capacity estimation of various radiator configurations for power transformers operated in ONAN mode
(Elsevier, 2024) Koca, Aliihsan; Şenturk, Oğuzkan; Çolak, Andaç Batur; Bacak, Aykut; Dalkılıç, Ahmet Selim
Power transformers submerged in oil are universally acknowledged as very useful elements in electrical power networks. A fraction of the electrical energy involved in the conversion from high to low voltages is lost as thermal energy, which is produced inside the transformer's windings and core. The effective dissipation of heat is of paramount significance and can be accomplished through the strategic installation of radiators on the tank. This study aims to examine the total cooling capacity of a radiator employed for cooling a power transformer operating in oil, natural air, and natural mode. The investigation is conducted by varying the design parameters, specifically the number of fins per radiator and the radiator length. The study also utilizes computational fluid dynamics results to achieve a substantial convergence with experimental findings. In the context of the verification study conducted for a specific study point, it was found that the simulation outcomes at a mass flow rate of 0.15 kg/s corresponded to about 7.4 % of the experimental results. Additionally, the cooling capacity value obtained was 7.2 %. In the context of machine learning, the performance of the numerical approach was assessed by employing the Bayesian regularization method. The evaluation revealed that the margin of deviation, mean squared error, and coefficient of determination (R2) metrics yielded values of ?0.001 %, 1.32E-02, and 0.99930, respectively.
An artificial neural network-based numerical estimation of the boiling pressure drop of different refrigerants flowing in smooth and micro-fin tubes
(De Gruyter, 2024) Çolak, Andaç Batur; Bacak, Aykut; Kayacı, Nurullah; Dalkılıç, Ahmet Selim
In thermal engineering implementations, heat exchangers need to have improved thermal capabilities and be smaller to save energy. Surface adjustments on tube heat exchanger walls may improve heat transfer using new manufacturing technologies. Since quantifying enhanced tube features is quite difficult due to the intricacy of fluid flow and heat transfer processes, numerical methods are preferred to create efficient heat exchangers. Recently, machine learning algorithms have been able to analyze flow and heat transfer in improved tubes. Machine learning methods may increase heat exchanger efficiency estimates using data. In this study, the boiling pressure drop of different refrigerants in smooth and micro-fin tubes is predicted using an artificial neural network-based machine learning approach. Two different numerical models are built based on the operating conditions, geometric specifications, and dimensionless numbers employed in the two-phase flows. A dataset including 812 data points representing the flow of R12, R125, R134a, R22, R32, R32/R134a, R407c, and R410a through smooth and micro-fin pipes is used to evaluate feed-forward and backward propagation multi-layer perceptron networks. The findings demonstrate that the neural networks have an average error margin of 10?percent when predicting the pressure drop of the refrigerant flow in both smooth and micro-fin tubes. The calculated R-values for the artificial neural network’s supplementary performance factors are found above 0.99 for all models. According to the results, margins of deviations of 0.3?percent and 0.05?percent are obtained for the tested tubes in Model 1, while deviations of 0.79?percent and 0.32?percent are found for them in Model 2.
Artificial neural networking estimation of skin friction coefficient at cylindrical surface: a Casson flow field
(Springer, 2023) Rehman, Khalil Ur; Shatanawi, Wasfi; Çolak, Andaç Batur
In this article, we constructed Artificial Neural Networking (ANN) models to predict values of the skin friction coefficient for two different flow regimes of non-Newtonian fluid. More specifically, flow of Casson fluid is considered toward an inclined surface with stagnation point and mixed convection effects. Energy equation is considered by means of thermal radiations, viscous dissipation, heat generation and temperature-dependent variable viscosity effects. The flow regime is carried as a two various models namely Model-I: Casson fluid flow in the presence of magnetic field and Model-II: Casson fluid flow in the absence of magnetic field. Mathematical formulation is presented for each model, and shooting method is used to obtain the numerical data of skin friction coefficient. In contrast to the Casson fluid, mixed convection, and velocities ratio parameters, the skin friction coefficient exhibits a direct relationship with the magnetic field parameter and the curvature parameter. The MoD values for both models (I, II) show that there is relatively little variation between targeted and the projected values produced from the constructed ANN models.
