35 results on '"Karakaya, Fuat"'
Search Results
2. Construction of phase diagrams to estimate phase transitions at high pressures: A critical point at the solid liquid transition for benzene
- Author
-
Ibrahimoglu, Beycan, Uner, Deniz, Veziroglu, Ayfer, and Karakaya, Fuat
- Published
- 2021
- Full Text
- View/download PDF
3. FPGA implementation of neuro-fuzzy system with improved PSO learning
- Author
-
Karakuzu, Cihan, Karakaya, Fuat, and Çavuşlu, Mehmet Ali
- Published
- 2016
- Full Text
- View/download PDF
4. SIH: segmented intensity histogram for orientation estimation in image matching
- Author
-
Peker, Murat and Karakaya, Fuat
- Published
- 2016
- Full Text
- View/download PDF
5. Position of plasma in the phase diagram
- Author
-
Karakaya, Fuat, Ibrahimoglu, Beycan, and YILMAZOĞLU, MUSTAFA ZEKİ
- Abstract
The study of phase transitions and critical parameters has a history of nearly one century. The interest shown in this field of science has provided an accumulation of knowledge about phase transitions and critical states of matter through significant researches. On the other hand, theoretical and experimental studies still show that the pressure and temperature diagram of the material is not completed. In particular, the absence of plasma in the (p-T) phase diagram, the uncertainty of the pressure-dependent boundary range of the liquid phase and the state of the equilibrium curves are the focus of the studies. In the phase diagram of pure substance (p-T), the representation of only three phases and the boundary curves separating these phases with two basic points limits the phase diagram. However, as long as the plasma, which constitutes 98% of the universe and is accepted as the fourth phase of matter, does not take place in the phase diagram (p-T) and the liquid phase remains uncertain, phase transitions and critical parameters cannot be fully resolved.
- Published
- 2022
6. Neural identification of dynamic systems on FPGA with improved PSO learning
- Author
-
Cavuslu, Mehmet Ali, Karakuzu, Cihan, and Karakaya, Fuat
- Published
- 2012
- Full Text
- View/download PDF
7. Faz diyagramında plazmanın konumu
- Author
-
İBRAHİMOĞLU, Beycan, primary, YİLMAZOGLU, Zeki, additional, KARAKAYA, Fuat, additional, and İBRAHİMOĞLU, Beycan, additional
- Published
- 2021
- Full Text
- View/download PDF
8. Determination of the Real Critical Pressure and Critical Density of Substances
- Author
-
KARAKAYA, Fuat, primary, IBRAHIMOGLU, Beycan, additional, and IBRAHIMOGLU, Beycan Jr., additional
- Published
- 2021
- Full Text
- View/download PDF
9. Effect of the duty cycle on the spark-plug plasma synthetic jet actuator
- Author
-
Seyhan Mehmet, Akansu Yahya Erkan, Karakaya Fuat, Yesildag Cihan, and Akbıyık Hürrem
- Subjects
Physics ,QC1-999 - Abstract
A promising novel actuator called Spark-Plug Plasma Synthetic Jet (SPSJ) has been developed in Atmospheric Plasma Research Laboratory at Niğde University. It generates electrothermally high synthetic jet velocity by using high voltage. SPSJ actuator can be utilized to be an active flow control device having some advantages such as no moving parts, low energy consumption and easy to integrate the system. This actuator consists of two main components: semi-surface spark plug (NGK BUHW) as an anode electrode and a cap having an orifice as a cathode electrode. The cap, having a jet exit orifice diameter of 2 mm, has diameter of 4.4 mm and height of 4.65 mm. This study presents the characteristics of SPSJ actuator by using the hot wire anemometer in order to approximately determine jet velocity in quiescent air. Peak velocity as high as 180 m/s was obtained for fe= 100 and duty cycle 50%. The flow visualization indicated that the actuator’s jet velocity is enough to penetrate the developed boundary layer.
- Published
- 2016
- Full Text
- View/download PDF
10. Sterilization and plasma processing of room temperature surfaces with a one atmosphere uniform glow discharge plasma (OAUGDP)
- Author
-
Gadri, Rami Ben, Roth, J.Reece, Montie, Thomas C., Kelly-Wintenberg, Kimberly, Tsai, Peter P.-Y., Helfritch, Dennis J., Feldman, Paul, Sherman, Daniel M., Karakaya, Fuat, and Chen, Zhiyu
- Published
- 2000
- Full Text
- View/download PDF
11. Air Filter Sterilization Using a One Atmosphere Uniform Glow Discharge Plasma (the Volfilter)
- Author
-
Kelly-Wintenberg, Kimberly, Sherman, Daniel M., Tsai, Peter P.-Y., Gadri, Rami Ben, Karakaya, Fuat, Chen, Zhiyu, Roth, J. Reece, and Montie, Thomas C.
- Subjects
Plasma (Ionized gases) -- Usage ,Air -- Purification ,Decontamination (from gases, chemicals, etc.) -- Methods ,Sick building syndrome -- Prevention ,Sanitary microbiology -- Research ,Business ,Chemistry ,Electronics ,Electronics and electrical industries - Abstract
Two characteristics of microorganisms--extremely small size (0.01 to a few micrometers) and the ability to reproduce--hinder the effective filtration of bacteria and viruses from indoor air. The microorganisms captured by a filter in spite of their small size can reproduce in situ and be released into the airstream, giving rise to the 'sick building syndrome.' The application of the One Atmosphere Uniform Glow Discharge Plasma (OAUGDP) to a filter can address both these issues. At University of Tennessee at Knoxville, we have recently developed the 'Volfilter,' a planar version of the OAUGDP produced by attaching strip electrodes to both sides of a sheet of dielectric filter material and energizing the electrodes with a high-voltage, low-frequency RF source. After the filter material removes microorganisms from the airstream, the OAUGD plasma kills the captured microorganisms. The combination of an appropriate filter material and periodic application of the OAUGDP results in an effective capture and sterilization device even for the smallest microorganisms and requires minimum maintenance. This paper will describe results obtained during the operation of a laboratory-scale 'Volfilter' challenged by two kinds of microorganisms, S. aureus and the bacterial virus Phi X 174. An objective of this work is to demonstrate that a 'Volfilter' exposed to a OAUGDP will have the number of captured microorganisms on its surface reduced by a factor of one million. Index Terms--Air filtration, atmospheric glow discharge, atmospheric plasma, OAUGDP, plasma sterilization, sick building syndrome, sterilization.
