6746 research outputs found
Sort by
Aggregation-Induced Red-Shift Emission From Self-Assembled Planar Naphthalene Diimide Dye: Interlayer in a Schottky-Type Photodiode and Dft Studies
No full textIn this study, a planar, soluble, thin film-forming and self-assembled small naphthalene diimide (3) molecule with a subtle moiety at the imide-nitrogen was synthesized, and applied for the first time in literature as an interfacial layer between Al and p-Si layers in a Schottky-type photodiode. The morphology of the compound was examined by scanning electron microscopy (SEM) and atomic force microscopy (AFM). The thin film structure and morphology affected the optical and electrical properties. The energy levels of the highest occupied molecular orbitals and lowest unoccupied molecular orbitals of 3 were calculated as -6.14 eV and -4.02 eV, corresponding to the band gap of 2.12 eV consistent with density functional theory (DFT) results. Differential scanning calorimetry (DSC) studies revealed a relatively high Tg value at 208 degrees C, indicating high-temperature applicability of the crystalline structure. The I-V measurements of Al/3/p-Si heterostructure were performed under dark and various light power intensities. The current steadily rose with each incremental 20 mW increase in light intensity. The reverse current increased almost 10-fold at 100 mW/cm2 illumination compared to dark measurement. The photodiode's responsivity, photosensitivity, and detectivity factors were elucidated. The photodiode's characteristic values, such as Io, n, phi b, and Rs, were obtained as 3.50 x 10-6 A, 8.24, 0.588 eV and 2.266 k Omega, respectively. The fabricated Schottky-type diode showed promising results for the optoelectronic field. The compound's perfect solubilities in a wide range of solvents, processability, excellent chemical and photochemical stabilities, and exciting optical, thermal and electrochemical properties make it an ideal candidate for thin film and molecular electronics applications.The support from Eastern Mediterranean University BAP-Projects Research Funding (BAPC-04-21-06) is acknowledged. We want to thank Prof. Dr. Murat Y ; imath;ld ; imath;r ; imath;m and Ali Akbar Hussaini from Selcuk University for providing equipment for photodiode studies.Eastern Mediterranean University BAP-Projects Research Funding [BAPC-04-21-06
Highly Conductive and Uniform Pedot on Poly(acrylic Acid-Vinylbenzyl Chloride) Functionalized Surfaces
No full textThis study demonstrates increasing the uniformity and conductivity of poly(3,4-ethylene dioxythiophene) (PEDOT) thin films synthesized by vapor phase polymerization (VPP) through the utilization of a thin interfacial prime layer on the substrate surface. The prime layer, which is a copolymer of acrylic acid (AA) and vinylbenzyl chloride (VBC), was coated on the substrate surface using initiated chemical vapor deposition (iCVD) method. FTIR and XPS were used to analyze the structure of as-deposited films. The use of P(AA-VBC) copolymer as a prime layer allowed uniform and complete coverage of the oxidant solution on the substrate surface due to the hydrophilic nature of the AA constituent. During the VPP, the existence of the chlorine ions originating from the VBC constituent allowed in-situ doping of the as-deposited polymer, which contributes to the increased uniformity and the conductivity. The experimental studies were carried out to show the increase in uniformity, conductivity and adherence of the as-deposited PEDOT film in the presence of the prime layer. There was nearly a 4-fold increase in the conductivity of as-deposited PEDOT in the presence of the prime layer, with measured conductivity uniformity as high as 96% over a 5x5 cm2 glass surface.Acknowledgements This study was supported by the Scientific Research Projects Council of Konya Technical University with grant number 241116012.Scientific Research Projects Council of Konya Technical University [241116012
Reel-To Coating of a Conductive Polymer on Synthetic Textile Yarns in a Semi-Closed Batch Oxidative Cvd System