Artificial Neural Networking Magnification for Heat Transfer Coefficient in Convective Non-Newtonian Fluid with Thermal Radiations and Heat Generation Effects
(MDPI, 2023) Rehman, Khalil Ur; Shatanawi, Wasfi; Çolak, Andaç Batur
In this study, the Casson fluid flow through an inclined, stretching cylindrical surface is considered. The flow field is manifested with pertinent physical effects, namely heat generation, viscous dissipation, thermal radiations, stagnation point flow, variable thermal conductivity, a magnetic field, and mixed convection. In addition, the flow field is formulated mathematically. The shooting scheme is used to obtain the numerical data of the heat transfer coefficient at the cylindrical surface. Further, for comparative analysis, three different thermal flow regimes are considered. In order to obtain a better estimation of the heat transfer coefficient, three corresponding artificial neural networks (ANN) models were constructed by utilizing Tan-Sig and Purelin transfer functions. It was observed that the heat transfer rate exhibits an inciting nature for the Eckert and Prandtl numbers, curvature, and heat generation parameters, while the Casson fluid parameter, temperature-dependent thermal conductivity, and radiation parameter behave oppositely. The present ANN estimation will be helpful for studies related to thermal energy storage that have Nusselt number involvements.
Assessment of heat transfer characteristics of a corrugated heat exchanger based on various corrugation parameters using artificial neural network approach
(Elsevier B.V., 2024) Çolak, Andaç Batur; Kırkar, Şafak Metin; Gönül, Alişan; Dalkılıç, Ahmet Selim
The complexity of the fluid flow process involved makes it difficult to estimate the characteristics of corrugated tubes. Heat exchangers are often designed to be more efficient by using numerical techniques. Recently, machine learning algorithms have become a viable method for assessing the behaviors of flow and heat transfer in corrugated tubes. Based on a set of data, machine learning algorithms can better estimate how efficient a heat exchanger is. In the present study, an artificial neural network is conducted to figure out the Nusselt number, friction factor, and performance evaluation criteria for heat transfer in straight corrugated tubes based on flow rate and corrugation parameters. The Reynolds number varies between 480 and 6100, spanning various flow regimes in the corrugated tubes, while the corrugation pitch and corrugation depth change between 6 mm and 18 mm and 0.6 and 1.0 mm, respectively. After totaling 220 data points, the network structure with a multilayer perceptron structure is trained. The Levenberg-Marquardt algorithm is performed for training with 17 neurons in the hidden layer. The established neural network structure forecasts Nusselt number, friction factor, and performance evaluation criteria parameters with deviation rates of 0.11 %, ?0.63 %, and 0.17 %, respectively. The neural network exhibits higher performance when compared to related correlations from the literature. This study is a novel one in open sources due to using artificial neural networks to estimate the flow and thermal behaviors in corrugated tubes operating at low flow rates. The current recommended approach may be regarded as a beneficial tool particularly for thermal systems as it aids designers in enhancing the system efficiency with accurate estimations.
Carreau Akışkanının Dikey Germe Silindirindeki Akış Karakteristiklerinin Yapay Zeka Yaklaşımıyla Analizi
(2023) Çolak, Andaç Batur
Bu çalışmada, Carreau akışkan akışının gözenekli bir ortama daldırılmış dikey bir germe silindiri üzerindeki akış karakteristikleri, yapay zeka yaklaşımı ile detaylı olarak analiz edilmiştir. Akış parametreleri olarak lokal yüzey sürtünmesi, lokal Nusselt sayısı ve lokal Sherwood sayısı parametreleri ele alınmıştır. Akış parametrelerini tahmin etmek için çok katmanlı algılayıcı mimarisine sahip üç farklı yapay sinir ağı modeli tasarlanmıştır. Literatürden elde edilmiş nümerik veri seti kullanılarak eğitilmiş olan ağ modellerinde Bayesian Düzenlileştirme eğitim algoritması kullanılmıştır. Farklı performans parametreleri dikkate alınarak optimize edilen yapay sinir ağlarında tahmin performansı en yüksek olan modeller tercih edilmiştir. Elde edilen tahmini değerler, hedef verilerle karşılaştırılmıştır. Ayrıca performans parametreleri de hesaplanmış ve ağ modellerinin performansları kapsamlı bir şekilde analiz edilmiştir. Çalışma bulguları, geliştirilmiş olan yapay sinir ağlarının, doğal taşınımlı Carreau akışına ait parametreleri yüksek doğrulukta tahmin edebildiğini ortaya koymuştur.