- Published
- 2000
12. A Remote Exposure Reactor (RER) for Plasma Processing and Sterilization by Plasma Active Species at One Atmosphere
- Author
-
Roth, J. Reece, Sherman, Daniel M., Gadri, Rami Ben, Karakaya, Fuat, Chen, Zhiyu, Montie, Thomas C., Kelly-Wintenberg, Kimberly, and Tsai, Peter P.-Y.
- Subjects
Plasma (Ionized gases) -- Usage ,Sterilization -- Methods ,Decontamination (from gases, chemicals, etc.) -- Methods ,Business ,Chemistry ,Electronics ,Electronics and electrical industries - Abstract
We have developed a remote exposure reactor (RER) in which the active species of air and other gases responsible for sterilization and processing effects are generated on flat panels in a surface layer of one atmosphere uniform glow discharge plasma (OAUGDP). These active species are convected by forced airflow at one atmosphere and near room temperature to a remote exposure chamber in which the workpiece is located. This allows workpieces of any size or shape to be sterilized or processed without direct contact with the plasma. Here, we report operation of the RER as a sterilizer with both single-pass and recirculating active species flow through the remote exposure chamber. We used the RER to reduce the numbers of two genera of microorganisms (Esherichia coli and Staphylococcus aureus) on test samples of polypropylene fabric. When the recirculating airflow configuration was employed, the population both of E. coli and S. aureus cells was reduced by at least five decades after only 25 s of exposure. Tests in the single pass airflow configuration produced similar results, with the E. coli and S. aureus populations decreased by at least four decades after 25 s of exposure. Index Terms--Atmospheric plasma, OAUGDP, plasma processing, remote exposure, sterilization.
- Published
- 2000
13. Implementation of edge dependent interpolation based de-interlacer on FPGA
- Author
-
Cavuslu, Mehmet Ali, primary, Karakaya, Fuat, additional, and Balkoca, Alisan, additional
- Published
- 2016
- Full Text
- View/download PDF
14. Hardware implementation of a scale and rotation invariant object detection algorithm on FPGA for real-time applications
- Author
-
PEKER, Murat, primary, ALTUN, Halis, additional, and KARAKAYA, Fuat, additional
- Published
- 2016
- Full Text
- View/download PDF
15. Hardware emulation of HOG and AMDF based scale and rotation invariant robust shape detection
- Author
-
Peker, Murat, primary, Altun, Halis, additional, and Karakaya, Fuat, additional
- Published
- 2012
- Full Text
- View/download PDF
16. Hardware implementation of Discrete Wavelet Transform and Inverse Discrete Wavelet Transform on FPGA
- Author
-
Cavuslu, Mehmet Ali, primary and Karakaya, Fuat, additional
- Published
- 2010
- Full Text
- View/download PDF
17. Implementation of HOG algorithm for real time object recognition applications on FPGA based embedded system
- Author
-
Karakaya, Fuat, primary, Altun, Halis, additional, and Cavuslu, Mehmet Ali, additional
- Published
- 2009
- Full Text
- View/download PDF
18. The effect of genetic algorithm parameters on the solution of plate location detection
- Author
-
Peker, Murat, primary, Altun, Halis, additional, and Karakaya, Fuat, additional
- Published
- 2008
- Full Text
- View/download PDF
19. Increasing the Surface Energy and Sterilization of Nonwoven Fabrics by Exposure to a One Atmosphere Uniform Glow Discharge Plasma (OAUGDP)*
- Author
-
Roth, J. Reece, primary, Chen, Zhiyu, additional, Sherman, Daniel M., additional, Karakaya, Fuat, additional, Tsai, Peter P.-Y., additional, Kelly-Wintenberg, Kimberly, additional, and Montie, Thomas C., additional
- Published
- 2001
- Full Text
- View/download PDF
20. Faz diyagramında plazmanın konumu.
- Author
-
İbrahimoğlu, Beycan, Yılmazoğlu, Mustafa Zeki, and Karakaya, Fuat
- Subjects
- *
PHASE diagrams , *PHASE transitions , *PHASE equilibrium , *PHASES of matter , *CRITICAL point (Thermodynamics) - Abstract
The study of phase transitions and critical parameters has a history of nearly one century. The interest shown in this field of science has provided an accumulation of knowledge about phase transitions and critical states of matter through significant researches. On the other hand, theoretical and experimental studies still show that the pressure and temperature diagram of the material is not completed. In particular, the absence of plasma in the (p-T) phase diagram, the uncertainty of the pressure-dependent boundary range of the liquid phase and the state of the equilibrium curves are the focus of the studies. In the phase diagram of pure substance (p-T), the representation of only three phases and the boundary curves separating these phases with two basic points limits the phase diagram. However, as long as the plasma, which constitutes 98% of the universe and is accepted as the fourth phase of matter, does not take place in the phase diagram (p-T) and the liquid phase remains uncertain, phase transitions and critical parameters cannot be fully resolved. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
21. Automated Exploration of the ASIC Design Space for Minimum Power-Delay-Area Product at the Register Transfer Level
- Author
-
Karakaya, Fuat
- Subjects
- Electrical and Computer Engineering
- Abstract
Exploring the integrated circuit design space for minimum power-delay-area (PDA) product can be time-consuming and tedious, especially when the target standard-cell library has hundreds of options. In this dissertation, heuristic algorithms that automate this process have been developed, implemented and validated at the reg- ister transfer level. In some cases, the PDA product was 1.9 times better than the initial baseline solution. The parallel search algorithm exhibited 9x speed up when executed on 10 machines simultaneously. These two new methods also characterize the design space for the given RTL code by generating power-delay-area points in addition to the minimum PDA point in case the designer wishes to select a different solution that is a tradeoff among these metrics. As a final step, these two search algorithms are integrated into a fully automated ASIC design flow.