No full textIn this manuscript, we demonstrate the ability to use a reel-to-reel processing technology for conductive surface functionalization of textile yarns using oxidative chemical vapor deposition in a continuous manner. We designed and built a vacuum deposition system, which allows the winding of yarns into the oCVD reactor by unreeling from the outside atmosphere, where the yarn is pre-treated with oxidant solution. Iron(III)chloride (FeCl3) and 3,4-ethylenedioxythiophene were used as the oxidant and monomer, respectively, to deposit thin films of poly(3,4-ethylenedioxythiophene) (PEDOT) thin films on the synthetic PET yarn surfaces. FTIR and XPS analyses were carried out to verify the chemical structure of as-deposited PEDOT films. Effects of temperature, oxidant concentration, and winding speed on the electrical conductivities of the yarns after oCVD were studied. All yarns exhibited non-zero conductivity values independent of the deposition conditions studied. Very high conductivity uniformities were observed along the longitudinal direction of the yarns even at the highest studied winding speed of 24 cm/min.The authors thank to the Korteks Textile Industry and Trade Company Inc. for the financial support of this study.KORTEKS; Korteks Textile Industry and Trade Company Inc
Decomposed Fuzzy Analytical Hierarchy Process Method for Business Processes Management Software Selection
No full textInternational Conference on Intelligent and Fuzzy Systems, INFUS 2024 -- 16 July 2024 through 18 July 2024 -- Canakkale -- 318029Software has been developed to help organizations manage their business processes. This is called Business Process Management (BPM) software. Organizations may have to make a decision whether to improve their business processes with the help of this software or to replace one software module with another software module. It is important to determine the criteria to satisfy the needs and get the opinions of the decision makers to make an efficient decision in this problem. For this purpose, Decomposed Fuzzy Analytic Hierarchy Process (DF-AHP) method is proposed in this study. The applicability of the proposed method in a financial institution is demonstrated. It was found that the module used by the organization is systematically backward. It needs to be changed. The business process life cycle of the financial institution has been evaluated in 6 main groups as criteria. These groups are Modeling, Design, Deployment, Execution ; Operation, Monitoring ; Control and Analysis. The business process management software identified were Activiti, Bonita, jBPM, Process Maker, uEngine, Camunda and YAWL. As a result, jBPM was selected as the best software in the Business Process Management category. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024
Environmentally Friendly Alternative Depressants in Chalcopyrite Flotation
No full textEfficient and eco-friendly technologies are required to reduce the environmental effects of the beneficiation plants. Recently, there has been an interest in environmentally friendly alternative depressants. This study explored the effects of various depressants (AERO633, AERO7260, DP3124, DP3125, carboxymethyl cellulose (CMC), Cyquest4000, and sodium silicate) on chalcopyrite flotation. The experimental results indicated that AERO7260 has the best effect on chalcopyrite flotation, and the dosage is 50 g/t, it can improve the chalcopyrite recovery up to 93.35% as well as chalcopyrite grade up to 36.97%. Meanwhile, AERO7260 had an obvious depression effect on pyrite, reducing pyrite grade to 61.10% and pyrite recovery to 3.35%. Given its low reagent dosage and high selectivity, AERO7260 is a promising depressant that can be used for chalcopyrite flotation. The use of organic depressants will significantly aid the transition to ecologically friendly and sustainable processes. This study shows that organic depressants in chalcopyrite flotation can replace the existing inorganic depressants used in the depression of gangue minerals and that metals can be recovered at high performance by processing appropriately.This study was funded by the Scientific Research Projects Funding Unit of the Selcuk University [Project number 18401029].Scientific Research Projects Funding Unit of the Selcuk University [18401029
