Elektrik - Elektronik Mühendisliği Bölümü Koleksiyonu

Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/202

Browse

Now showing 1 - 20 of 158

Adaptive Fault Detection Scheme Using an Optimized Self-healing Ensemble Machine Learning Algorithm
(CHINA ELECTRIC POWER RESEARCH INST, 2022) Yavuz, Levent; Soran, Ahmet; Onen, Ahmet; Li, Xiangjun; Muyeen, S. M.; 0000-0003-1398-9447; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Yavuz, Levent; Soran, Ahmet
This paper proposes a new cost-efficient, adaptive, and self-healing algorithm in real time that detects faults in a short period with high accuracy, even in the situations when it is difficult to detect. Rather than using traditional machine learning (ML) algorithms or hybrid signal processing techniques, a new framework based on an optimization enabled weighted ensemble method is developed that combines essential ML algorithms. In the proposed method, the system will select and compound appropriate ML algorithms based on Particle Swarm Optimization (PSO) weights. For this purpose, power system failures are simulated by using the PSCAD-Python co-simulation. One of the salient features of this study is that the proposed solution works on real-time raw data without using any pre-computational techniques or pre-stored information. Therefore, the proposed technique will be able to work on different systems, topologies, or data collections. The proposed fault detection technique is validated by using PSCAD-Python co-simulation on a modified and standard IEEE-14 and standard IEEE-39 bus considering network faults which are difficult to detect.
All-polymer ultrasonic transducer design for an intravascular ultrasonography application
(TUBITAK SCIENTIFIC & TECHNICAL RESEARCH COUNCIL TURKEY, ATATURK BULVARI NO 221, KAVAKLIDERE, ANKARA, 00000, TURKEY, 2019) Hah, Dooyoung; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü; Hah, Dooyoung
Intravascular ultrasonography (IVUS), a medical imaging modality, is used to obtain cross-sectional views of blood vessels from inside. In IVUS, transducers are brought to the proximity of the imaging targets so that high-resolution images can be obtained at high frequency without much concern of signal attenuation. To eliminate mechanical rotation rendered in conventional IVUS, it is proposed to manufacture a transducer array on a flexible substrate and wrap it around a cylindrical frame. The transducer of consideration is a capacitive micromachined ultrasonic transducer (CMUT). The whole device needs to be made out of polymers to be able to endure a high degree of bending (radius: 1 mm) Bending of the devices leads to considerable changes in the device characteristics, including resonant frequency and pull-in voltage due to geometrical dimension changes and stress induced. The main purpose of this work is to understand the effect of bending on the device characteristics by means of finite element analysis. Another objective of the work is to understand the relationships between such an effect and the device geometries. It is learned that the bending-induced stress depends strongly on anchor width, membrane thickness, and substrate thickness. It is also learned that resonant frequency and pull-in voltage become lower in most cases because of using a flexible substrate in comparison to those of the device on a rigid substrate. Bending-induced stress increases the spring constant and hence increases resonant frequency and pull-in voltage, although this effect is relatively weaker. For most of the device geometries, pull-in voltage is too high for the polymer material to endure. This is the main drawback of the all-polymer CMUT. In order to meet the design goal of 20 MHz resonant frequency, the membrane radius has to be smaller than 7.7 mu m for a thickness of 3 mu m.
All-Surface Induction Heating With High Efficiency and Space Invariance Enabled by Arraying Squircle Coils in Square Lattice
(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA, 2018) Kilic, Veli Tayfun; Unal, Emre; Yilmaz, Namik; Demir, Hilmi Volkan; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
This paper reports an all-surface induction heating system that enables efficient heating at a constant speed all over the surface independent of the specific location on the surface. In the proposed induction system, squircle coils are placed tangentially in a two-dimensional square lattice as opposed to commonly used hexagonal packing. As a proof-of-concept demonstration, a simple model setup was constructed using a 3 x 3 coil array along with a steel plate to be inductively heated. To model surface heating, a set of six locations for the plate was designated considering symmetry points. For all of these cases, power dissipated by the system and the plate's transient heating were recorded. Independent from the specific plate position, almost equal heating speeds were measured for the similar levels of dissipated energies in the system. Using full three-dimensional electromagnetic solutions, the experimental results were also verified. The findings indicate that the proposed system is proved to enable energy efficient space-invariant heating in all-surface induction hobs.