- Published
- 2004
22. SIH: segmented intensity histogram for orientation estimation in image matching
- Author
-
Fuat Karakaya, Murat Peker, 0-Belirlenecek, peker, murat -- 0000-0001-9877-5493, and [Peker, Murat -- Karakaya, Fuat] Nigde Univ, Elect & Elect Engn Fac, Nigde, Turkey
- Subjects
Computation ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Scale-invariant feature transform ,02 engineering and technology ,Image (mathematics) ,SIFT ,Keypoint matching ,Histogram ,0202 electrical engineering, electronic engineering, information engineering ,Computer vision ,Electrical and Electronic Engineering ,Orientation estimation ,Dominant orientation ,Mathematics ,Image matching ,business.industry ,Orientation (computer vision) ,Histogram matching ,020206 networking & telecommunications ,Pattern recognition ,Local features ,Rotation invariance ,Computer Science::Computer Vision and Pattern Recognition ,Signal Processing ,020201 artificial intelligence & image processing ,Artificial intelligence ,business ,Intensity (heat transfer) - Abstract
WOS: 000382363600019, In this paper, we propose a fast and effective new method to reduce the overhead cost of orientation estimation. The proposed method uses the summation of intensity values from segments of image patches and forms a histogram based on those values. As a result, it is faster than SIFT-like algorithms because it does not require computation of gradient orientations and magnitudes. Also, it is as fast as other intensity-based algorithms with better image matching performance. Proposed method could be easily integrated to any image matching algorithms. Test results indicate that SIFT integrated with proposed orientation estimation method produces comparable results as the original multi-angle SIFT algorithm with less execution time.
- Published
- 2016
23. Hardware implementation of a scale and rotation invariant object detection algorithm on FPGA for real-time applications
- Author
-
Fuat Karakaya, Halis Altun, Murat Peker, 0-Belirlenecek, altun, halis -- 0000-0002-2126-8757, peker, murat -- 0000-0001-9877-5493, and [Peker, Murat -- Karakaya, Fuat] Nigde Univ, Dept Elect & Elect Engn, Nigde, Turkey -- [Altun, Halis] KTO Karatay Univ, Dept Elect & Elect Engn, Fac Engn, Karatay, Konya, Turkey
- Subjects
General Computer Science ,Computer science ,Image processing ,02 engineering and technology ,Average magnitude difference function,field programmable gate arrays,histogram of oriented gradients,image processing ,histogram of oriented gradients ,Difference function ,0202 electrical engineering, electronic engineering, information engineering ,Computer vision ,Electrical and Electronic Engineering ,Invariant (mathematics) ,MATLAB ,Field-programmable gate array ,computer.programming_language ,field programmable gate arrays ,business.industry ,020206 networking & telecommunications ,Object detection ,image processing ,Histogram of oriented gradients ,020201 artificial intelligence & image processing ,Artificial intelligence ,business ,Average magnitude difference function ,computer ,Computer hardware - Abstract
WOS: 000378097800077, A hardware implementation of a computationally light, scale, and rotation invariant method for shape detection on FPGA is devised. The method is based on histogram of oriented gradients (HOG) and average magnitude difference function (AMDF). AMDF is used as a decision module that measures the similarity/dissimilarity between HOG vectors of an image in order to classify the object. In addition, a simulation environment implemented on MATLAB is developed in order to overcome the time-consuming and tedious process of hardware verification on the FPGA platform. The simulation environment provides specific tools to quickly implement the proposed methods. It is shown that the simulator is able to produce exactly the same results as those obtained from FPGA implementation. The results indicate that the proposed approach leads to a shape detection method that is computationally light, scale, and rotation invariant, and, therefore, suitable for real-time industrial and robotic vision applications.
