Search results for: single layer model

Authors: Al-Kashif M., Abdel-Mooty M., Fahmy E., Abou Zeid M., Haroun M.

Abstract:

Sandwich panels are widely used in the construction industry for their ease of assembly, light weight and efficient thermal performance. They are composed of two RC thin outer layers separated by an insulating inner layer. In this research the inner insulating layer is made of lightweight Autoclaved Aerated Concrete (AAC) blocks which has good thermal insulation properties and yet possess reasonable mechanical strength. The shear strength of the AAC infill is relied upon to replace the traditionally used insulating foam and to provide the shear capacity of the panel. A comprehensive experimental program was conducted on full scale sandwich panels subjected to bending. In this paper, detailed numerical modeling of the tested sandwich panels is reported. Nonlinear 3-D finite element modeling of the composite action of the sandwich panel is developed using ANSYS. Solid elements with different crashing and cracking capabilities and different constitutive laws were selected for the concrete and the AAC. Contact interface elements are used in this research to adequately model the shear transfer at the interface between the different layers. The numerical results showed good correlation with the experimental ones indicating the adequacy of the model in estimating the loading capacity of panels.

Keywords: Autoclaved Aerated Concrete, Concrete Sandwich Panels, Finite Element Modeling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3091

Authors: Kejun Zhuang, Zhaohui Wen

Abstract:

This paper deals with a delayed single population model on time scales. With the assistance of coincidence degree theory, sufficient conditions for existence of periodic solutions are obtained. Furthermore, the better estimations for bounds of periodic solutions are established.

Keywords: Coincidence degree, continuation theorem, periodic solutions, time scales

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1355

Authors: Ahmed A. D. Sarhan, Chong Feng Duan, Mum Wai Yip, M. Sayuti

Abstract:

Rapid Prototyping (RP) technologies enable physical parts to be produced from various materials without depending on the conventional tooling. Fused Deposition Modeling (FDM) is one of the famous RP processes used at present. Tensile strength and compressive strength resistance will be identified for different sample structures and different layer orientations of ABS rapid prototype solid models. The samples will be fabricated by a FDM rapid prototyping machine in different layer orientations with variations in internal geometrical structure. The 0° orientation where layers were deposited along the length of the samples displayed superior strength and impact resistance over all the other orientations. The anisotropic properties were probably caused by weak interlayer bonding and interlayer porosity.

Keywords: Building orientation, compression strength, rapid prototyping, tensile strength.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1718

Authors: Properties, Approximation Distribution Reductions in Multigranulation Rough Set Model

Abstract:

Some properties of approximation sets are studied in multi-granulation optimist model in rough set theory using maximal compatible classes. The relationships between or among lower and upper approximations in single and multiple granulation are compared and discussed. Through designing Boolean functions and discernibility matrices in incomplete information systems, the lower and upper approximation sets and reduction in multi-granulation environments can be found. By using examples, the correctness of computation approach is consolidated. The related conclusions obtained are suitable for further investigating in multiple granulation RSM.

Keywords: Incomplete information system, maximal compatible class, multi-granulation rough set model, reduction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 870

Authors: Yunyong Guo, Sudhakar Ganti, Bryan Guo

Abstract:

The rapid growth of telehealth Internet of Things (IoT) devices has raised concerns about energy consumption and efficient data processing. This paper presents an energy-efficient model that integrates telehealth IoT devices with a fog and cloud computing-based platform, offering a sustainable and robust solution to overcome these challenges. Our model employs fog computing as a localized data processing layer while leveraging cloud computing for resource-intensive tasks, significantly reducing energy consumption. We incorporate adaptive energy-saving strategies. Simulation analysis validates our approach's effectiveness in enhancing energy efficiency for telehealth IoT systems integrated with localized fog nodes and both private and public cloud infrastructures. Future research will focus on further optimization of the energy-saving model, exploring additional functional enhancements, and assessing its broader applicability in other healthcare and industry sectors.