The Novelty of Silver Extraction by Leaching in Acetic Acid With Hydrogen Peroxide as an Organic Alternative Lixiviant for Cyanide
No full textIn this article, the dissolution kinetics of pure metallic silver in acetic acid with a hydrogen peroxide solution were carried out. The effects of stirring speed, acetic acid concentration, hydrogen peroxide concentration, and temperature were examined. The results show that increasing the stirring speed decomposes the hydrogen peroxide and negatively affects the dissolution rate of silver. In addition, an acetic acid concentration in the range of 0.25–1 M has a positive effect and a negative effect in the range of 1–3 M. A hydrogen peroxide concentration in the range of 0.5–2 M has a significantly positive effect on the dissolution rate, while it has a negative effect in the range of 2–3 M. The temperature in the range of 61.5–70 °C has a negative effect due to the decomposition of hydrogen peroxide. The shrinking core model was applied to all parameters to obtain the dissolution kinetics. The dissolution process of silver was controlled by the surface reaction-controlled shrinking core model, i.e. 1-(1-X)1/3 = kst, with an activation energy of 28.80 kJ/mol. © 2024 The AuthorsKonya Teknik Üniversitesi, KTÜ
Pid-Based Pitch Angle Controller Scheme for Pmsg-Based Wind Energy Conversion System
No full textIn recent years, the Wind Energy Conversion System (WECS) usage has been increasing with rapid expansion. WECS has a highly nonlinear characteristic and the wind speed is also inherently uncertain. The output power regulation of the WECS at operating conditions above the rated wind speed is one of the subjects of intense interest to researchers. This regulation may be implemented through controlling the pitch angle of the wind turbine. This study presents a PID-based pitch angle controller scheme to realize this purpose in PMSG-based WECS. System configuration and whole controller designs are modeled and created in the simulation environment. In addition, a wind speed profile including both below-rated and above-rated wind speeds is created and the PID-based pitch angle controller is tested in detail for these conditions. According to the results obtained, it is seen that a good control dynamic can be obtained and power regulation is performed effectively by reaching and remaining the pitch angle to needed reference values at above-rated wind speeds. Moreover, with the designed scheme, appropriate adaptation to the changes in different operating points is provided. © 2024 IEEE
Portable Acceleration of Cms Computing Workflows With Coprocessors as a Service
No full textComputing demands for large scientific experiments, such as the CMS experiment at the CERN LHC, will increase dramatically in the next decades. To complement the future performance increases of software running on central processing units (CPUs), explorations of coprocessor usage in data processing hold great potential and interest. Coprocessors are a class of computer processors that supplement CPUs, often improving the execution of certain functions due to architectural design choices. We explore the approach of Services for Optimized Network Inference on Coprocessors (SONIC) and study the deployment of this as-a-service approach in large-scale data processing. In the studies, we take a data processing workflow of the CMS experiment and run the main workflow on CPUs, while offloading several machine learning (ML) inference tasks onto either remote or local coprocessors, specifically graphics processing units (GPUs). With experiments performed at Google Cloud, the Purdue Tier-2 computing center, and combinations of the two, we demonstrate the acceleration of these ML algorithms individually on coprocessors and the corresponding throughput improvement for the entire workflow. This approach can be easily generalized to different types of coprocessors and deployed on local CPUs without decreasing the