- Published
- 2016
24. Implementation of edge dependent interpolation based de-interlacer on FPGA
- Author
-
Fuat Karakaya, Alisan Balkoca, Mehmet Ali Cavuslu, [Cavuslu, Mehmet Ali] KOC Bilgi & Savunma Teknol AS, Ankara, Turkey -- [Karakaya, Fuat] Nigde Univ, Elekt Elekt Muhendsligi Bolumu, Nigde, Turkey -- [Balkoca, Alisan] Kagitsokagi ArGe, Ankara, Turkey, and 0-Belirlenecek
- Subjects
De-interlacer ,business.industry ,Computer science ,Reading (computer) ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Interlacing ,020207 software engineering ,02 engineering and technology ,High-definition video ,VHDL ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,Point (geometry) ,Computer vision ,Artificial intelligence ,Enhanced Data Rates for GSM Evolution ,business ,Field-programmable gate array ,computer ,FPGA ,Computer hardware ,Interpolation ,computer.programming_language - Abstract
24th Signal Processing and Communication Application Conference (SIU) -- MAY 16-19, 2016 -- Zonguldak, TURKEY, WOS: 000391250900025, Interlacing technique aims to lower the costs of high definition video systems by reducing the data amount sent to receiver unit. Regeneration of image at the receiver unit is an important point of interlacing method. In this study, regeneration (de-interlacing) of frames that are sent to receiver unit is implemented by using edge dependent interpolation method. The method is implemented using VHDL on Altera Cyclone-II FPGA. The method avoids reading of redundant data which yields to reduced operation time. Implementation occupies only %3 of the FPGA that is used in this study., IEEE, Bulent Ecevit Univ, Dept Elect & Elect Engn, Bulent Ecevit Univ, Dept Biomed Engn, Bulent Ecevit Univ, Dept Comp Engn
- Published
- 2016
25. A new approach to visual navigation systems for autonomous vehicles
- Author
-
Peker, Murat, Karakaya, Fuat, Elektrik-Elektronik Mühendisliği Ana Bilim Dalı, Fen Bilimleri Enstitüsü, and Peker, Murat
- Subjects
Görsel Seyir Sistemi ,Yer İzi Tanımlayıcıları ,Elektrik ve Elektronik Mühendisliği ,İmage Matching ,Visual Navigation System ,Öznitelik Eşleştirme ,Digital image processing ,Görüntü Eşleştirme ,Feature Matching ,Electrical and Electronics Engineering ,Landmark Descriptors - Abstract
Bu tez çalışmasında, görsel algılayıcılar ile elde edilen görüntülerden, özgün algoritmalar yardımıyla çıkarılan yer izleri ve tanımlayıcılar kullanılarak, insansız araçlar için gürbüz görsel seyir sistemi gerçekleştirilmiştir. Görsel yer izi tabanlı konum değişimi hesaplama yöntemleri, bir görüntü akışı içerisindeki görüntülerde tekrarlanan yer izlerini tespit ederek konum değişimini bulmaya yarar. Tez kapsamında yeni bir yer izi tespit algoritması önerilmiştir. Önerilen yer izi tespit algoritması literatürdeki veri kümeleri üzerinde test edilmiş ve yaygın olarak kullanılan yer izi bulma algoritmasına (SIFT) göre daha fazla yer izi tespit ettiği benzetimlerle gösterilmiştir. İkinci olarak bu tez çalışmasında, yönelim açısı hesabı gerektiren yer izi tanımlayıcı algoritmalarında kullanılabilecek başarımı yüksek ve hızlı bir yönelim açısı hesaplama algoritması önerilmiştir. Önerilen yönelim açısı hesaplama algoritmasının, çeşitli görüntü bozulmalarında literatürde mevcut diğer algoritmalara kıyasla daha az etkilendiği veri kümeleri üzerinde gösterilmiştir. Üçüncü olarak yönelim açısı hesabını ortadan kaldıran bir tanımlayıcı eşleştirme algoritması ve bu algoritmada kullanılabilecek bir tanımlayıcı deseni önerilmiştir. Son olarak bu çalışma kapsamında önerilen algoritmalar ve Kalman filtresi kullanılarak gerçekleştirilen görsel seyir sistemi iki farklı veri kümesi üzerinde test edilerek elde edilen sonuçlar incelenmiştir., This thesis presents a robust visual navigation system for autonomous unmanned vehicles. For this purpose novel algorithms are proposed to extract features and descriptors from images captured by visual sensors. Visual landmark-based localization methods detect repeating landmarks from images in a video stream to determine location change. In this thesis a new landmark detection algorithm is proposed. The proposed landmark detection algorithm was tested on commonly used benchmark problems and it was shown that the proposed algorithm could detect more landmarks than algorithm (SIFT) well known in the literature. As a second contribution of the thesis, a fast and effective new method is proposed to reduce the overhead cost of orientation estimation and it could be easily integrated to any image matching algorithms which require orientation estimation. It was shown that the proposed orientation estimation algorithm performs better immunity to some of the image distortion types than other algorithms available in the literature. Also, a new feature matching algorithm, aimed to eliminate the need for orientation estimation, and a descriptor pattern, used with the algorithm, is proposed. Finally, the visual navigation system, utilizing the proposed algorithms and Kalman filter, was tested on two different benchmark data sets and the results are presented.
- Published
- 2016
26. Effect of the duty cycle on the spark-plug plasma synthetic jet actuator
- Author
-
Hürrem Akbıyık, Yahya Erkan Akansu, Mehmet Seyhan, Fuat Karakaya, Cihan Yesildag, Dancova, P, Vesely, M, [Seyhan, Mehmet] Karadeniz Tech Univ, Mech Engn Dept, TR-61080 Trabzon, Turkey -- [Akansu, Yahya Erkan -- Yesildag, Cihan] Nigde Univ, Mech Engn Dept, TR-51200 Nigde, Turkey -- [Karakaya, Fuat] Nigde Univ, Elect & Elect Engn Dept, TR-51200 Nigde, Turkey -- [Akbiyik, Hurrem] Cukurova Univ, Mech Engn Dept, TR-01380 Adana, Turkey, Çukurova Üniversitesi, and 0-Belirlenecek
- Subjects
Flow visualization ,Physics ,Jet (fluid) ,010308 nuclear & particles physics ,QC1-999 ,Acoustics ,Mechanical engineering ,Atmospheric-pressure plasma ,01 natural sciences ,7. Clean energy ,0-Belirlenecek ,law.invention ,Duty cycle ,law ,0103 physical sciences ,Synthetic jet ,010306 general physics ,Actuator ,Spark plug ,Body orifice - Abstract
10th Anniversary International Conference on Experimental Fluid Mechanics -- NOV 17-20, 2015 -- Prague, CZECH REPUBLIC, WOS: 000400395300106, A promising novel actuator called Spark-Plug Plasma Synthetic Jet (SPSJ) has been developed in Atmospheric Plasma Research Laboratory at Nigde University. It generates electrothermally high synthetic jet velocity by using high voltage. SPSJ actuator can be utilized to be an active flow control device having some advantages such as no moving parts, low energy consumption and easy to integrate the system. This actuator consists of two main components: semi-surface spark plug (NGK BUHW) as an anode electrode and a cap having an orifice as a cathode electrode. The cap, having a jet exit orifice diameter of 2 mm, has diameter of 4.4 mm and height of 4.65 mm. This study presents the characteristics of SPSJ actuator by using the hot wire anemometer in order to approximately determine jet velocity in quiescent air. Peak velocity as high as 180 m/s was obtained for f(e) = 100 and duty cycle 50%. The flow visualization indicated that the actuator's jet velocity is enough to penetrate the developed boundary layer., DANTEC Dynam GmbH, LAVISION, LENAM, MECAS ESI s r o, MIT s r o, TSI GmbH, Scientific and Technological Research Council of Turkey (TUBITAK) [213M179], The authors would like to acknowledge the financial support of this work by the Scientific and Technological Research Council of Turkey (TUBITAK) under the Contract Number of 213M179.