Keywords: Energy-efficient, fog computing, IoT, telehealth.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 187

Authors: Mohammed Farag, Mina Attari, S. Andrew Gadsden, Saeid R. Habibi

Abstract:

Battery state of charge (SOC) estimation is an important parameter as it measures the total amount of electrical energy stored at a current time. The SOC percentage acts as a fuel gauge if it is compared with a conventional vehicle. Estimating the SOC is, therefore, essential for monitoring the amount of useful life remaining in the battery system. This paper looks at the implementation of three nonlinear estimation strategies for Li-Ion battery SOC estimation. One of the most common behavioral battery models is the one state hysteresis (OSH) model. The extended Kalman filter (EKF), the smooth variable structure filter (SVSF), and the time-varying smoothing boundary layer SVSF are applied on this model, and the results are compared.

Keywords: State of charge estimation, battery modeling, one-state hysteresis, filtering and estimation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1715

Authors: Kazuyoshi Mori

Abstract:

In this paper, we consider the parametrization of the discrete-time systems without the unit-delay element within the framework of the factorization approach. In the parametrization, we investigate the number of required parameters. We consider single-input single-output systems in this paper. By the investigation, we find, on the discrete-time systems without the unit-delay element, three cases that are (1) there exist plants which require only one parameter and (2) two parameters, and (3) the number of parameters is at most three.

Keywords: Linear systems, parametrization, Coprime Factorization, number of parameters.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 816

Authors: M. Raciti Castelli, A. Castelli, E. Benini

Abstract:

The flow field over a flat roof model building has been numerically investigated in order to determine threedimensional CFD guidelines for the calculation of the turbulent flow over a structure immersed in an atmospheric boundary layer. To this purpose, a complete validation campaign has been performed through a systematic comparison of numerical simulations with wind tunnel experimental data. Wind tunnel measurements and numerical predictions have been compared for five different vertical positions, respectively from the upstream leading edge to the downstream bottom edge of the analyzed model. Flow field characteristics in the neighborhood of the building model have been numerically investigated, allowing a quantification of the capabilities of the CFD code to predict the flow separation and the extension of the recirculation regions. The proposed calculations have allowed the development of a preliminary procedure to be used as guidance in selecting the appropriate grid configuration and corresponding turbulence model for the prediction of the flow field over a three-dimensional roof architecture dominated by flow separation.

Keywords: CFD, roof, building, wind

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1740

Authors: A. Amar, N. Burtebayev, Kerimkulov Zhambul, Sh. Hamada, N. Amangeldi

Abstract:

the elastic scattering of protons, deuterons and 3He on 6Li at different incident energies have been analyzed in the framework of the optical model using ECIS88 as well as SPI GENOA codes. The potential parameters were extracted in the phenomenological treatment of measured by us angular distributions and literature data. A good agreement between theoretical and experimental differential cross sections was obtained in whole angular range. Parameters for real part of potential have been also calculated microscopically with singleand double-folding model for the p and d, 3He scattering, respectively, using DFPOT code. For best agreement with experiment the normalization factor N for the potential depth is obtained in the range of 0.7-0.9.

Keywords: Elastic scattering, Esis88 Code single and doublefolding model, phenomenological, DWBA.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1422

Authors: Amit Kumar Gautam, Sudipta Majumdar

Abstract:

This paper presents parameter estimation of a single-phase rectifier using extended Kalman filter (EKF). The state space model has been obtained using Kirchhoff’s current law (KCL) and Kirchhoff’s voltage law (KVL). The capacitor voltage and diode current of the circuit have been estimated using EKF. Simulation results validate the better accuracy of the proposed method as compared to the least mean square method (LMS). Further, EKF has the advantage that it can be used for nonlinear systems.

