DSpace@ATÜ (Adana Alparslan Türkeş Bilim ve Teknoloji Universiti)
Not a member yet
3463 research outputs found
Sort by
Electromagnetic shielding effectiveness of Ba(Zn1/3Nb2/3)O3: chopped strands composites for wide frequency applications
No full textIn this work, the conventional mixed oxide process was used to form Ba(Zn1/3Nb2/3)O-3-chopped strands composites. The single phase compound with the chemical equation Ba(Zn1/3Nb2/3)O-3 was generated after sintering at 1250 degrees C for 4 h. Varied amounts of Ba(Zn1/3Nb2/3)O-3-chopped strands were fabricated for the structural examination. X-ray diffraction (XRD), energy-dispersive X-ray spectroscopy (EDS) and scanning electron microscopy (SEM) were carried out for the structural analysis, and the results showed that the second phase did not form in Ba(Zn1/3Nb2/3)O-3. In addition to, the Ba(Zn1/3Nb2/3)O-3-strands composites were created by hot pressing using the compositions of Ba(Zn1/3Nb2/3)O-3-chopped strands in various rates and epoxy. The Ba(Zn1/3Nb2/3)O-3-chopped strands compound created in varied weights, and epoxy powder were utilized to manufacture microwave shielding effect composites. Using a two-port vector network analyser (VNA) device, the microwave shielding effect of Ba(Zn1/3Nb2/3)O-3-chopped strands composites was examined in a range of 0-8 GHz. A minimal of -44.75 dB shielding effectiveness value was reached at 6.24 GHz and 1.4 mm in thick. The Ba(Zn1/3Nb2/3)O-3-chopped strands were fabricated as composite and their properties were defined characterized for shielding effectiveness. To modify microwave shielding effectiveness, the content of components in the specimens may be adjusted for the wider and required frequency ranges
Determination of electronic band structure of quaternary ferromagnetic Ga0.97-yMn0.03CryAs epitaxial layers
No full textIntroducing transition metals to conventional III-V semiconductors anomalously changes their fundamental characteristics, such as electronic, magnetic, and structural properties. In this study, we show that the valence band anti-crossing (VBAC) model can be exploited to calculate the electronic band structure of the quaternary Ga0.97-xMn0.03CrxAs epitaxial layers. In this model, the localized Mn and Cr defect states interact with the valence band states (VB), reconstructing VBs and splitting each VB state. The splitting top of the VB state forms an impurity band (IB) and fundamental VB edge. Photomodulated reflectance (PR) spectroscopy is exploited to determine optical transition energies at room temperature. PR spectra were analyzed with the third derivative functional form (TDFF) signal's line shape. The experimental optical transition energies, including band-to-band and spin split-off band transitions, match the calculated optical transition energies by the VBAC model. In the calculation, the interaction energy between localized Mn/Cr-energy level and valence band edges is experi-mentally determined as 0.7 eV.Scientific Research Projects Coor-dination Unit of Istanbul University [FDP-2020-36589, FBG-2022-38573]; [FDK-2019-35447]This study was supported by the Scientific Research Projects Coor-dination Unit of Istanbul University (Project No: FDK-2019-35447, FDP-2020-36589, and FBG-2022-38573) . We acknowledge Prof Hiroyuki AKINAGA for supplying the samples
Voltage Sag Detection and Compensation Signal Extraction for Power Quality Mitigation Devices
No full textThe importance of voltage quality is continuously increasing in electrical networks due to the rising manufacturing costs resulting from system faults and disturbances in utility dynamics. Researchers generally prefer reference-frame transformation-based methods to detect and mitigate these disturbances. However, these methods are adversely affected during unbalanced loading and disturbances due to their direct dependence on system dynamics (currents and voltages). In this study, a new and simple method based on Clarke transformation is proposed to detect disturbances and generate compensation signals for Power Quality Mitigation Devices. The aim is to address the deficiencies of existing approaches. Firstly, the Clarke transformation is introduced through the vector presentation. Then, the mathematical derivation of the proposed method is provided to enhance readers' understanding. The voltage sag detection and compensation signal extraction of its control algorithm for a Dynamic Voltage Restorer is illustrated graphically. Subsequently, a simple power system is created using a simulation program. Balanced and unbalanced voltage disturbances are applied to the test system to demonstrate the validation of the proposed method under distorted system conditions. The results of voltage sag detection and compensation signal extraction for both the proposed and existing methods are compared at the end of the case studies