- Published
- 2016
27. Neural identification of dynamic systems on FPGA with improved PSO learning
- Author
-
Cihan Karakuzu, Mehmet Ali Cavuslu, Fuat Karakaya, 0-Belirlenecek, Karakuzu, Cihan -- 0000-0003-0569-098X, and [Karakaya, Fuat] Nigde Univ, Fac Engn, Dept Elect & Elect Engn, Nigde, Turkey -- [Cavuslu, Mehmet Ali] Koc Information & Def Technol Inc, METU Technopolis, ODTU Teknokent, TR-06800 Ankara, Turkey -- [Karakuzu, Cihan] Bilecik Seyh Edebali Univ, Fac Engn, Dept Comp Engn, TR-11210 Bilecik, Turkey
- Subjects
Artificial neural network ,business.industry ,Computer science ,Computer Science::Neural and Evolutionary Computation ,System identification ,Particle swarm optimization ,Function (mathematics) ,Dynamical system ,Artificial neural networks (ANN) ,Term (time) ,Computer Science::Hardware Architecture ,Identification (information) ,Computer engineering ,Particle swarm optimization (PSO) ,Artificial intelligence ,business ,Field-programmable gate array ,FPGA ,Software - Abstract
WOS: 000306107900001, This work introduces hardware implementation of artificial neural networks (ANNs) with learning ability on field programmable gate array (FPGA) for dynamic system identification. The learning phase is accomplished by using the improved particle swarm optimization (PSO). The improved PSO is obtained by modifying the velocity update function. Adding an extra term to the velocity update function reduced the possibility of stucking in a local minimum. The results indicates that ANN, trained using improved PSO algorithm, converges faster and produces more accurate results with a little extra hardware utilization cost. (C) 2012 Elsevier B.V. All rights reserved.
- Published
- 2012
28. FPGA implementation of neuro-fuzzy system with improved PSO learning
- Author
-
Mehmet Ali Cavuslu, Fuat Karakaya, Cihan Karakuzu, 0-Belirlenecek, Karakuzu, Cihan -- 0000-0003-0569-098X, and [Karakuzu, Cihan] Bilecik Seyh Edebali Univ, Fac Engn, Dept Comp Engn, Gulumbe Campus, TR-11210 Bilecik, Turkey -- [Karakaya, Fuat] Nigde Univ, Fac Engn, Dept Elect & Elect Engn, Cent Campus, TR-51240 Nigde, Turkey -- [Cavuslu, Mehmet Ali] Koc Informat & Def Technol Inc, METU Technopolis, ODTU Teknokent, TR-06800 Ankara, Turkey
- Subjects
Neuro-fuzzy ,Computer science ,Cognitive Neuroscience ,Real-time computing ,02 engineering and technology ,Neuro-fuzzy network ,Machine Learning ,Fuzzy Logic ,Artificial Intelligence ,VHDL ,0202 electrical engineering, electronic engineering, information engineering ,Field-programmable gate array ,System identification ,Metaheuristic ,computer.programming_language ,020208 electrical & electronic engineering ,Particle swarm optimization ,Computer engineering ,FPGA implementation ,Benchmark (computing) ,020201 artificial intelligence & image processing ,computer ,Membership function ,Algorithms ,Metaheuristic learning - Abstract
WOS: 000376234300011, PubMed ID: 27136666, This paper presents the first hardware implementation of neuro-fuzzy system (NFS) with its metaheuristic learning ability on field programmable gate array (FPGA). Metaheuristic learning of NFS for all of its parameters is accomplished by using the improved particle swarm optimization (iPSO). As a second novelty, a new functional approach, which does not require any memory and multiplier usage, is proposed for the Gaussian membership functions of NFS. NFS and its learning using iPSO are implemented on Xilinx Virtex5 xc5vlx110-3ff1153 and efficiency of the proposed implementation tested on two dynamic system identification problems and licence plate detection problem as a practical application. Results indicate that proposed NFS implementation and membership function approximation is as effective as the other approaches available in the literature but requires less hardware resources. (C) 2016 Elsevier Ltd. All rights reserved.
- Published
- 2015
29. Doğrudan metanol yakıt pillerinin elektronik kontrolü
- Author
-
Bozay, Ali, Altuncu, Yasemin, Karakaya, Fuat, Elektrik-Elektronik Mühendisliği Ana Bilim Dalı, and Bozay, Ali
- Subjects
Microprocessor ,Mikrodenetleyici ,Direct Methanol Fuel Cell (DMFC) ,Elektrik ve Elektronik Mühendisliği ,Electronic Card ,Elektronik Kontrol ,Doğrudan Metanol Yakıt Pili (DMYP) ,Electrical and Electronics Engineering - Abstract
Bu tez çalışmasında doğrudan metanol yakıt pilinin (DMYP) çalışabilmesi için gerekli parametreler belirlenmiş, elektronik kontrol kartı tasarlanmış ve gerçekleştirilmiştir. Gerçekleştirilen kontrol kartında merkezi işlem birimi olarak PIC 16F877A mikrodenetleyicisi kullanılmış ve Assembly dili ile kontrol programı yazılmıştır. Yakıt pili sisteminden 3 Volt 30 Watt çıkış verebilen 10 hücreli doğrudan metanol yakıt pili hücre grubu kullanılmış, DC/DC çevirici ile çıkış gerilimi 12 Volta yükseltilmiştir. Gerçekleştirilen doğrudan metanol yakıt pilinin verimini etkileyen parametreler farklı durumlar için incelenmiştir. Yapılan deneyler ile doğrudan metanol yakıt pilinin (DMYP) yüksek verimle çalışması için parametre değerlerinin ne olması gerektiği ayrı ayrı belirlenmiştir., In this thesis, the desired parameters for working of Direct Methanol Fuel Cell (DMFC) were determined and also electronic control card of the DMFC was designed. In order to control DMFC, PIC 16F877A microprocessor was used as a central process unit and desired program was written by using Assembly programming language. The fuel cell which has 10 direct methanol cell and 3 Volt 30 Watt output was used and then the output voltage was increased to 12 Volt by using DC to DC converter. The efficiency parameters for designed DMFC were investigated for different cases. Thus, the best working parameter values for DMFC‟s high efficiency operation were determined one by one.
- Published
- 2014
30. Hardware implementaton of artificial neural network training using gradient based and global search algorithms on FPGA
- Author
-
Çavuşlu, Mehmet Ali, Karakaya, Fuat, Elektrik-Elektronik Mühendisliği Ana Bilim Dalı, Çavuşlu, Mehmet Ali, and Fen Bilimleri Enstitüsü
- Subjects
Yapay Sinir Hücresi ,Geriye Yayılım Algoritması ,Elektrik ve Elektronik Mühendisliği ,Yapay Arı Koloni Algoritması ,Levenberg & Marquardt Algorithm ,Artificial Bee Colony Algorithm ,Particle Swarm Optimization Algorithm ,Kayan Noktalı Sayı ,Back Propagation Algorithm ,Yapay Sinir Ağları ,Levenberg & Marquardt Algoritması ,Artificial Neural Cell ,Floating Point Number ,Parçacık Sürü Optimizasyon Algoritması ,FPGA ,Artificial Neural Networks ,Electrical and Electronics Engineering - Abstract
Yapay sinir ağları (YSA), sistem giriş ve çıkışları arasındaki karmaşık ilişkiyi etkili bir şekilde modelleyebilme yeteneği ile bilimsel çalışmalarda yaygın olarak kullanılmaktadır. Güncel çalışmalarda YSA eğitiminde gradyen tabanlı algoritmaların yanında global arama özelliğine sahip algoritmalar da kullanılmaktadır. Bu çalışmada, gradyen tabanlı algoritmalardan geriye yayılım (GY) ve Levenberg & Marquardt (LM) algoritmaları ile sezgisel arama özelliğine sahip algoritmalardan parçacık sürü optimizasyon (PSO) ve yapay arı koloni (YAK) algoritmaları kullanılarak YSA eğitimi FPGA üzerinde donanımsal olarak gerçeklenmiştir. Gerçeklemelerde sağlamış olduğu dinamiklik ve hassasiyetten ötürü IEEE 754 kayan noktalı sayı formatı kullanılmıştır. FPGA üzerinde YSA gerçeklemesinde en kritik aşama olan aktivasyon fonksiyonunun gerçeklenmesinde matematiksel yaklaşımlar tercih edilmiştir. Donanımsal gerçeklemeler dinamik sistem tanıma ve araç plaka bölgesi belirleme problemleri kullanılarak test edilmiştir. Eğitilen YSA'lar eğitim fazında ağa gösterilmeyen girişçıkış örnekleri ile test edilmiş ve her örnek için yukarıda bahsi geçen algoritmaların YSA eğitimindeki başarım oranları kıyaslamalı olarak verilmiştir., Artificial neural networks (ANNs) are commonly used in scientific studies due to their ability to effectively model complex relationship between input and output of a system. In recent studies global search algorithms are also utilized in addition to gradient based algorithms in ANN training. In this study, hardware implementation of ANN training on FPGA is realized using gradient based algorithms such as Back Propagation (BP) and Levenberg&Marquardt, and heuristic algorithms such as Particle Swarm Optimization (PSO) and Artificial Bee Colony (ABC). For implementation, floating point numbers are chosen as number format due to its dynamism and accuracy. Mathematical approachesare preferred for hardware implementation of activation functions, which is the most critical stage of ANN implementation on FPGA. The hardware implementations on FPGA are tested using dynamic system identification and license plate recognition problems. The trained ANNs are tested using input-output data sets which are not used in training and results for each example are given in a comparative manner.