Keywords: Extended Kalman filter, parameter estimation, single phase rectifier, state space modelling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 913

Authors: Qiuyi Lyu, Bin Gong

Abstract:

Breadth-First Search (BFS) is a core graph algorithm that is widely used for graph analysis. As it is frequently used in many graph applications, improving the BFS performance is essential. In this paper, we present a graph ordering method that could reorder the graph nodes to achieve better data locality, thus, improving the BFS performance. Our method is based on an observation that the sibling relationships will dominate the cache access pattern during the BFS traversal. Therefore, we propose a frequency-based model to construct the graph order. First, we optimize the graph order according to the nodes’ visit frequency. Nodes with high visit frequency will be processed in priority. Second, we try to maximize the child nodes’ overlap layer by layer. As it is proved to be NP-hard, we propose a heuristic method that could greatly reduce the preprocessing overheads.We conduct extensive experiments on 16 real-world datasets. The result shows that our method could achieve comparable performance with the state-of-the-art methods while the graph ordering overheads are only about 1/15.

Keywords: Breadth-first search, BFS, graph ordering, graph algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 649

Authors: O. Frolov, K. Kolacek, J. Schmidt, J. Straus, V. Prukner, A. Shukurov

Abstract:

Many applications require surface modification and micro-structuring of polymers. For these purposes is mainly used ultraviolet (UV) radiation from excimer lamps or excimer lasers. However, these sources have a decided disadvantage - degrading the polymer deep inside due to relatively big radiation penetration depth which may exceed 100 μm. In contrast, extreme ultraviolet (EUV) radiation is absorbed in a layer approximately 100 nm thick only. In this work, the radiation from a discharge-plasma EUV source (with wavelength 46.9 nm) based on a capillary discharge driver is focused with a spherical Si/Sc multilayer mirror for surface modification of PMMA sample or thin gold layer (thickness about 40 nm). It was found that the focused EUV laser beam is capable by one shot to ablate PMMA or layer of gold, even if the focus is significantly influenced by astigmatism.

Keywords: ablation, capillary discharge, EUV laser, surface modification

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1572

Authors: J. K. Adedeji, S. T. Ijatuyi

Abstract:

The application of neural network using pattern recognition to study the fluid dynamics and predict the groundwater reservoirs properties has been used in this research. The essential of geophysical survey using the manual methods has failed in basement environment, hence the need for an intelligent computing such as predicted from neural network is inevitable. A non-linear neural network with an XOR (exclusive OR) output of 8-bits configuration has been used in this research to predict the nature of groundwater reservoirs and fluid dynamics of a typical basement crystalline rock. The control variables are the apparent resistivity of weathered layer (p₁), fractured layer (p₂), and the depth (h), while the dependent variable is the flow parameter (F=λ). The algorithm that was used in training the neural network is the back-propagation coded in C++ language with 300 epoch runs. The neural network was very intelligent to map out the flow channels and detect how they behave to form viable storage within the strata. The neural network model showed that an important variable g_r (gravitational resistance) can be deduced from the elevation and apparent resistivity p_a. The model results from SPSS showed that the coefficients, a, b and c are statistically significant with reduced standard error at 5%.

Keywords: Neural network, gravitational resistance, pattern recognition, non-linear.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 809

Authors: Renata L. C. Leão, João E. Nascimento, Natalia C. E. S. Nascimento, Elaine S. Vasconcelos, Mércia L. Oliveira

Abstract:

The quality control procedures of a radiopharmaceutical include the assessment of its chemical purity. The method suggested by international pharmacopeias consists of a thin layer chromatographic run. In this paper, the method proposed by the United States Pharmacopeia (USP) is compared to a direct method to determine the final concentration of aminopolyether in Fludeoxyglucose (¹⁸F-FDG) preparations. The approach (no chromatographic run) was achieved by placing the thin-layer chromatography (TLC) plate directly on an iodine vapor chamber. Both methods were validated and they showed adequate results to determine the concentration of aminopolyether in ¹⁸F-FDG preparations. However, the direct method is more sensitive, faster and simpler when compared to the reference method (with chromatographic run), and it may be chosen for use in routine quality control of ¹⁸F-FDG.