Analysis of optical gyroscopes with vertically stacked ring resonators
(TUBITAK SCIENTIFIC & TECHNICAL RESEARCH COUNCIL TURKEYATATURK BULVARI NO 221, KAVAKLIDERE, ANKARA 00000, TURKEY, 2021) Hah, Dooyoung; 0000-0002-1290-0597; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Hah, Dooyoung
Without any moving part, optical gyroscopes exhibit superior reliability and accuracy in comparison to mechanical sensors. Microring-resonator-based optical gyroscopes emerged as alternatives for bulky conventional Sagnac interferometer sensors, especially attractive for applications with limited footprints. Previously, it has been reported that planar incorporation of multiple resonators does not bring about improvement in sensitivity for a given area because the increase in Sagnac phase accumulation does not outrun the increase of area. Therefore, it was naturally suggested to consider vertical stacking of ring resonators because then, the resonators can share the same footprint. In this work, sensitivity performances of such configurations with vertically stacked microring resonators are analyzed and compared to that of a basic (single-resonator) configuration. Through comprehensive study, it is learned that the sensitivity performance of the devices with vertically-stacked resonators (either with a single bus waveguide or with two bus waveguides) does not exceed that of the basic sensor device (single resonator with one bus waveguide), i.e. the basic structure is yet to be remained as the most efficient configuration.
Arousal state transitions occlude sensory-evoked neurovascular coupling in neonatal mice
(NATURE PORTFOLIO, 2023) Gheres, Kyle W. W.; Unsal, Hayreddin S. S; Han, Xu; Zhang, Qingguang; Turner, Kevin L.; Zhang, Nanyin; Drew, Patrick J.; 0009-0000-6906-2144; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Unsal, Hayreddin S. S
In the adult sensory cortex, increases in neural activity elicited by sensory stimulation usually drive vasodilation mediated by neurovascular coupling. However, whether neurovascular coupling is the same in neonatal animals as adults is controversial, as both canonical and inverted responses have been observed. We investigated the nature of neurovascular coupling in unanesthetized neonatal mice using optical imaging, electrophysiology, and BOLD fMRI. We find in neonatal (postnatal day 15, P15) mice, sensory stimulation induces a small increase in blood volume/BOLD signal, often followed by a large decrease in blood volume. An examination of arousal state of the mice revealed that neonatal mice were asleep a substantial fraction of the time, and that stimulation caused the animal to awaken. As cortical blood volume is much higher during REM and NREM sleep than the awake state, awakening occludes any sensory-evoked neurovascular coupling. When neonatal mice are stimulated during an awake period, they showed relatively normal (but slowed) neurovascular coupling, showing that that the typically observed constriction is due to arousal state changes. These result show that sleep-related vascular changes dominate over any sensory-evoked changes, and hemodynamic measures need to be considered in the context of arousal state changes. A combination of optical imaging, electrophysiology, and BOLD fMRI in unanesthetized neonatal mice reveals that sleep-related vascular changes dominate over sensory-evoked changes.
Artificial Intelligence Based Intrusion Detection System for IEC 61850 Sampled Values Under Symmetric and Asymmetric Faults
(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC445 HOES LANE, PISCATAWAY, NJ 08855-4141, 2021) Ustun, Taha Selim; Hussain, S. M. Suhail; Yavuz, Levent; Onen, Ahmet; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Yavuz, Levent; Onen, Ahmet
Modern power systems require increased connectivity to implement novel coordination and control schemes. Wide-spread use of information technology in smartgrid domain is an outcome of this need. IEC 61850-based communication solutions have become popular due to a myriad of reasons. Object-oriented modeling capability, interoperable connectivity and strong communication protocols are to name a few. However, power system communication infrastructure is not well-equipped with cybersecurity mechanisms for safe operation. Unlike online banking systems that have been running such security systems for decades, smartgrid cybersecurity is an emerging field. A recent publication aimed at equipping IEC 61850-based communication with cybersecurity features, i.e. IEC 62351, only focuses on communication layer security. To achieve security at all levels, operational technology-based security is also needed. To address this need, this paper develops an intrusion detection system for smartgrids utilizing IEC 61850's Sampled Value (SV) messages. The system is developed with machine learning and is able to monitor communication traffic of a given power system and distinguish normal data measurements from falsely injected data, i.e. attacks. The designed system is implemented and tested with realistic IEC 61850 SV message dataset. Tests are performed on a Modified IEEE 14-bus system with renewable energy-based generators where different fault are applied. The results show that the proposed system can successfully distinguish normal power system events from cyberattacks with high accuracy. This ensures that smartgrids have intrusion detection in addition to cybersecurity features attached to exchanged messages.
Artificial neural networks based harmonics estimation for real university microgrids using hourly solar irradiation and temperature data
(ELSEVIER, 2023) Yarar, Nurcan; Yagci, Mustafa; Bahceci, Serkan; Onen, Ahmet; Ustun, Taha Selim; 0000-0001-7086-5112; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Onen, Ahmet
The need for renewable energy is increasing day by day due to different factors such as increasing energy demand, environmental considerations as well as the will to decrease the share of fossil fuel-based generation. Due to their relative low-cost and ease of installation, PV systems are leading the way for renewable energy deployments around the globe. However, there are meticulous studies that need to be undertaken for realization of such projects. Studying local weather and load patterns for proper panel sizing or considering grid components to determine cable and transformer sizing can be named as some examples for pre-installation studies. In addition to these, post-installation impact studies, e.g. accurate harmonic analysis contribution, is more important to ensure safe and secure operation of the overall system. These steps need to be taken for all PV installation projects. The aim of this study is to show a step-by-step analysis of the effect of a real PV system on the network and to improve the prediction and give a new perspective to the harmonic estimation by using the hourly temperature and radiation data together. At the first phase of the study, a detail real-time 250 kW PV system was modeled for real university campus, and then harmonic estimation based on hourly solar irradiation and hourly temperature was performed with artificial neural networks (ANN) and nonlinear autoregressive exogenous (NARX). The accuracy of the prediction made with ANN was 0.98, and the accuracy of the prediction made with NARX was 0.96.Researchers in PV sizing and control field as well as engineers in power quality area would find these findings beneficial and useful. Use of ANNs and NARX for such analysis indicates the trend in this field that can be targeted by new research projects.
Assessment of Battery Storage Technologies for a Turkish Power Network
(MDPI, ST ALBAN-ANLAGE 66, CH-4052 BASEL, SWITZERLAND, 2019) Kocer, Mustafa Cagatay; Cengiz, Ceyhun; Gezer, Mehmet; Gunes, Doruk; Cinar, Mehmet Aytac; Alboyaci, Bora; Onen, Ahmet; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
Population growth has brought an increase in energy demand and cost that has a meaningful impact on personal and government expenses. In this respect, governments attach importance to investments in renewable energy resources (RER), which are a sustainable and clean energy source. However, the unpredictable characteristics of RER are a major problem for these clean sources and RER need auxiliary assets. Battery energy storage systems (BESS) are one of the promising solutions for these issues. Due to the high investment cost of BESS, governments act cautiously about accepting and implementing BESS in their power network. Recently, with the improvement of technology, the cost of BESS has been reduced, and therefore battery technologies have begun to be applied to conventional systems. In this study, first, we will review and discuss the current globally state-of-the-art BESS and their applications. Later, attention will be turned to a country-specific study for Turkey.