- Published
- 2013
31. Gerçek zamanlı uygulamalar için kalman süzgeci tabanlı takip algoritmalarının FPGA üzerinde etkin gerçekleştirilmesi
- Author
-
Kösten, Mehmet Muzaffer, Karakaya, Fuat, and Fen Bilimleri Enstitüsü
- Subjects
Mapping ,Kalman Süzgeci ,Genişletilmiş Kalman Süzgeci ,Extended Kalman Filter ,Konumlama ,Eş Zamanlı Haritalandırma ,Simultaneous Localization ,Kalman Filter ,FPGA - Abstract
Teknolojinin gelişmesi gündelik hayatta kullandığımız çoğu cihaz daha akıllı ve kullanıcı dostu haline gelmiştir. Bu gelişimin arka planına bakıldığında gelişmiş kontrol yöntemlerinin ve algoritmaların kullanılmaya başladığı görülmektedir. Özellikle bir sistemin durumunu kontrol etmek ve gözlemlemek gibi işlemlerde yoğun miktarda veri toplanmakta ve bu verilerin anlamlı bir şekilde takip edilerek yorumlanması gerekir. Fiziksel dünyadan alınan verilerin gerek ortam gürültüsü gerekse ölçüm yöntemlerinden kaynaklanan hatalar içermesi nedeni ile süzülmesi gerekmektedir. Bu gözleme ve süzme işlemi için pek çok yöntem geliştirilmiş olmakla birlikte günümüzde çoğu uygulama için Kalman Süzgeci ve onun genişletilmiş hali olan Genişletilmiş kalman süzgeci kullanılmaktadır. Bu çalışmada da bu yöntem FPGA üzerine Kayan Noktalı sayılar kullanılarak uygulanmış ve iki örnek problem üzerinde uyarlanarak FPGA üzerinde gerçekleştirilmiştir. İlk uygulama olarak eğik atış problemi seçilmiş olup, ikinci uygulama olarak da Eş Zamanlı Haritalandırma ve Konumlama probleminde kullanılan Genişletilmiş Kalman Süzgeci tercih edilmiştir. Tasarlanan yapılar Matlab üzerinde denenmiş ve daha sonra Xilinx ISE programında sentezlenmiştir. Elde edilen sonuçlar bilgisayar üzerinde MATLAB kullanılarak değerlendirilmiştir., Thanks to the developments in the technology, many devices that we use in our daily life have become more useful. When we look at the backgrounds of these developments, we can see the usage of advanced control systems and algortihms. Especially, in the process of observing and controlling of a system huge amount of data may be collected and this data should be deciphered in a logical way. Because of the inconvenience of knowledge which is collected from the real world, both the noise of the environment and the errors which are caused by methods of measurement should be filtered. For this observing and filtering process, many different methods have been developed but the mostly used ones are Kalman Filtering and it's developed version Extended Kalman Filtering. And, in this study these two methods have been applied by using Floating Point Numbers on FPGA and implemented on it with the help of two example case. For the first implementation Projectile Motion Problem has been chosen and for the second implementation, Extended Kalman Filtering which is used in Simultaneous Localization and Mapping Problem has been applied. Designed structures have been tested on Matlab and then synthesized on Xilinx ISE Program. Obtained results have been evaluated with computer on Matlab.
- Published
- 2013
32. Investigating the effects of plasma actuator on the flow control around NACA2415 airfoil
- Author
-
Şanlisoy, Aytaç, Akansu, Yahya Erhan, Karakaya, Fuat, Şanlısoy, Aytaç, Akansu, Yahya Erkan, and Makine Mühendisliği Ana Bilim Dalı
- Subjects
Uçak Kanadı ,Bilim ve Teknoloji ,Mechanical Engineering ,Plazma Aktüatör ,Akış Kontrol ,Peristaltik Uyarım ,Makine Mühendisliği ,Uçak Mühendisliği ,Aircraft Engineering ,Science and Technology - Abstract
Bu çalışmada, aerodinamik akış kontrolünde paraelektrik ve peristaltik sinyal sürümünün plazma aktüatörün performansına etkisi araştırılmıştır. Sinyal modülasyonları ile uyarılan, NACA2415 model uçak kanadı üzerine yerleştirilmiş plazma aktüatörün aktif ve pasif olduğu durum için Reynolds sayılarının 8000-90000 aralığındaki etkileri rüzgâr tünelinde incelenmiştir. Uçak kanadına etki eden aerodinamik kuvvet katsayıları ölçülmüş, iz bölgesine yerleştirilen kızgın tel probu ile hız ölçümleri alınmış ve duman-tel yöntemi kullanılarak uçak kanadı etrafındaki akış yapısı görüntülenmiştir. Plazma aktüatörün açık ve kapalı durumu karşılaştırıldığında, kaldırma kuvvetinin arttığı ve iz bölgesinin daraldığı görülmüştür. Düşük Reynolds sayılarında, plazma aktuatör ile etkili akış kontrolü sağlanarak uçak kanadı üzerindeki tutunma kaybının daha büyük hücum açılarına kaydırılması sağlanmıştır. Reynolds sayısının 36000 değerinde akış ayrılması 18??ye kadar iyileştirilebilmiş ve 14??de maksimum kaldırma kuvveti, plazmasız durumun iki katı olarak elde edilmiştir. Bununla beraber, peristaltik sürümünün çalışılan elektriksel parametre aralığında dikkate değer bir etkisinin olmadığı ortaya konulmuştur, In this master thesis, the effects of paraelectric and peristaltic signal drive on plasma actuator performance is investigated at active flow control. In the case of the plasma actuator located on NACA2415 airfoil, driven by modulated signal, was activated and deactivated. The effect of actuator to active flow control is examined at Reynolds number between 8000 and 90000 in the wind tunnel. The lift force which acted to airfoil was measured by force balance. The velocity measurements were taken by hot wire probe, located to wake region, and the flow around model is visualized by smoke wire method. When the plasma compared with active and passive case, it is observed that the lift force was increased and wake region was narrowed. The stall angle shifted to higher attack angle by accomplishing effective active flow control at low Reynolds number. The flow separation was enhanced up to 18°, the maximum lift force occurred at 14° attack angle which is doubled compared with when the plasma actuator is closed. On the other hand, the peristaltic actuation has no significant effect at the values of considered experimental parameters …