CFD and ANN analyses for the evaluation of the heat transfer characteristics of a rectangular microchannel heat sink with various cylindrical pin-fins
(Springer, 2024) Malazi, Mahdi Tabatabaei; Kaya, Kenan; Çolak, Andaç Batur; Dalkılıç, Ahmet Selim
Electrical equipment extensively uses Microchannels (MCs) for cooling. Due to their complexity, it is challenging to evaluate the features of the fluid flow and heat transfer processes in MC pin-fin heat sinks. Numerical approaches have been frequently employed in MC design to enhance efficiency. Machine learning methods have recently enabled the assessment of flow and heat transfer research in these devices. In this study, numerical calculations have been made to obtain outlet fluid temperature, the average Nusselt number, and pressure drop, using the computational fluid dynamics (CFD) software, ANSYS Fluent. Previous experimental work validates the numerical model by examining the average Nusselt number and the apparent friction factor. Three distinct ratios of fin spacing to fin diameter (l/d = 2, 4, and 6) and five different values of Reynolds number (Re = 50, 75, 100, 125, and 150) are considered. A constant ratio of fin height to channel height (h/H = 0.25) is maintained, and the inlet fluid temperature is set to 291.15, 294.15, 297.15, and 300.15 K. Numerical calculations have been conducted for cases of uniform and non-uniform heating, where bottom wall temperatures of 323.15 K and 317.15 K were considered, respectively, for a fixed fin surface temperature of 323.15 K. Using the results of the numerical simulations, a multi-layer perceptron (MLP)-structured artificial neural network (ANN) is trained. The Levenberg-Marquardt (LM) training method is employed in the hidden layer, using 17 neurons for the training procedure. The results of the numerical simulations show that the average Nusselt number increases linearly with the Reynolds number, except for the non-uniform heating case of Re = 50. The average Nusselt number and pressure drop are inversely proportional to fin spacing for all cases. There is also a linear increase in pressure drop with the Reynolds number, since the flow regime considered in this study is laminar. The ANN model predicts the outlet fluid temperature, the average Nusselt number, and the pressure drop, with variation rates of -0.0027%, -0.075%, and - 0.0004%, respectively.
A comparative analysis of maximum likelihood estimation and artificial neural network modeling to assess electrical component reliability
(John Wiley and Sons, 2022) Çolak, Andaç Batur; Sindhu, Tabassum Naz; Lone, Showkat Ahmad; Akhtar, Md Tanwir; Shafiq, Anum
This study focuses on accurately predicting the behavior of new power functiondistribution using neural network and optimizing it using maximum likelihoodestimation. The main motivation of this study is that there is no study inthe literature that optimizes and predicts the reliability analysis of lifetimemodels by combining artificial neural networks and maximum likelihoodestimation methods. The numerical findings of the reliability investigationsand the values got from maximum likelihood estimation and artificial neuralnetwork modeling have been examined and investigated carefully. For theartificial neural network models, the R value was 0.99999 and the deviationratios were lower than 0.08%. The findings reveal that artificial neural networksare a powerful and useful mathematical tool for analyzing the reliabilityof lifetime models and numerical study findings via maximum likelihoodestimation are completely in accord with artificial neural network predictionresults.
Comparative analysis to study the Darcy–Forchheimer Tangent hyperbolic flow towards cylindrical surface using artificial neural network: An application to Parabolic Trough Solar Collector
(Elsevier, 2024) Shafiq, Anum; Çolak, Andaç Batur; Sindhu, Tabassum Naz
Solar thermal collectors convert sunlight into useful thermal energy by absorbing its incoming radiation. Concentrated solar power technologies use the parabolic trough solar collector to collect solar energy with temperatures ranging from 325– 700 K. The tangent hyperbolic fluid model is one of the most important non-Newtonian fluid models. Laboratory studies demonstrate that this model accurately predicts the shear thinning phenomenon. In addition, tangent hyperbolic fluid has a better heat transfer performance due to its rheological bearing at various shear rates. The current study investigates the heat transmission performance of Darcy–Forchheimer tangent hyperbolic radiative inclined cylindrical film movement in parabolic trough solar collector with an irregular heat sink/source utilizing the Levenberg–Marquardt technique and backpropagated neural networks. Through the implementation of required transformations, this system is turned into an equivalent nonlinear ordinary differential system. The findings are investigated for Newtonian and tangent hyperbolic fluid cases to understand the rheological characteristics. The outcomes are considered using graphical and mathematical evaluations. Fluids featuring tangent hyperbolic rheological conductivity are obligatory for active rate of heat diffusion. As a consequence, these fluids may be employed in Parabolic Trough Solar Collector for increased heat transmission rate and operational usage of solar energy. Furthermore, We create a dataset using the Runge–Kutta fourth-order shooting technique to create the proposed multilayer perceptron artificial neural network. The data points representing the MoD values are observed to be closely clustered around the zero deviation line. Additionally, it is important to highlight that these data points have relatively small numerical values. Moreover, when calculating the average MoD values for each output, it becomes evident that they are consistently very low.