Keywords: Chemical purity, Kryptofix 222, thin layer chromatography, validation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 855

Authors: A. Pourahmad, Sh. Gharipour

Abstract:

The present work describes the preparation and characterization of nanosized SiO₂@PbS core-shell particles by using a simple wet chemical route. This method utilizes silica spheres formation followed by successive ionic layer adsorption and reaction method assisted lead sulphide shell layer formation. The final product was characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), UV–vis spectroscopic, infrared spectroscopy (IR) and transmission electron microscopy (TEM) experiments. The morphological studies revealed the uniformity in size distribution with core size of 250 nm and shell thickness of 18 nm. The electron microscopic images also indicate the irregular morphology of lead sulphide shell layer. The structural studies indicate the face-centered cubic system of PbS shell with no other trace for impurities in the crystal structure.

Keywords: Core-shell, nanostructure, semiconductor, optical property, XRD.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1209

Authors: Mahdi Sadeghian, Somaye Sadeghian, Reza Dinarvand

Abstract:

This paper is going to discuss two issues encountered in using PLAXIS. Both issues were monitored during application of PLAXIS to estimate the excavation-induced displacement. Column Soil Mixing (CSM) was applied to stabilise the excavation. It was understood that the estimated excavation induced deformation at the top of the CSM blocks highly depends on the material type defining pavement material adjacent to the CSM blocks. Cohesive material for pavement will result in the unrealistic connection between pavement and CSM even by defining an interface element. To find the most realistic approach, the interface defined in three different manners (1) no interface elements were applied (2) a non-cohesive soil layer was defined between pavement and CSM block to represent the friction between these materials (3) built-in interface elements in PLAXIS was used to define the boundary between the pavement and the CSM block. The result showed that the option 2 would result in more realistic results. The second issue was in the modelling of the contact line between the CSM block and an inclined layer underneath. The analysis result showed that the excavation-induced deformation highly depends on how the PLAXIS user defines the contact area. It was understood that if the contact area had defined as a point in which CSM block had intersected the layer underneath the estimated lateral displacement of CSM block would be unrealistically lower than the model in which the contact area was defined as a line.

Keywords: PLAXIS, FEM, CSM, excavation-induced deformation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 646

Authors: Keny Ordaz-Hernandez, Xavier Fischer, Fouad Bennis

Abstract:

This article illustrates a model selection management approach for virtual prototypes in interactive simulations. In those numerical simulations, the virtual prototype and its environment are modelled as a multiagent system, where every entity (prototype,human, etc.) is modelled as an agent. In particular, virtual prototyp ingagents that provide mathematical models of mechanical behaviour inform of computational methods are considered. This work argues that selection of an appropriate model in a changing environment,supported by models? characteristics, can be managed by the deter-mination a priori of specific exploitation and performance measures of virtual prototype models. As different models exist to represent a single phenomenon, it is not always possible to select the best one under all possible circumstances of the environment. Instead the most appropriate shall be selecting according to the use case. The proposed approach consists in identifying relevant metrics or indicators for each group of models (e.g. entity models, global model), formulate their qualification, analyse the performance, and apply the qualification criteria. Then, a model can be selected based on the performance prediction obtained from its qualification. The authors hope that this approach will not only help to inform engineers and researchers about another approach for selecting virtual prototype models, but also assist virtual prototype engineers in the systematic or automatic model selection.

Keywords: Virtual prototype models, domain, qualification criterion, model qualification, model assessment, environmental modelling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2047

Authors: Yaw-Juen Wang, Po-Chun Hsu

Abstract:

This paper presents an equivalent circuit model based on piecewise linear parallel branches (PLPB) to study solar cell modules which are partially shaded. The PLPB model can easily be used in circuit simulation software such as the ElectroMagnetic Transients Program (EMTP). This PLPB model allows the user to simulate several different configurations of solar cells, the influence of partial shadowing on a single or multiple cells, the influence of the number of solar cells protected by a bypass diode and the effect of the cell connection configuration on partial shadowing.