throughput performance. We emphasize that the SONIC approach enables high coprocessor usage and enables the portability to run workflows on different types of coprocessors. © The Author(s) 2024.Council of Scientific and Industrial Research, India, CSIR; Ministry of Business, Innovation and Employment, MBIE; Ministry of Education and Science, MES; Benemérita Universidad Autónoma de Puebla, BUAP; Department of Atomic Energy, Government of India, DAE; PCTI; National Academy of Sciences of Ukraine, NASU; National Science and Technology Development Agency, สวทช; National Research Foundation of Korea, NRF; MSES; Ministerio de Educación, Cultura y Deporte, MECD; National Science Foundation, NSF; Institut National de Physique Nucléaire et de Physique des Particules, IN2P3; Latvijas Zinātnes Padome; Science and Technology Facilities Council, STFC; Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, CINVESTAV; Ministry of Science, ICT and Future Planning, MSIP; Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, FAPERJ; Ministerio de Ciencia e Innovación, MCIN; Universiti Malaya, UM; Ministry of Science and Technology, Taiwan, MOST; Hellenic Foundation for Research and Innovation, ΕΛ.ΙΔ.Ε.Κ; National Science Council, NSC; Ministry of Science,Technology and Research, MoSTR; Hispanics in Philanthropy, HIP; Instituto Nazionale di Fisica Nucleare, INFN; Secretaría de Educación Pública, SEP; Austrian Science Fund, FWF; Department of Science and Technology, Ministry of Science and Technology, India, DST; Chulalongkorn Academic; Consejo Nacional de Humanidades, Ciencias y Tecnologías, Conahcyt; Belgian Federal Science Policy Office, BELSPO; Centre National de la Recherche Scientifique, CNRS; Bundesministerium für Bildung und Forschung, BMBF; Fonds Wetenschappelijk Onderzoek, FWO; Türkiye Bilimsel ve Teknolojik Araştırma Kurumu, TÜBİTAK; Kavli Foundation; Helmholtz-Gemeinschaft, HGF; Commissariat à l'Énergie Atomique et aux Énergies Alternatives, CEA; Research Council of Finland, AKA; Conselho Nacional de Desenvolvimento Científico e Tecnológico, CNPq; Pakistan Atomic Energy Commission, PAEC; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, CAPES; Türkiye Enerji, Nükleer ve Maden Araştırma Kurumu, TENMAK; Ministry of Education - Singapore, MOE; European Commission, EC; Ministarstvo Prosvete, Nauke i Tehnološkog Razvoja, MPNTR; Science Foundation Ireland, SFI; U.S. Department of Energy, USDOE; International Council of Shopping Centers, ICSC; Fundação de Amparo à Pesquisa do Estado de São Paulo, FAPESP; Cosmetic Surgery Foundation, CSF; Agencia Estatal de Investigación, AEI; Programa Severo Ochoa del Principado de Asturias; General Secretariat for Research and Innovation, GSRI; Bulgarian National Science Fund, BNSF; Direktion für Entwicklung und Zusammenarbeit, DEZA; Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie, BMBWF; Ministerio de Ciencia, Tecnología e Innovación; Alfred P. Sloan Foundation, APSF; Maryland Ornithological Society, MOS; Chinese Academy of Sciences, CAS; Ministry of Higher Education, Science, Research and Innovation, Thailand, MHESRI; Fonds De La Recherche Scientifique - FNRS, FNRS; Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul, FAPERGS; Weston Havens Foundation; Institute for Research in Fundamental Sciences, IPM; Magyar Tudományos Akadémia, MTA; A.G. Leventis Foundation; Fonds pour la Formation à la Recherche dans l’Industrie et dans l’Agriculture, FRIA; Laboratorio Nacional de Supercómputo del Sureste de Mexico, LNS; European Regional Development Fund, ERDF; Nvidia; CERN; Louisiana Academy of Sciences, LAS; Ministerstvo Školství, Mládeže a Tělovýchovy, MŠMT; Secretaría de Educación Superior, Ciencia, Tecnología e Innovación, SENESCYT; Fundamental Research Funds for the Central Universities; Agentschap voor Innovatie door Wetenschap en Technologie, IWT; Universidad Autónoma de San Luis Potosí, UASLP; National Natural Science Foundation of China, NSFC; Beijing Municipal Science and Technology Commission, Adminitrative Commission of Zhongguancun Science Park, (Z191100007219010); Beijing Municipal Science and Technology