Comparison of experimental thermal conductivity of water-based Al2O3–Cu hybrid nanofluid with theoretical models and artificial neural network output
(Springer Science and Business Media B.V., 2024) Çolak, Andaç Batur; Bayrak, Mustafa
The research aimed to experimentally test the thermal conductivity of five distinct Al2O3–Cu/water hybrid nanofluids. These nanofluids were generated at volumetric concentrations of 0.0125, 0.025, 0.05, 0.1, and 0.2. The measurements were conducted within a temperature range of 10–65 °C. The primary objective of this research is to tackle the insufficient empirical data on hybrid nanofluids and establish a dependable artificial neural network model for forecasting their thermal conductivity. A multilayer perceptron feed forward back propagation artificial neural network has been created using the acquired experimental thermal conductivity data. The experimental thermal conductivity data have been compared with four commonly used mathematical correlations and the outputs of an artificial neural network. The findings demonstrated that the constructed artificial neural network accurately forecasted the thermal conductivity of the Al2O3–Cu/water hybrid nanofluid, with an average deviation of just 0.4%. Nevertheless, Maxwell’s mathematical correlation proved to be the most accurate model in predicting the experimental findings, with an average error margin of just 0.08%. © Akadémiai Kiadó, Budapest, Hungary 2024.
Computational Analysis on Magnetized and Non-Magnetized Boundary Layer Flow of Casson Fluid Past a Cylindrical Surface by Using Artificial Neural Networking
(MDPI, 2023) Rehman, Khalil Ur; Shatanawi, Wasfi; Çolak, Andaç Batur
In this article, we constructed an artificial neural networking model for the stagnation point flow of Casson fluid towards an inclined stretching cylindrical surface. The Levenberg–Marquardt training technique is used in multilayer perceptron network models. Tan–Sig and purelin transfer functions are carried in the layers. For better novelty, heat and mass transfer aspects are taken into account. The viscous dissipation, thermal radiations, variable thermal conductivity, and heat generation effects are considered by way of an energy equation while the chemical reaction effect is calculated by use of the concentration equation. The flow is mathematically modelled for magnetic and non-magnetic flow fields. The flow equations are solved by the shooting method and the outcomes are concluded by means of line graphs and tables. The skin friction coefficient is evaluated at the cylindrical surface for two different flow regimes and the corresponding artificial neural networking estimations are presented. The coefficient of determination values’ proximity to one and the low mean squared error values demonstrate that each artificial neural networking model predicts the skin friction coefficient with high accuracy
Construction of neural network based intelligent computing for treatment of darcy–forchheimer sisko nanofluid flow with rosseland’s radiative process
(Begell House Inc., 2023) Shafiq, Anum; Çolak, Andaç Batur; Sindhu, Tabassum Naz
A generalization of Newtonian and power-law fluids is the Sisko model. It foretells dilatants and fluid pseudoplasticity. It was first suggested to use the Sisko fluid model to gauge high shear rates in lubricating greases. Three constants in this model are easily selectable for certain fluids, and it is demonstrated that the model is a good predictor of shear thickening and thinning. The study of nanofluids is gaining popularity quickly because of unique thermal, mechanical, and chemical characteristics of nanomaterials. Sisko nanofluids are also required for the production of nanoscale materials because of the superb wetting and dispersing capabilities they possess. In the present investigation, the Levenberg-Marquardt method with backpropagated neural networks is used to evaluate the nanomaterial flow of Darcy–Forchheimer Sisko fluid model. Thermophoresis and Brownian motion effects are considered when developing the nanofluid model. By applying the necessary transformations, the original nonlinear coupled partial differential system representing fluidic model are converted to an analogous nonlinear ordinary differential system. For different fluid model scenarios, a dataset for the proposed multilayer perceptron artificial neural network is produced by altering the necessary variables via the Galerkin weighted residual approach. An artificial neural network called a multilayer perceptron has been created in order to forecast the multilayer perceptron values.