Keywords: Cell Connection Configurations, EMTP, Equivalent Circuit, Partial Shading, Photovoltaic Module

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2946

Authors: Chia-Ling Chang, Chung-Sheng Liao

Abstract:

The present study focuses on the discussion over the parameter of Artificial Neural Network (ANN). Sensitivity analysis is applied to assess the effect of the parameters of ANN on the prediction of turbidity of raw water in the water treatment plant. The result shows that transfer function of hidden layer is a critical parameter of ANN. When the transfer function changes, the reliability of prediction of water turbidity is greatly different. Moreover, the estimated water turbidity is less sensitive to training times and learning velocity than the number of neurons in the hidden layer. Therefore, it is important to select an appropriate transfer function and suitable number of neurons in the hidden layer in the process of parameter training and validation.

Keywords: Artificial Neural Network (ANN), sensitivity analysis, turbidity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2824

Authors: Ling-fei Xu, Qi Zhao, Yan-ru Chen, Mu-chun Zhou, Meng Zhang, Shi-xue Xu

Abstract:

SVM ( Support Vector Machine ) is a new method in the artificial neural network ( ANN ). In the steel making, how to use computer to predict the end point of BOF accuracy is a great problem. A lot of method and theory have been claimed, but most of the results is not satisfied. Now the hot topic in the BOF end point predicting is to use optical way the predict the end point in the BOF. And we found that there exist some regular in the characteristic curve of the flame from the mouse of pudding. And we can use SVM to predict end point of the BOF, just single spectrum intensity should be required as the input parameter. Moreover, its compatibility for the input space is better than the BP network.

Keywords: SVM, predict, BOF, single spectrum intensity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1374

Authors: Thiang, Handry Khoswanto, Rendy Pangaldus

Abstract:

Inverse kinematics analysis plays an important role in developing a robot manipulator. But it is not too easy to derive the inverse kinematic equation of a robot manipulator especially robot manipulator which has numerous degree of freedom. This paper describes an application of Artificial Neural Network for modeling the inverse kinematics equation of a robot manipulator. In this case, the robot has three degree of freedoms and the robot was implemented for drilling a printed circuit board. The artificial neural network architecture used for modeling is a multilayer perceptron networks with steepest descent backpropagation training algorithm. The designed artificial neural network has 2 inputs, 2 outputs and varies in number of hidden layer. Experiments were done in variation of number of hidden layer and learning rate. Experimental results show that the best architecture of artificial neural network used for modeling inverse kinematics of is multilayer perceptron with 1 hidden layer and 38 neurons per hidden layer. This network resulted a RMSE value of 0.01474.

Keywords: Artificial neural network, back propagation, inverse kinematics, manipulator, robot.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2296

Authors: R. Karmouch

Abstract:

A thin gold metal layer was deposited on the top of silicon oxide films containing embedded Si nanocrystals (Si-nc). The sample was annealed in a gas containing nitrogen, and subsequently characterized by photoluminescence. We obtained 3-fold enhancement of photon emission from the Si-nc embedded in silicon dioxide covered with a Gold layer as compared with an uncovered sample. We attribute this enhancement to the increase of the spontaneous emission rate caused by the coupling of the Si-nc emitters with the surface plasmons (SP). The evolution of PL emission with laser irradiated time was also collected from covered samples, and compared to that from uncovered samples. In an uncovered sample, the PL intensity decreases with time, approximately with two decay constants. Although the decrease of the initial PL intensity associated with the increase of sample temperature under CW pumping is still observed in samples covered with a gold layer, this film significantly contributes to reduce the permanent deterioration of the PL intensity. The resistance to degradation of light-emitting silicon nanocrystals can be increased by SP coupling to suppress the permanent deterioration. Controlling the permanent photodeterioration can allow to perform a reliable optical gain measurement.