The study of the quality parameters of the tortilla chips products formulated from mixtures of corn flour and legumes
No full textCurrent study was designed to evaluate chemical composition and quality characteristics of tortilla chips made from corn flour with chickpea flour and red lentil flour in 20%, 50%, 80%, and 100%. The results suggested that moisture, and fat contents and acidity values increased with increasing levels of chickpea and lentil flour. However peroxide value increased with increasing levels of chickpea flour and decreased with increasing levels of lentil flour. Most studies that were carried out on tortilla chips made from corn flour and chickpeas revealed the presence of an unsaturated fatty acid content of up to 90%. Between 17.8 g of unsaturated fatty acids were present in each serving of tortilla chips loaded with chickpeas and samples. These may be monounsaturated or polyunsaturated, with approximately 50% being polyunsaturated. The two most important PUFA acids, linoleic acid and gamma-linoleic acid are essential fatty acids. In terms of the composition profile, the best possible results were achieved by samples that contain a 50% addition. In conclusion, it has been shown that corn flour incorporating chickpea is more nutritious than cornmeal alone. The purpose for which lentils and chickpeas were chosen is due to the high protein content and the entire nutritional composition
Ripple and Passive Components Reduction on All SiC Multi-Device Interleaved Boost Converter for Hybrid Electric Vehicles
No full textDue to a smaller environmental footprint than conventional vehicles, electric vehicles (EVs) may help in the fight against climate change. In spite of their many advantages, especially those in maintenance, disadvantages such as less traveling range, long recharging time and the high price tag-compared to gas-powered vehicles-are essential barriers to extensive usage of EVs. The forenamed limitations can be partly overcome by advancing electrical equipment in terms of efficiency, weight and compactness. The development of novel materials for circuit components and improvements in circuit design have been the primary focus of advancement with regard to DC/DC converters, one of the fundamental building blocks of electrical equipment. Therefore, this paper proposes an All-SiC-based multi-device interleaved boost converter (MDIBC) to replace the Silicon insulated gate bipolar transistors (Si IGBT)-based BC topology in Toyota Prius. To achieve high efficiency from the proposed converter, the power efficiency is examined with respect to the input current ripple values. The obtained results disclose the best efficiency is achieved for a 10% input current ripple with a 50% decrease in output voltage ripple. Additionally, for these ripple values, other decisive parameters are determined and compared to the benchmark converters. The outcomes reveal that (at 70 kHz switching frequency), the best values of inductor and capacitor with reduction in their values of 83.55% and 96.45%, respectively – are achieved. This considerable decline not only results in downsizing the passive components but also leads to a higher power density that has a crucial importance for power efficiency. © 2023, Tafila Technical University. All rights reserved
Uyarlanabilir K-değeri seçimi ile birleştirilmiş komşuların çoğunluk oylamasının ve ortalama uzaklıklarının harmonik ortalamasına bağlı yeni bir KNN sınıflandırıcı
No full textLisansüstü Eğitim Enstitüsü, Bilgisayar Mühendisliği Ana Bilim DalıBu çalışma, KNN algoritmasının geliştirilmiş bir versiyonunu sunarak geleneksel KNN'nin sınıflandırma doğruluğunu iyileştirmeyi amaçlamaktadır. Bu tez, uyarlanabilir ve artımlı k-değeri seçimi ile çoğunluk oylamasının ve ortalama uzaklık harmonik ortalamasına dayanan geliştirilmiş bir KNN sınıflandırıcısı olan HMAKNN'ı önermektedir. Bu çalışma kapsamında, bir ağırlıklandırma mekanizmasının olup olmamasına bağlı olarak HMAKNN'nin sıradan ve ağırlıklı olmak üzere iki farklı türü tasarlanmıştır. Bu metotların adları sırasıyla HMAKNNR ve HMAKNNW'dir. Bu HMAKNN sınıflandırıcıları, sekizi PRTools'tan elde edilen sentetik ve yirmi altısı UCI ve Kaggle depolarından alınan gerçek kıyaslama veri setleri olmak üzere toplam otuz dört veri seti üzerinde değerlendirilmiştir. Önerilen yaklaşımların sınıflandırma etkinliğini ve başarısını belirlemek için geleneksel KNN ve dört iyi bilinen ağırlıklı KNN modeli