Dairy factory milk product processing and sustainable of the shelf-life extension with artificial intelligence: a model study
(Frontiers Media SA, 2024) Öztuna Taner, Öznur; Çolak, Andaç Batur
This study models milk product processing and sustainable of the shelf-life extension in a dairy factory using artificial intelligence. The Cappadocia dairy factory was used to study chemical processes and computational system modeling and simulation. Levenberg–Marquardt algorithm was used to create an artificial neural network model from real-time data. An AI-based method utilizing a Multilayer Perceptron (MLP) Artificial Neural Network (ANN) model was employed to precisely analyze productivity data in dairy factories. There are 9 product types and production quantities used as input parameters, and 90 datasets of actual dairy products used as output values. The model was trained using the Levenberg–Marquardt algorithm on 62 datasets for training, 14 for validation, and 14 for testing. The accuracy of the model is affected by the optimal data segmentation. The model showed how AI algorithms can improve processes and industrial production by increasing dairy production efficiency from 20 to 40%. Model efficiency values were compared to observed values to determine prediction accuracy. Model mean squared error was 4.02E-06, and coefficient of determination was 0.99984. Model efficiency predictions and observed values differed by ?0.04% on average. This study investigated using artificial intelligence to optimize salvage processes and systems to increase energy efficiency and reduce environmental impact. The results show that a neural network model trained with real data can predict dairy plant productivity.
A decreasing failure rate model with a novel approach to enhance the artificial neural network's structure for engineering and disease data analysis
(Elsevier, 2024) Sindhu, Tabassum Naz; Çolak, Andaç Batur; Lone, Showkat Ahmad; Shafiq, Anum; Abushal, Tahani A.
The study focuses on key metrics used to examine the characteristics of a lifetime random variable distribution in reliability and survival theory research. In this analysis, metrics including the probability density function time, mean residual lifespan, mean time between failures, hazard rate, and reliability function are essential. The focus of the inquiry is these important parameters in relation to the Burr-Hatke exponential model specifically. The study focuses on key metrics used to examine the characteristics of a lifetime random variable distribution in reliability and survival theory research. In this analysis, metrics including the probability density function time, mean residual lifespan, mean time between failures, hazard rate, and reliability function are essential The focus of the inquiry is these important parameters in relation to the Burr-Hatke exponential model specifically. A key component of the research is a comparison of the outcomes from the artificial intelligence approach and those from conventional literature-based methodologies. This comparison study sheds light on how well the artificial neural network framework performs while evaluating the Burr-Hatke exponential model’s technical features. The study allows a comprehensive analysis of the training and prediction capabilities of the growing neural network by calculating multiple performance measures. This comprehensive strategy improves our comprehension of the model’s survival traits and reliability, offering significant contributions to the larger field of study. The network structure’s mean square error was estimated to be 5.19E-04, and its coefficient of determination value was 0.99987 for the first neural network model. For the second neural network model, the coefficient of determi nation value was 0.99999 and the mean square error value was 4.58E-06. The outcomes amply revealed the neural network structure’s extraordinarily high prediction accuracy and the degree to which the prediction outputs agree with those of the Maximum Likelihood Estimation technique.