Commission, Adminitrative Commission of Zhongguancun Science Park; Programa Estatal de Fomento de la Investigación Científica y Técnica de Excelencia María de Maeztu, (MDM-2017-0765); National Science, Research and Innovation Fund, (B37G660013); Excellence of Science, (30820817); Nemzeti Kutatási Fejlesztési és Innovációs Hivatal, NKFIH, (K 131991, K 133046, K 143477, K 124845, K 129058, K 143460, K 128713, K 124850, TKP2021-NKTA-64, 20202.2.1-ED-2021-00181, K 138136, K 128786); Nemzeti Kutatási Fejlesztési és Innovációs Hivatal, NKFIH; European Research Council, ERC, (TK202); European Research Council, ERC; Shota Rustaveli National Science Foundation, SRNSF, (FR-22-985); Shota Rustaveli National Science Foundation, SRNSF; Alexander von Humboldt-Stiftung, AvH, (22rl-037); Alexander von Humboldt-Stiftung, AvH; Deutsche Forschungsgemeinschaft, DFG, (400140256—GRK2497, 390833306); Deutsche Forschungsgemeinschaft, DFG; Horizon 2020, (724704, 758316, 824093, 765710, 752730, 675440); Horizon 2020; European Cooperation in Science and Technology, COST, (CA16108); European Cooperation in Science and Technology, COST; Welch Foundation, (C-1845); Welch Foundation; Narodowe Centrum Nauki, NCN, (2021/43/B/ST2/01552, 2021/41/B/ST2/01369); Narodowe Centrum Nauki, NCN; Ministry of Education and Science, (2022/WK/14); Ministry of Education and Science; Fundação para a Ciência e a Tecnologia, FCT, (CEECIND/01334/2018); Fundação para a Ciência e a Tecnologia, FCT; Qatar National Research Fund, QNRF, (MCIN/AEI/10.13039/501100011033); Qatar National Research Fund, QNR
Investigation of Positioning Accuracy Based on Received Signal Strength Indicator (rssi) Values From Radio Waves
No full textWith the development and civilian deployment of GPS by the United States, many countries around the world have realized the strategic and commercial importance of having positioning systems. This has led to the emergence and increased use of various satellite-based position systems operating on a regional or global scale. However, the biggest disadvantage of these systems is the need for a clear line of sight between the satellite and the user on the ground. In other words, when using these systems, an obstacle (tree, wall, etc.) between the satellites and the receiver on the ground can greatly hinder the use of the system. For this reason, alternative systems and methods are being developed to meet the need for positioning in underground and confined spaces. One of these methods is positioning with RSSI. In this study, experiments have been carried out to determine the location in confined spaces using RSSI value and machine learning algorithms and to examine and improve the accuracy of the obtained position. In the study, Trilateration and Fingerprint methods were used as positioning methods. Due to the biggest disadvantage of the trilateration method, which is the effect of the placement geometry of the reference points on the calculated position accuracy, the reference RF transmitter stations used in the study were placed in two different geometries in the experimental field. For both geometries, RSSI and coordinate values were measured at points spread homogeneously over the experimental field and training data were collected and data sets were created for machine learning based regression models. In addition, RSSI values were measured with 10 epoch measurements and filtered with statistical methods. In this way, RSSI measurements that may be inaccurate due to environmental effects were tried to be eliminated. The models trained with this training data were hyper parameter optimized with an algorithm developed to increase the fit to the training data. In this way, it is aimed to obtain the highest performance from each model used in the study. The models with the highest accuracy were selected from the models whose parameters were adjusted and added to the prepared mobile application and real-time tests were performed for both geometries in the application area. For the distance and cooridnate values obtained as a result of the tests, methods to increase the position accuracy obtained by balancing