ile karşılaştırılmıştır. Diğer ağırlıklandırma yöntemlerinin aksine, her iki HMAKNN sınıflandırıcısı da çoğunluk oylaması ve ortalama uzaklık arasındaki sinerjinin yanı sıra k-değerini uyarlamalı olarak değiştirme becerisini kullanarak sınıflandırma doğruluğunu önemli ölçüde artırmıştır. Gerçek kıyaslama veri setlerinin sınıflandırılması sonuçlarına göre, HMAKNNW %83,02 ile en yüksek ACC değerini üretirken, HMAKNNR %82,83 ile ikinci en yüksek değere ulaşmıştır. Karşılaştırılan diğer beş yöntemden elde edilen ortalama sonuçlara göre HMAKNNW ve HMAKNNR sırasıyla %2,69 ve %2,50 üstünlük sağlamıştır. Ayrıca, gerçek kıyaslama veri setlerinin sınıflandırma sonuçları, her iki HMAKNN yönteminin de istatistiksel olarak diğer ağırlıklı KNN yöntemlerinden daha iyi performans gösterdiğini göstermektedir.This study aims to improve the classification accuracy of traditional KNN by presenting an improved version of the KNN algorithm. This thesis proposes HMAKNN, an improved KNN classifier based on the harmonic mean of the majority voting and average distance phenomena with adaptive and incremental k-value selection. Within the purview of this study, two variants of HMAKNN, regular and weighted, were designed based on the presence or absence of a weighting mechanism. The names of these variants are HMAKNNR and HMAKNNW, respectively. These HMAKNN classifiers were evaluated on a total of thirty-four data sets, of which eight were synthetic obtained from PRTools and twenty-six were real benchmark data sets from UCI and Kaggle repositories. To determine the classification effectiveness and success of the proposed approaches, they were compared with traditional KNN and four well-known weighted KNN models. In contrast to other weighting methods, both HMAKNN classifiers utilize the constructive interaction between majority voting and average distance, as well as the ability to adaptively modify the k-value, thereby substantially enhancing classification accuracy. Based on the results of the classification of real benchmark data sets, HMAKNNW produced the highest ACC value with 83.02%, and HMAKNNR reached the second-highest value with 82.83%. HMAKNNW and HMAKNNR provided an advantage of 2.69% and 2.50% over the other five methods compared, respectively. Furthermore, the classification results of real benchmark data sets indicate that both HMAKNN methods statistically outperform other weighted KNN methods
Casein-Hydrolysate-Loaded W/O Emulsion Preparation as the Primary Emulsion of Double Emulsions: Effects of Varied Phase Fractions, Emulsifier Types, and Concentrations
No full textStable primary emulsion formation in which different parameters such as viscosity and droplet size come into prominence for their characterization is a key factor in W/O/W emulsions. In this study, different emulsifiers (Crill (TM) 1, Crill (TM) 4, AMP, and PGPR) were studied to produce a casein-hydrolysate-loaded stable primary emulsion with lower viscosity and droplet size. Viscosity, electrical conductivity, particle size distribution, and emulsion stability were determined for three different dispersed phase ratios and three emulsifier concentrations. In 31 of the 36 examined emulsion systems, no electrical conductivity could be measured, indicating that appropriate emulsions were formed. While AMP-based emulsions showed non-Newtonian flow behaviors with high consistency coefficients, all PGPR-based emulsions and most of the Crill (TM)-1- and -4-based ones were Newtonian fluids with relatively low viscosities (65.7-274.7 cP). The PGPR-based emulsions were stable for at least 5 days and had D(90) values lower than 2 mu m, whereas Crill (TM)-1- and -4-based emulsions had phase separation after 24 h and had minimum D(90) values of 6.8 mu m. PGPR-based emulsions were found suitable and within PGPR-based emulsions, and the best formulation was determined by TOPSIS. Using 5% PGPR with a 25% dispersed phase ratio resulted in the highest relative closeness value. The results of this study showed that PGPR is a very effective emulsifier for stable casein-hydrolysate-loaded emulsion formations with low droplet size and viscosity
Instant soluble roselle (Hibiscus sabdariffa L.) powder rich in bioactive compounds: Effect of the production process on volatile compounds
No full textThis study investigated the optimum extraction conditions of bioactive compounds from roselle (Hibiscus sabdariffa), the effect of the air inlet temperatures of spray drying on the final product, and variations in the volatile compounds during converting the plant extracts into soluble powder forms. The extraction was conducted at 54 degrees C for 120min to obtain extracts with superior total phenolic content (TPC) (2032.71mg GAE/100g), total anthocyanin content (TAC) (995.37mg Dp-3-sam/100g), DPPH radical scavenging activity (9.48g sample/g DPPH), and ABTS radical scavenging activity (3326.15mg TEAC/100g). The powder forms of the extracts were produced at air inlet temperatures of 165 and 200 degrees C by spray drying and the process was confirmed by FTIR spectroscopy. An increment in air inlet temperature improved the microencapsulation yield (MY) (76.53% for 165 degrees C and 80.83% for 200 degrees C) but adversely affected the solubility (96.86% for 165 degrees C and 90.22% for 200 degrees C). The HPLC results showed that anthocyanins tended to degrade in powders produced at higher temperatures. No interaction was detected between process temperatures and antioxidant properties (p > 0.05). However, there were remarkable changes in the concentration of volatile compounds during the process.Erciyes University Scientific Research Projects Unit [FYL-2018-8029]This study was financially supported by Erciyes University Scientific Research Projects Unit (Project no: FYL-2018-8029)
Which Sentence Representation is More Informative: An Analysis on Text Classification
No full textGeorgetown College of Arts and Sciences; Georgetown Department of Linguistics; Georgetown Faculty of Languages and Linguistics7th International Conference on Dependency Linguistics (Depling, GURT/SyntaxFest 2023), Depling 2023 -- 9 March 2023 through 12 March 2023 -- Virtual, Online -- 192806Text classification is a popular and well-studied problem in Natural Language Processing. Most previous work on text classification has focused on deep neural networks such as LSTMs and CNNs. However, text classification studies using syntactic and semantic information are very limited in the literature. In this study, we propose a model using Graph Attention Network (GAT) that incorporates semantic and syntactic information as input for the text classification task. The semantic representations of UCCA and AMR are used as semantic information and the dependency tree is used as syntactic information. Extensive experimental results and in-depth analysis show that UCCA-GAT model, which is a semantic-aware model outperforms the AMR-GAT and DEP-GAT, which are semantic and syntax-aware models respectively. We also provide a comprehensive analysis of the proposed model to understand the limitations of the representations for the problem. © 2023 Association for Computational Linguistics
Pterostilbene loaded poly(vinyl alcohol)-gelatin cryogels as potential bioactive wound dressing material
No full textCryogels are support materials which are good at mimicking extracellular matrix due to their excellent hydrophilicity, biocompatibility, and macroporous structure, thus they are useful in facilitating cell activities during healing process. In this study, polyvinyl alcohol-gelatin (PVA-Gel) based cryogel membranes loaded with pterostilbene (trans-3,5-dimethoxy-4-hydroxystilbene; PTS) (PVA-Gel/PTS) was synthesized as wound dressing materials. PVA-Gel and PVA-Gel/PTS were synthesized with the polymerization yields of 96% ± 0.23% and 98% ± 0.18%, respectively, and characterized by swelling tests, Brunauer–Emmett–Teller (BET) and scanning electron microscopy (SEM) analysis. The swelling ratios were calculated as 98.6% ± 4.93% and 102% ± 5.1%, macroporosities were determined as 85% ± 2.13% and 88% ± 2.2%, for PVA-Gel and PVA-Gel/PTS, respectively. It was determined that PVA-Gel and PVA-Gel/PTS have 17 m2/g ± 0.76 m2/g and 20 m2/g ± 0.92 m2/g surface areas, respectively. SEM studies were demonstrated that they have ~100 ?m pore sizes. According to 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), trypan blue exclusion and live-dead assay results, it was observed that cell proliferation, cell number and cell viability were higher in PVA-Gel/PTS cryogel at 24, 48, and 72 h compared to PVA-Gel. A strong and transparent fluorescent light intensity was observed indicating higher cell population in PVA-Gel/PTS in comparison with PVA-Gel, according to 4?,6-diamidino-2-phenylindole (DAPI) staining. SEM, F-Actin, Giemsa staining and inverted-phase microscope image of fibroblasts in PVA-Gel/PTS cryogels revealed that dense fibroblast proliferation and spindle-shaped morphology of cells were preserved. Moreover, DNA agarose gel data demonstrated that PVA-Gel/PTS cryogels had no effect on DNA integrity. Consequently, produced PVA-Gel/PTS cryogel can be used as wound dressing material to promote wound therapies, inducing cell viability and proliferation. © 2023 Wiley Periodicals LLC.Sabinsa Corporatio