Determination of optimum insulation thickness in submarines
(Yildiz Technical University, 2023) Durmaz, Savaş; Çolak, Andaç Batur; Mercan, Hatice; Dalkiliç, Ahmet Selim
One of the most effective ways to save energy for cooling and heating applications is thermal insulation. Because of this, determining the ideal insulation thickness is a popular topic for publications. The purpose of this study is to determine the appropriate insulation thickness needed for a submarine’s external construction while it is cruising in various locations. Since seawater makes up a submarine’s external environment, situations involving five distinct sea-water temperatures from around the globe have been studied. There are five of them: the Med-iterranean, Marmara, Aegean, Black Sea, and Sakhalin, which is in the North Pacific Ocean and has the coldest seawater on earth. By using the idea of degree-days, the annual cooling and heating needs of submarines in various regions have been computed. Based on life cycle cost analysis, optimization has been accomplished. In the beginning, the results of a study published in the literature supported the calculation methods utilized. The use of insulation materials such as rock wool, glass wool, polyurethane, expanded polystyrene, fiberglass, and foam glass, as well as fuel oil to run the generator, has been taken into account in a number of calculations, including the best insulation thickness, annual savings value, annual energy cost, and payback period. The findings indicate that depending on seawater temperatures and insulation materials, the ideal insulation thicknesses range between 2 and 12 cm, energy savings between 8.5% and 90%, and payback periods between 1.1 and 10 years.
Development of an intelligent computing system using neural networks for modeling bioconvection flow of second-grade nanofluid with gyrotactic microorganisms
(Taylor and Francis Ltd., 2023) Shafiq, Anum; Çolak, Andaç Batur; Sindhu, Tabassum Naz
Nanoparticles are carried in bioconvective fluid flow by convective motion caused by living tissues. This flow has important applications in cell and tissue engineering because it demonstrates the mechanics of particle trans fer between cells and fluids. This type of flow is used in medicine delivery systems that particularly target cancer cells in real life. Nanofluids are cru cial suspensions that allow nanomaterials to disperse and behave in a homogeneous and stable environment. The bioconvective second-grade nanofluid flow, on the other hand, is distinguished by a more complex process that permits nanoparticle motion to be controlled by external fields and pressures. This type of flow has numerous applications, including biology, the environment, and energy. It is particularly useful in medical imaging, cancer hyperthermia treatment, and nanodrug delivery systems. The primary purpose of this research is to use an artificial neural network to examine the rate of heat, mass, and motile microbe movement in the convective flow of magnetohydrodynamic second-grade nanofluid toward vertical surface. Suspended nanoparticles are effectively stabilized by the action of microorganisms, facilitated through bioconvection. This process is influenced by both nanoparticle attributes and buoyancy forces. In add ition to thermophoretic dynamics and Brownian motion, the model consid ers radiation and Newtonian heating effects. Nonlinear equation systems are obtained using appropriate transformations. The non-linear simplified equations underwent numerical calculations utilizing the fourth-order Runge-Kutta shooting method. The Sherwood number, Nusselt number, and density of motile microorganism coefficient were determined using various parameters, and three distinct artificial neural networks were built employing the findings.
Discharging performance prediction of experimentally tested sorption heat storage materials with machine learning method
(Elsevier, 2022) Çolak, Andaç Batur; Aydin, Devrim; Al-Ghosini, Abdullah; Dalkilic, Ahmet Selim
In this study, the usability of the machine learning method in predicting the discharge performance of experi- mentally tested sorption heat storage materials was investigated. Experimental data was obtained from a lab scale fixed-bed thermochemical heat storage unit. 9 candidate composites were tested under different inlet conditions. Based on the experimental data, moisture sorption rates, heat output, exergy output and energy storage densities were determined. For the 6 cycles testing, highest average heat and exergy output were ob- tained with vermiculite/LiCl composite with the values of 0.83 kW and 0.013 kW, respectively. On the other hand, P-CaCl2 was found as the most durable material in terms of energy storage density (296 ? 209 kWh/m3). A multilayer perceptron artificial neural network was established to evaluate measured data and its prediction performance was extensively studied. In the model 54 experimental data sets were utilized, consisting of 6 cycles testing of 9 different composite sorbents. Levenberg-Marquardt algorithm was benefited as the training one in the artificial neural network model established and the Tan-Sig and Purelin functions were selected as the transfer one in the multilayer neural network with 7 neurons in the hidden layer. According to the mathematical defi- nition of the discussed statistical metrics, experimental data were used to compare them to the predicted output in order to verify the reliability of the proposed ANN model; and the analysis of the model was performed by examining the coefficient of determination, mean squared error, and deviation values, which were assumed as performance parameters, in detail. The deviation rate between the prediction values acquired from the artificial neural network and the practical data was determined as less than ±5 %. The acquired findings showed that artificial neural networks, which is one of the common machine learning algorithms, is a preferable method that can be employed to estimate the discharge performance of sorption heat storage materials.

Yazar "Çolak, Andaç Batur" seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri