4858 results
Search Results
Now showing 1 - 10 of 4858
Article Citation - WoS: 39Citation - Scopus: 49Focus Variation Measurement and Prediction of Surface Texture Parameters Using Machine Learning in Laser Powder Bed Fusion(Asme, 2020) Ozel, Tugrul; Altay, Ayca; Kaftanoglu, Bilgin; Leach, Richard; Senin, Nicola; Donmez, AlkanThe powder bed fusion-based additive manufacturing process uses a laser to melt and fuse powder metal material together and creates parts with intricate surface topography that are often influenced by laser path, layer-to-layer scanning strategies, and energy density. Surface topography investigations of as-built, nickel alloy (625) surfaces were performed by obtaining areal height maps using focus variation microscopy for samples produced at various energy density settings and two different scan strategies. Surface areal height maps and measured surface texture parameters revealed the highly irregular nature of surface topography created by laser powder bed fusion (LPBF). Effects of process parameters and energy density on the areal surface texture have been identified. Machine learning methods were applied to measured data to establish input and output relationships between process parameters and measured surface texture parameters with predictive capabilities. The advantages of utilizing such predictive models for process planning purposes are highlighted.Article Citation - WoS: 9Citation - Scopus: 9Existence and Uniqueness of Best Proximity Points Under Rational Contractivity Conditions(Walter de Gruyter Gmbh, 2016) Karapinar, Erdal; Roldan-Lopez-de-Hierro, Antonio-Francisco; Sadarangani, KishinThe main aim of this paper is to present some theorems in order to guarantee existence and uniqueness of best proximity points under rational contractivity conditions using very general test functions. To illustrate the variety of possible test functions, we include some examples of pairs of functions which are included in innovative papers published in the last years. As a consequence, we prove that our results unify and extend some recent results in this field.Article Citation - WoS: 11Citation - Scopus: 20Reinforcement Learning Using Fully Connected, Attention, and Transformer Models in Knapsack Problem Solving(Wiley, 2022) Yildiz, Beytullah; Yıldız, Beytullah; Yıldız, BeytullahKnapsack is a combinatorial optimization problem that involves a variety of resource allocation challenges. It is defined as non-deterministic polynomial time (NP) hard and has a wide range of applications. Knapsack problem (KP) has been studied in applied mathematics and computer science for decades. Many algorithms that can be classified as exact or approximate solutions have been proposed. Under the category of exact solutions, algorithms such as branch-and-bound and dynamic programming and the approaches obtained by combining these algorithms can be classified. Due to the fact that exact solutions require a long processing time, many approximate methods have been introduced for knapsack solution. In this research, deep Q-learning using models containing fully connected layers, attention, and transformer as function estimators were used to provide the solution for KP. We observed that deep Q-networks, which continued their training by observing the reward signals provided by the knapsack environment we developed, optimized the total reward gained over time. The results showed that our approaches give near-optimum solutions and work about 40 times faster than an exact algorithm using dynamic programming.Article W-Band RCS Prediction of Small Objects: Comparing Two Widely Used Methods with Experimental Validation(Gazi University, 2025) Kara, Ali; Aydın, Elif; Yardım, Funda Ergün; Sezgin, Deniz; Ergun Yardim, FundaThis paper compares the accuracy of Shooting and Bouncing Rays and Electric Field Integral Equation methods for Radar Cross Section prediction of small objects at 77-81 GHz band. Existing studies on RCS prediction methods often lack comprehensive comparisons between computational and experimental results, particularly for small objects measured with a 77 GHz radar. This study addresses this gap by presenting an in-depth analysis of both simulation and measurement data. In this work, three targets with varying geometries and materials were measured with a frequency modulated continuous wave radar and simulated using Ansys HFSS and CST Studio Suite. The measurements were performed with a commercial off-the-shelf (COTS) frequency modulated continuous wave radar operating at 77–81 GHz. This study aims to emphasize the importance of considering both efficiency and accuracy when opting for an RCS prediction method. Overall, the outcomes of both methods have largely demonstrated good alignment. It has been noted that, while Shooting and Bouncing Rays method offers promising time-saving advantages, Electric Field Integral Equation method remains a valuable tool for complex geometries where precise results are crucial.Article Citation - WoS: 1Citation - Scopus: 1Novel Frequency-Domain Criterion for Elimination of Limit Cycles in a Class of Digital Filters With Single Saturation Nonlinearity(Pergamon-elsevier Science Ltd, 2008) Singh, VimalA frequency-domain criterion for the elimination of limit cycles in a class of digital filters utilizing single saturation nonlinearity is presented. The criterion is derived by exploiting the structural properties of the system under consideration in a greater detail. A novel feature of the criterion is that it takes the form of a matrix inequality, despite the fact that there is single nonlinearity in the system. An example showing the effectiveness of the criterion is given. (c) 2006 Elsevier Ltd. All rights reserved.Article Citation - WoS: 6Citation - Scopus: 7Construction of Self-Assembled Vertical Nanoflakes on Cztsse Thin Films(Iop Publishing Ltd, 2019) Terlemezoglu, M.; Surucu, O. Bayrakli; Colakoglu, T.; Abak, M. K.; Gullu, H. H.; Ercelebi, C.; Parlak, M.; Bayrakli Sürücü, O.Cu2ZnSn(S, Se)(4) (CZTSSe) is a promising alternative absorber material to achieve high power conversion efficiencies, besides its property of involving low-cost and earth-abundant elements when compared to Cu(In, Ga) Se-2 (CIGS) and cadmium telluride (CdTe), to be used in solar cell technology. In this study, a novel fabrication technique was developed by utilizing RF sputtering deposition of CZTSSe thin films having a surface decorated with self-assembled nanoflakes. The formation of nanoflakes was investigated by detailed spectroscopic method of analysis in the effect of each stacked layer deposition in an optimized sequence and the size of nanoflakes by an accurate control of sputtering process including film thickness. Moreover, the effects of substrate temperature on the formation of nanoflakes on the film surface were discussed at a fixed deposition route. One of the main advantages arising from the film surface with self-assembled nanoflakes is the efficient light trapping which decreases the surface reflectance. As a result of the detailed production and characterization studies, it was observed that there was a possibility of repeatable and controllable fabrication sequence for the preparation of CZTSSe thin films with self-textured surfaces yielding low surface reflectance.Article Citation - WoS: 7Citation - Scopus: 10Comparison of Three Different Learning Methods of Multilayer Perceptron Neural Network for Wind Speed Forecasting(Gazi Univ, 2021) Bulut, Mehmet; Tora, Hakan; Buaisha, Dr.magdi; Buaisha, MagdiIn the world, electric power is the highest need for high prosperity and comfortable living standards. The security of energy supply is an essential concept in national energy management. Therefore, ensuring the security of electricity supply requires accurate estimates of electricity demand. The share of electricity generation from renewables is significantly growing in the world. This kind of energy types are dependent on weather conditions as the wind and solar energies. There are two vital requirements to locate and measure specific systems to utilize wind power: modelling and forecasting of the wind velocity. To this end, using only 4 years of measured meteorological data, the present research attempts to estimate the related speed of wind within the Libyan Mediterranean coast with the help of ANN (artificial neural networking) with three different learning algorithms, which are Levenberg-Marquardt, Bayesian Regularization and Scaled Conjugate Gradient. Conclusions reached in this study show that wind speed can be estimated within acceptable limits using a limited set of meteorological data. In the results obtained, it was seen that the SCG algorithm gave better results in tests in this study with less data.Article Citation - WoS: 31Citation - Scopus: 40Variational Mode Decomposition-Based Radio Frequency Fingerprinting of Bluetooth Devices(Ieee-inst Electrical Electronics Engineers inc, 2019) Aghnaiya, Alghannai; Ali, Aysha M.; Kara, AliRadio frequency fingerprinting (RFF) is based on identification of unique features of RF transient signals emitted by radio devices. RF transient signals of radio devices are short in duration, non-stationary and nonlinear time series. This paper evaluates the performance of RF fingerprinting method based on variational mode decomposition (VMD). For this purpose, VMD is used to decompose Bluetooth (BT) transient signals into a series of band-limited modes, and then, the transient signal is reconstructed from the modes. Higher order statistical (HOS) features are extracted from the complex form of reconstructed transients. Then, Linear Support Vector Machine (LVM) classifier is used to identify BT devices. The method has been tested experimentally with BT devices of different brands, models and series. The classification performance shows that VMD based RF fingerprinting method achieves better performance (at least 8% higher) than time-frequency-energy (TFED) distribution based methods such as Hilbert-Huang Transform. This is demonstrated with the same dataset but with smaller number of features (nine features) and slightly lower (2-3 dB) SNR levels.Article Citation - WoS: 4Citation - Scopus: 5Forecasting Turkish Local Elections(Elsevier, 2012) Toros, EmreThe literature on political forecasting is large, although the main focus of this literature is limited to a number of countries. Nevertheless, and despite the major differences between political systems, scientific forecasting work has proved to be broadly possible, with noteworthy extensions to new countries. This article extends the literature further by developing a new forecasting model for local elections in Turkey. The basic motivation of this article is to test the usefulness of political forecasting in the contexts of alternative democratic settings. Turkey, in that sense, seems to be an interesting case for a number of reasons. First, the Turkish Republic has been a multi-party democracy since the mid-1940s. Although it has been interrupted by three military coups, the party and election system in Turkey has brought real alternations in the government starting from very early years of the multi-party system. So, it is plausible to argue that Turkish voters have the tradition of evaluating the performances of political parties, as in any other Western-type democracy. That is to say, the dynamics of evaluations of political parties in Turkey follow a similar pattern to other contemporary democracies, being driven by economic and political forces. The main contribution of this analysis is the introduction of an explicit model, which can forecast the impact of economic and political variables across local elections in Turkey by using reliable, public, and macro-level data. In particular, this study offers a new forecasting model which tries to forecast the Justice and Development Party's (Adalet ve Kalkinma Partisi, AKP) vote share in 81 cities. (C) 2012 International Institute of Forecasters. Published by Elsevier B.V. All rights reserved.Article Citation - WoS: 10Citation - Scopus: 18Evaluation Criteria for Object-Oriented Metrics(Budapest Tech, 2011) Misra, Sanjay; Computer EngineeringIn this paper an evaluation model for object-oriented (OO) metrics is proposed. We have evaluated the existing evaluation criteria for OO metrics, and based on the observations, a model is proposed which tries to cover most of the features for the evaluation of OO metrics. The model is validated by applying it to existing OO metrics. In contrast to the other existing criteria, the proposed model is simple in implementation and includes the practical and important aspects of evaluation; hence it suitable to evaluate and validate any OO complexity metric.