Blockchain-Based Energy Applications: The DSO Perspective
(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC445 HOES LANE, PISCATAWAY, NJ 08855-4141, 2021) Yagmur, Ahmet; Dedeturk, Beyhan Adanur; Soran, Ahmet; Jung, Jaesung; Onen, Ahmet; AGÜ, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümü; Yagmur, Ahmet; Dedeturk, Beyhan Adanur; Soran, Ahmet
This paper discusses blockchain-based energy applications from the distribution system operator (DSO) perspective. Blockchain has a potential impact on newly emergent actors, such as electric vehicles (EVs) and the charging facility units (CFUs) of the electricity grid. Although Blockchain offers magnificent decentralized solutions, the central management of DSOs still plays a significant, non-negligible role, owing to the reality of the existing grid structure. Numerous related studies of proposed blockchain-based EV systems have investigated the energy costs of EVs, fast and efficient charging, privacy and security, P2P energy trading, sharing economy, the selection of appropriate CFUs location, and scheduling. However, cooperation with DSO organizations has not been adequately addressed. Blockchain-based solutions mainly suggest an entirely distributed and decentralized approach for energy trading; however, converting the entire power system infrastructure is considerably expensive. Building a thoroughly decentralized electricity network in a short time is nearly impossible, particularly at the national grid level. In this regard, the applicability of the solutions is as significant as their appropriateness, especially from the DSO perspective, and must be examined closely. We searched and analyzed the blockchain literature related to EVs, CFUs, DERs, microgrids, marketing, and DSOs to define the DSO-based requirements for potential blockchain applications in the energy sector, specifically EV evolution.
BRCA Variations Risk Assessment in Breast Cancers Using Different Artificial Intelligence Models
(MDPIST ALBAN-ANLAGE 66, CH-4052 BASEL, SWITZERLAND, 2021) Senturk, Niyazi; Tuncel, Gulten; Dogan, Berkcan; Aliyeva, Lamiya; Dundar, Mehmet Sait; Ozemri Sag, Sebnem; Mocan, Gamze; Temel, Sehime Gulsun; Dundar, Munis; Ergoren, Mahmut Cerkez; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Dundar, Mehmet Sait
Artificial intelligence provides modelling on machines by simulating the human brain using learning and decision-making abilities. Early diagnosis is highly effective in reducing mortality in cancer. This study aimed to combine cancer-associated risk factors including genetic variations and design an artificial intelligence system for risk assessment. Data from a total of 268 breast cancer patients have been analysed for 16 different risk factors including genetic variant classifications. In total, 61 BRCA1, 128 BRCA2 and 11 both BRCA1 and BRCA2 genes associated breast cancer patients' data were used to train the system using Mamdani's Fuzzy Inference Method and Feed-Forward Neural Network Method as the model softwares on MATLAB. Sixteen different tests were performed on twelve different subjects who had not been introduced to the system before. The rates for neural network were 99.9% for training success, 99.6% for validation success and 99.7% for test success. Despite neural network's overall success was slightly higher than fuzzy logic accuracy, the results from developed systems were similar (99.9% and 95.5%, respectively). The developed models make predictions from a wider perspective using more risk factors including genetic variation data compared with similar studies in the literature. Overall, this artificial intelligence models present promising results for BRCA variations' risk assessment in breast cancers as well as a unique tool for personalized medicine software.
Can Laws Be a Potential PET Image Texture Analysis Approach for Evaluation of Tumor Heterogeneity and Histopathological Characteristics in NSCLC?
(SPRINGER, 233 SPRING ST, NEW YORK, NY 10013 USA, 2018) Karacavus, Seyhan; Yılmaz, Bülent; Tasdemir, Arzu; Kayaaltı, Ömer; Kaya, Eser; İçer, Semra; Ayyıldız, Oguzhan; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
We investigated the association between the textural features obtained from F-18-FDG images, metabolic parameters (SUVmax(,) SUVmean, MTV, TLG), and tumor histopathological characteristics (stage and Ki-67 proliferation index) in non-small cell lung cancer (NSCLC). The FDG-PET images of 67 patients with NSCLC were evaluated. MATLAB technical computing language was employed in the extraction of 137 features by using first order statistics (FOS), gray-level co-occurrence matrix (GLCM), gray-level run length matrix (GLRLM), and Laws' texture filters. Textural features and metabolic parameters were statistically analyzed in terms of good discrimination power between tumor stages, and selected features/parameters were used in the automatic classification by k-nearest neighbors (k-NN) and support vector machines (SVM). We showed that one textural feature (gray-level nonuniformity, GLN) obtained using GLRLM approach and nine textural features using Laws' approach were successful in discriminating all tumor stages, unlike metabolic parameters. There were significant correlations between Ki-67 index and some of the textural features computed using Laws' method (r = 0.6, p = 0.013). In terms of automatic classification of tumor stage, the accuracy was approximately 84% with k-NN classifier (k = 3) and SVM, using selected five features. Texture analysis of FDG-PET images has a potential to be an objective tool to assess tumor histopathological characteristics. The textural features obtained using Laws' approach could be useful in the discrimination of tumor stage.
Cesium-lead based inorganic perovskite quantum-dots as interfacial layer for highly stable perovskite solar cells with exceeding 21% efficiency
(ELSEVIER, RADARWEG 29, 1043 NX AMSTERDAM, NETHERLANDS, 2019) Akin, Seckin; Altintas, Yemliha; Mutlugun, Evren; Sonmezoglu, Savas; AGÜ, Mühendislik Fakültesi, Malzeme Bilimi ve Nanoteknoloji Mühendisliği Bölümü
Despite the excellent photovoltaic performances of perovskite solar cells (PSCs), the instability of PSCs under severe environment (e.g. humidity, light-induced, etc.) limits further commercialization of such devices. Therefore, in recent years, research on the long-term stability improvement of PSCs has been actively carried out in perovskite field. To address these issues, we demonstrated the incorporation of ultra-thin interfacial layer of inorganic CsPbBr1.85I1.15 perovskite quantum-dots (PQDs) that can effectively passivate defects at or near to the perovskite/hole transport material (HTM) interface, significantly suppressing interfacial recombination. This passivation layer increased the open circuit voltage (V-oc) of triple-cation perovskite cells by as much as 50 mV, with champion cells achieving V-oc similar to 1.14 V. As a result, we obtained hysteresis-free cells with the efficiency beyond 21%. More importantly, devices based on such architecture are capable of resisting humidity and light-induced. Remarkably, the device employing CsPbBr1.85I1.15 demonstrated a superb shelf-stability aganist to humidity under ambient conditions (R.H. >= 40%), retaining nearly 91% of initial efficiency after 30 days, while the efficiency of control device rapidly dropped to 45% from its initial value under the same conditions. Besides benefiting from the high moisture resistivity as well as supressed ion migration, PSC5 based on PQDs showed better operational stability (retaining 94% of their initial performance) than that of the PQDs-free one under continuous light irradiation over 400 h. In addition, a faster PL decay time of 4.66 ns was attained for perovskite/PQDs structure (5.77 ns for only PQDs structure) due to the favorable energy transfer at the interface, indicating a Forster resonance energy transfer (FRET) mechanism. This work indicates that inorganic PQDs are important materials as interlayer in PSC5 to supremely enhance the device stability and efficiency.
Chaos in PID Controlled Nonlinear Systems
(SPRINGER SINGAPORE PTE LTD, #04-01 CENCON I, 1 TANNERY RD, SINGAPORE 347719, SINGAPORE, 2015) Ablay, Gunyaz; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü; Ablay, Gunyaz
Controlling nonlinear systems with linear feedback control methods can lead to chaotic behaviors. Order increase in system dynamics due to integral control and control parameter variations in PID controlled nonlinear systems are studied for possible chaos regions in the closed-loop system dynamics. The Lure form of the feedback systems are analyzed with Routh's stability criterion and describing function analysis for chaos prediction. Several novel chaotic systems are generated from second-order nonlinear systems including the simplest continuous-time chaotic system. Analytical and numerical results are provided to verify the existence of the chaotic dynamics.