with the EKK method were investigated. As a result, the method that gives the highest position accuracy values is selected and the parameters of the system for positioning with RSSI are proposed for the application area. It is thought that the developed method can be successfully applied in real application areas since the selected study area is closed, has limited visibility and physical obstacles similar to real world conditions.ABD tarafından GPS'in geliştirilmesi ve sivil kullanıma açılması birlikte, dünya genelinde konum belirleme sistemlerine sahip olmanın stratejik ve ticari önemi birçok ülke tarafından fark edilmiştir. Bu durum, günümüzde bölgesel veya küresel çapta faaliyet gösteren çeşitli uydu tabanlı konum belirleme sistemlerinin ortaya çıkmasına ve kullanımının artmasına neden olmuştur. Fakat bu sistemlerinin en büyük dezavantajı konum belirlenirken uydu ile yeryüzünde bulunan kullanıcı arasında görüş açıklığı bulunması gerekliliğidir. Yani bu sistemler kullanılırken uydular ile yeryüzünde bulunan alıcı arasında bulunan bir engel (ağaç, duvar vs.) sistemin kullanımını büyük ölçüde engelleyebilmektedir. Bu nedenle, yeraltı ve kapalı alanlarda konum belirleme ihtiyacını karşılamak amacıyla alternatif sistemler ve yöntemler geliştirilmektedir. Bu yöntemlerden biri de RSSI değeri ile konum belirlemedir. Bu çalışma ile RSSI değeri ve makine öğrenmesi algoritmaları kullanılarak kapalı mekanlarda konum belirleme ve elde edilen konumun doğruluğunun incelenerek arttırılması yönünde araştırmalar yapılmıştır. Çalışmada konum belirleme yöntemi olarak Trilaterasyon ve Parmak İzi (ing:Fingerprint) yöntemleri kullanılmıştır. Trilaterasyon yönteminin en büyük dezavantajı olan referans noktaların yerleşim geometrisinin hesaplanan konum doğruluğuna etkisi sebebi ile çalışmada kullanılan referans RF verici istasyonları deney sahasına iki farklı geometride yerleştirilmiştir. Her iki geometri için deney sahasına homojen olarak yayılan noktalarda RSSI ve konum değerleri ölçülmüş olup makine öğrenmesi tabanlı regresyon modelleri için eğitim verisi toplanarak veri setleri oluşturulmuştur. Ayrıca toplanan bu eğitim verilerinde RSSI değerleri 10 epok ölçülerek istatiksel yöntemler ile filtrelemeye tabi tutulmuştur. Bu sayede çevresel etkilerden dolayı hatalı olabilecek RSSI ölçümleri elemine edilmeye çalışılmıştır. Bu eğitim verileri ile eğitilen modellerin geliştirilen bir algoritma ile hiper parametre optimizasyonu sağlanarak eğitim verisine uyumu arttırılmaya çalışılmıştır. Bu sayede çalışmada kullanılan her bir modelden en yüksek performansın alınması amaçlanmıştır. Parametreleri ayarlanan modellerden en yüksek doğruluğa sahip olanları seçilmiş ve hazırlanan mobil uygulamaya eklenerek uygulama sahasında her iki geometri için gerçek zamanlı testler yapılmıştır. Testler sonucunda elde edilen mesafe ve konum değerleri için EKK yöntemi ile dengeleme yapılarak elde edilen konum doğruluğunun arttırılmasına yönelik yöntemler araştırılmıştır. Sonuç olarak elde edilen en yüksek konum doğruluğu değerlerini veren yöntem seçilerek uygulama sahası için RSSI ile konum belirlemeye yönelik sistemin parametreleri önerilmiştir. Seçilen çalışma sahasının gerçek dünya koşullarına benzer şekilde kapalı, görüş açısının kısıtlı ve fiziki engeller içermesi sebebiyle, geliştirilen yöntemin gerçek uygulama alanlarında da başarıyla uygulanabileceği düşünülmektedir
X-ray Görüntülerinden Gizlenmiş Elektronik Devrelerin Derin Öğrenme Yöntemleriyle Tespiti
No full textThis thesis focuses on the detection of explosive circuits hidden in electronic devices, such as laptops, using X-ray imaging technologies, which are commonly used for security purposes in public institutions. The detection of prohibited items by security personnel through X-ray images may not be efficient due to issues such as time constraints, labor, and lack of expertise. Therefore, various methods have been developed to automate this process and eliminate human factors. Deep learning methods hold the potential to overcome these issues by automating image analysis and object recognition. In the first phase of the study, an