Keywords: Photodeterioration, Silicon Nanocrystals, Ion Implantation, Photoluminescence, Surface Plasmons.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1878

Authors: A. Amar

Abstract:

Analysis of the elastic scattering of protons on ¹⁰B nuclei has been done in the framework of the optical model and single folding model at the beam energies up to 17 MeV. We could enhance the optical potential parameters using Esis88 Code, as well as SPI GENOA Code. Linear relationship between volume real potential (V₀) and proton energy (E_p) has been obtained. Also, surface imaginary potential W_D is proportional to the proton energy (E_p) in the range 0.400 and 17 MeV. The radiative reaction ¹⁰B(p,γ)¹¹C has been analyzed using potential model. A comparison between ¹⁰B(p,γ)¹¹C and ⁶Li(p,γ)⁷Be has been made. Good agreement has been found between theoretical and experimental results in the whole range of energy. The radiative resonance reaction ⁷Li(p,γ)⁸Be has been studied.

Keywords: Elastic scattering of protons on 10B nuclei, optical potential parameters, potential model, radiative reaction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 871

Authors: Zhen Peng, Baifeng Wu

Abstract:

Modern computing technology enters the era of parallel computing with the trend of sustainable and scalable parallelism. Single Instruction Multiple Data (SIMD) is an important way to go along with the trend. It is able to gather more and more computing ability by increasing the number of processor cores without the need of modifying the program. Meanwhile, in the field of scientific computing and engineering design, many computation intensive applications are facing the challenge of increasingly large amount of data. Data parallel computing will be an important way to further improve the performance of these applications. In this paper, we take the accurate collision detection in building information modeling as an example. We demonstrate a model for constructing a data parallel algorithm. According to the model, a complex object is decomposed into the sets of simple objects; collision detection among complex objects is converted into those among simple objects. The resulting algorithm is a typical SIMD algorithm, and its advantages in parallelism and scalability is unparalleled in respect to the traditional algorithms.

Keywords: Data parallelism, collision detection, single instruction multiple data, building information modeling, continuous scalability.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1248

Authors: Apoorva Vinod, Archana Mathur, Snehanshu Saha

Abstract:

Though there exists a plethora of Activation Functions (AFs) used in single and multiple hidden layer Neural Networks (NN), their behavior always raised curiosity, whether used in combination or singly. The popular AFs – Sigmoid, ReLU, and Tanh – have performed prominently well for shallow and deep architectures. Most of the time, AFs are used singly in multi-layered NN, and, to the best of our knowledge, their performance is never studied and analyzed deeply when used in combination. In this manuscript, we experiment on multi-layered NN architecture (both on shallow and deep architectures; Convolutional NN and VGG16) and investigate how well the network responds to using two different AFs (Sigmoid-Tanh, Tanh-ReLU, ReLU-Sigmoid) used alternately against a traditional, single (Sigmoid-Sigmoid, Tanh-Tanh, ReLU-ReLU) combination. Our results show that on using two different AFs, the network achieves better accuracy, substantially lower loss, and faster convergence on 4 computer vision (CV) and 15 Non-CV (NCV) datasets. When using different AFs, not only was the accuracy greater by 6-7%, but we also accomplished convergence twice as fast. We present a case study to investigate the probability of networks suffering vanishing and exploding gradients when using two different AFs. Additionally, we theoretically showed that a composition of two or more AFs satisfies Universal Approximation Theorem (UAT).