- Published
- 2013
33. Implementation of artificial bee colony (ABC) algorithm on FPGA for real-time applications
- Author
-
Avcı, Gökmen, Karakaya, Fuat, Avcı, Gökmen, Elektrik-Elektronik Mühendisliği Ana Bilim Dalı, and Fen Bilimleri Enstitüsü
- Subjects
ABC Algoritması ,Artificial neural networks ,ABC Algorithm ,Artificial Neural Networks(ANNs) ,FPGAs ,Engineering Sciences ,YSA ,FPGA ,Mühendislik Bilimleri - Abstract
Bu tez çalışmasında Yapay Arı Kolonisi(Artifical Bee Colony, ABC) algoritması ilk kez FPGA üzerinde donanımsal olarak gerçeklenmiştir. ABC algoritması sürü tabanlı sezgisel (metaheuristic) yeni bir algoritmadır. Arıların doğada polen arama davranışlarından adapte edilmiş ve bir çok optimizasyon probleminde başarı ile kullanılmıştır. Bu çalışmada ABC algoritmasının FPGA üzerinde etkin bir şekilde gerçeklenmesini sağlayan donanım mimarisi önerilmiştir. Önerilen mimari sayesinde algoritma FPGA üzerinde daha az yer kaplamakta ve PC tabanlı yazılım versiyonlarına göre daha hızlı çalışmaktadır. Önerilen mimarinin performansının test edilmesi için literatürden yüksek hesaplama gücü ve hız gerektiren fonksiyonlar seçilmiştir. Aynı zamanda ABC algoritması ile YSA eğitimi problemi de FPGA ortamında gerçeklenmiştir. Yapılan çalışmalar sonucunda donanım üzerinde gerçeklenen yapının hem optimizasyon problemlerinde hemde YSA eğitiminde PC tabanlı uygulamalarla aynı doğruluğu gösterirken, bu uygulamlara göre çok daha hızlı olduğu gözlemlenmiştir., In this thesis the first hardware implementation of the artificial bee colony (ABC) optimization on Field-Programmable Gate Arrays (FPGAs) is introduced. The artificial bee colony optimization is a new population?based metaheuristic algorithm. It is adopted from the natural foraging behavior of real honey bees and has been employed to find solutions to wide variety of optimization problems. In this paper we propose an efficient hardware architecture which leads to an efficient implementation with a less space requirement on FPGA and reduced runtime over software based solutions. Proposed architecture is tested with functions selected from the litrature. In addition to that Artificial Neural Network (ANN) tarining with ABC algorithm is also implemented on FPGA. The results indicate that the proposed architacture is as accurate as its software counterpart on PC but much faster than its corresponding software implementation in both function optimization and ANN training.
- Published
- 2011
34. Implementation of HOG algorithm for Real Time Object Recognition Applications on FPGA based Embedded System
- Author
-
Fuat Karakaya, Mehmet Ali Cavuslu, Halis Altun, altun, halis -- 0000-0002-2126-8757, [Karakaya, Fuat -- Altun, Halis -- Cavuslu, Mehmet Ali] Nigde Univ Elekt Elekt Muhendisligi, Nigde, Turkey, and 0-Belirlenecek
- Subjects
Artificial neural network ,Computer science ,business.industry ,Orientation (computer vision) ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Cognitive neuroscience of visual object recognition ,Object (computer science) ,Computational geometry ,0-Belirlenecek ,Edge detection ,Histogram ,Embedded system ,Computer vision ,Artificial intelligence ,business ,Algorithm - Abstract
IEEE 17th Signal Processing and Communications Applications Conference -- APR 09-11, 2009 -- Antalya, TURKEY, WOS: 000273935600185, Recent years HOG algorithm has been used to recognize objects in images, with complex content, with a very high success rate. Hardware implementation of this algorithm is very important because of the fact that it can be used in many object recognition applications. In this work HOG algorithm is implemented on FPGA to recognize different geometrical figures with a very high success rate. Objects vertical and horizontal edges have been sharpened using edge detection algorithms to calculate magnitude and angle of the local gradients. Obtained result are used to construct the histograms of gradient orientation. It is observed that each constructed histogram have distinctive features for every object. Rule based classifiers has been used to implement a successful real time object recognition approach on embedded system., IEEE
- Published
- 2009
35. The Effect of Genetic Algorithm Parameters on the Solution of Plate Location Detection
- Author
-
Murat Peker, Fuat Karakaya, Halis Altun, 0-Belirlenecek, altun, halis -- 0000-0002-2126-8757, and [Peker, Murat] Y Vizyon Sinyalizasyon Hacettepe Unv, Beytepe Kampusu Kulucka Ar Ge Binasi 12, Ankara, Turkey -- [Altun, Halis -- Karakaya, Fuat] Nigde Univ, Elekt Elekt Muhendisligi Bolumu, Nigde, Turkey
- Subjects
Artificial neural network ,Computer science ,Genetic algorithm ,Detector ,Image processing ,Data mining ,computer.software_genre ,computer ,Algorithm ,0-Belirlenecek ,Linear search ,Image (mathematics) - Abstract
IEEE 16th Signal Processing and Communications Applications Conference -- APR 20-22, 2008 -- Aydin, TURKEY, WOS: 000261359200068, In this study, a new method based on genetic algorithm and neural networks for determining licence plate location is proposed. The effect of genetic algorithm parameters on the quality of solutions is investigated. The method is able to successfully locate a licence plate in avearge 40 msn, on the image of 768x288 size. This score is 200 times quicker compared to sequential search method. Futhermore the method is able to find multiple plates in an image., IEEE
- Published
- 2008
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.