original dataset was created for detecting explosive circuits, and classification was performed using deep learning models on this dataset. However, due to the small size and high complexity of the dataset, various problems arose during the testing phase. To address these issues, in the second phase, a method was developed to improve classification performance by combining feature maps from deep learning models. In particular, 11 different models based on Convolutional Neural Networks (CNN) were designed for classification purposes. The first challenge in the study was the small size of the dataset, which led to overfitting during testing, resulting in lower test performance compared to training performance. To mitigate this issue and improve performance, a classifier method that requires shorter training time and is resistant to overfitting was proposed. Thus, a feature fusion method with a Random Weight Network (RWN)-based classifier was introduced, and its performance was compared with the deep learning methods in the literature, as well as with a method where only the classifier was changed, for explosive detection. In the final section of the study, an analysis was conducted using RWN and feature fusion. This analysis aimed to evaluate the performance and effectiveness of each approach under different intermediate layer and hyperparameter values. As a result, the contribution of the developed feature fusion method to success was demonstrated through experimental results.Kamu kurum ve kuruluşlarında vazgeçilmez olan güvenlik terör vb. eylemlere karşı en yüksek seviyeye çıkarılmaya çalışılır. Bunun için güvenlik görevlileri tarafından kullanılan X-ray görüntüleri üzerinden yasaklı maddelerin görevliler tarafından tespiti, zaman, iş gücü ve uzman eksiklikleri nedeniyle verimli olmayabilir. Bu ve benzer sebeplerden ötürü ilgili süreci otomatikleştirmek ve insan faktörünü ortadan kaldırmak için çeşitli yöntemler geliştirilmiştir. Derin öğrenme yöntemleri ile görüntü analizi ve nesne tanıma ise bu sorunları aşma potansiyeline sahiptir. Tez çalışmasında, dizüstü bilgisayarlar gibi elektronik cihazlarda gizlenmiş patlayıcı devrelerinin tespiti konusu incelenmiştir. Bu problemi çözmek için hipotezler sunulmuş ve bu hipotezlere dayanarak derin öğrenme tabanlı çözüm yaklaşımları geliştirilmiştir. Problemi çözmek için ilk aşamada yeni ve özgün bir veri seti oluşturulmuş ve bu veri seti üzerinde sınıflandırma literatürdeki derin öğrenme modelleri kullanılarak gerçekleştirilmiştir. Yapılan deney çalışmaları sonuçları incelendiğinde veri seti boyutu ve karmaşıklığı nedeniyle çeşitli problemler ortaya çıktığı görülmüştür. İkinci aşamada elde edilen sınıflandırma başarılarının arttırılması amacıyla derin öğrenme modelleri özellik haritalarını birleştirmeye yönelik bir yaklaşım geliştirilmiştir. Bu yaklaşımın, özellikle boyutun küçük ve karmaşıklığının yüksek olduğu veri setlerinde başarım artırımı için yol gösterici nitelikte olduğu düşünülmektedir. Çalışmanın ilk aşamasında derin öğrenme alanında yaygın olarak kullanılan Evrişimsel Sinir Ağları tabanlı 11 farklı model sınıflandırma amacıyla tasarlanmıştır. Çalışmada yaşanan ilk zorluk veri setinin boyutunun küçük olması ve karmaşıklığı nedeniyle test esnasında yaşanmıştır. Test başarısı eğitim başarısına göre düşük olmuş ve yöntemler aşırı öğrenme yapmıştır. Bu durumu azaltmak ve başarımı artırmak için daha kısa süre eğitime ihtiyaç duyan ve aşırı öğrenmeye dirençli bir sınıflandırıcı yöntem kullanılması hedeflenmiştir. Böylece RWN (Random Weight Network) tabanlı bir sınıflandırıcı içeren özellik füzyonu yöntemi önerilmiş, yapılan literatürdeki derin öğrenme yöntemlerine ait deneyler ve yalnızca sınıflandırıcının değiştirildiği yöntem ile kıyaslanmış patlayıcı tespitinde kullanılmıştır. Çalışmanın son bölümünde RWN ve özellik füzyonu kullanılarak bir analiz yapılmıştır. Bu analiz, farklı ara katman ve hiper parametreler değerleri altında her bir yaklaşımın performansını ve etkinliğini değerlendirmeyi amaçlamıştır. Sonuç olarak geliştirilen özellik füzyonu yönteminin başarıya katkısı deney sonuçlarıyla gösterilmiştir