Keywords: Activation Function, Universal Approximation function, Neural Networks, convergence.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 166

Authors: MK Lee, JS Hong, SM Sin, HU Oh

Abstract:

Heat pipes are used to control the thermal problem for electronic cooling. It is especially difficult to dissipate heat to a heat sink in an environment in space compared to earth. For solving this problem, in this study, the Poiseuille (Po) number, which is the main measure of the performance of a heat pipe, is studied by CFD; then, the heat pipe performance is verified with experimental results. A heat pipe is then fabricated for a spatial environment, and an in-house code is developed. Further, a heat pipe subsystem, which consists of a heat pipe, MLI (Multi Layer Insulator), SSM (Second Surface Mirror), and radiator, is tested and correlated with the TMM (Thermal Mathematical Model) through a commercial code. The correlation results satisfy the 3K requirement, and the generated thermal model is verified for application to a spatial environment.

Keywords: CFD, Heat pipe, Radiator, Space.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1652

Authors: A.F.A. Abidin, N.S.M. Usop

Abstract:

A computer cluster is a group of tightly coupled computers that work together closely so that in many respects they can be viewed as though they are a single computer. The components of a cluster are commonly, but not always, connected to each other through fast local area networks. Clusters are usually deployed to improve performance and/or availability over that provided by a single computer, while typically being much more cost-effective than single computers of comparable speed or availability. This paper proposed the way to implement the Beowulf Cluster in order to achieve high performance as well as high availability.

Keywords: Beowulf Cluster, grid computing, GridMPI, MPICH.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1685

Authors: Anupama Pande, Vishik Goel

Abstract:

A complex valued neural network is a neural network, which consists of complex valued input and/or weights and/or thresholds and/or activation functions. Complex-valued neural networks have been widening the scope of applications not only in electronics and informatics, but also in social systems. One of the most important applications of the complex valued neural network is in image and vision processing. In Neural networks, radial basis functions are often used for interpolation in multidimensional space. A Radial Basis function is a function, which has built into it a distance criterion with respect to a centre. Radial basis functions have often been applied in the area of neural networks where they may be used as a replacement for the sigmoid hidden layer transfer characteristic in multi-layer perceptron. This paper aims to present exhaustive results of using RBF units in a complex-valued neural network model that uses the back-propagation algorithm (called 'Complex-BP') for learning. Our experiments results demonstrate the effectiveness of a Radial basis function in a complex valued neural network in image recognition over a real valued neural network. We have studied and stated various observations like effect of learning rates, ranges of the initial weights randomly selected, error functions used and number of iterations for the convergence of error on a neural network model with RBF units. Some inherent properties of this complex back propagation algorithm are also studied and discussed.

Keywords: Complex valued neural network, Radial BasisFunction, Image recognition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2419

Authors: John Kabuba, Antoine Mulaba-Bafubiandi, Kim Battle

Abstract:

In this study a neural network (NN) was proposed to predict the sorption of binary mixture of copper-cobalt ions into clinoptilolite as ion-exchanger. The configuration of the backpropagation neural network giving the smallest mean square error was three-layer NN with tangent sigmoid transfer function at hidden layer with 10 neurons, linear transfer function at output layer and Levenberg-Marquardt backpropagation training algorithm. Experiments have been carried out in the batch reactor to obtain equilibrium data of the individual sorption and the mixture of coppercobalt ions. The obtained modeling results have shown that the used of neural network has better adjusted the equilibrium data of the binary system when compared with the conventional sorption isotherm models.

Keywords: Adsorption isotherm, binary system, neural network; sorption

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2051

Authors: Mohammad Shanbghazani, Vahid Heidarpour, Iraj Mirzaee

Abstract:

In this study a two dimensional axisymmetric, steady state and incompressible laminar flow in a rotating single disk is numerically investigated. The finite volume method is used for solving the momentum equations. The numerical model and results are validated by comparing it to previously reported experimental data for velocities, angles and moment coefficients. It is demonstrated that increasing the axial distance increases the value of axial velocity and vice versa for tangential and total velocities. However, the maximum value of nondimensional radial velocity occurs near the disk wall. It is also found that with increase rotational Reynolds number, moment coefficient decreases.

Keywords: Rotating disk, Laminar flow, Numerical, Momentum

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1593