Clinical probe utilizing surface enhanced Raman scattering
(A V S AMER INST PHYSICS, STE 1 NO 1, 2 HUNTINGTON QUADRANGLE, MELVILLE, NY 11747-4502 USA, 2014) Kim, Jeonghwan; Hah, Dooyoung; Daniels-Race, Theda; Feldman, Martin; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
Conventional Raman scattering is a well-known technique for detecting and identifying complex molecular samples. In surface enhanced Raman scattering, a nanorough metallic surface close to the sample enormously enhances the Raman signal. In previous work, the metallic surface was a thin layer of gold deposited on a rough transparent epoxy substrate. The advantage of the clear substrate was that the Raman signal could be obtained by passing light through the substrate, on to opaque samples simply placed against its surface. In this work, a commercially available Raman spectrometer was coupled to a distant probe. Raman signals were obtained from the surface, and from the interior, of a solid specimen located more than 1 m away from the spectrometer. The practical advantage of this arrangement is that it opens up surface enhanced Raman spectrometry to a clinical environment, with a patient simply sitting or lying near the spectrometer. (C) 2014 American Vacuum Society.
Colloidal Quantum Dot Light-Emitting Diodes Employing Phosphorescent Small Organic Molecules as Efficient Exciton Harvesters
(AMER CHEMICAL SOC, 1155 16TH ST, NW, WASHINGTON, DC 20036 USA, 2014) Mutlugun, Evren; Guzelturk, Burak; Abiyasa, Agus Putu; Gao, Yuan; Sun, Xiao Wei; Demir, Hilmi Volkan; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
Nonradiative energy transfer (NRET) is an alternative excitation mechanism in colloidal quantum dot (QD) based electroluminescent devices (QLEDs). Here, we develop hybrid highly spectrally pure QLEDs that facilitate energy transfer pumping via NRET from a phosphorescent small organic molecule-codoped charge transport layer to the adjacent QDs. A partially codoped exciton funnelling electron transport layer is proposed and optimized for enhanced QLED performance while exhibiting very high color purity of 99%. These energy transfer pumped hybrid QLEDs demonstrate a 6-fold enhancement factor in the external quantum efficiency over the conventional QLED structure, in which energy transfer pumping is intrinsically weak.
Comparative analysis of dimensionality reduction techniques for cybersecurity in the SWaT dataset
(SPRINGER, 2024) Bozdal, Mehmet; Ileri, Kadir; Ozkahraman, Ali; 0000-0002-2081-7101; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Bozdal, Mehmet
The Internet of Things (IoT) has revolutionized the functionality and efciency of distributed cyber-physical systems, such as city-wide water treatment systems. However, the increased connectivity also exposes these systems to cybersecurity threats. This research presents a novel approach for securing the Secure Water Treatment (SWaT) dataset using a 1D Convolutional Neural Network (CNN) model enhanced with a Gated Recurrent Unit (GRU). The proposed method outperforms existing methods by achieving 99.68% accuracy and an F1 score of 98.69%. Additionally, the paper explores dimensionality reduction methods, including Autoencoders, Generalized Eigenvalue Decomposition (GED), and Principal Component Analysis (PCA). The research fndings highlight the importance of balancing dimensionality reduction with the need for accurate intrusion detection. It is found that PCA provided better performance compared to the other techniques, as reducing the input dimension by 90.2% resulted in only a 2.8% and 2.6% decrease in the accuracy and F1 score, respectively. This study contributes to the feld by addressing the critical need for robust cybersecurity measures in IoT-enabled water treatment systems, while also considering the practical trade-of between dimensionality reduction and intrusion detection accuracy.
Comparison of deep learning and conventional machine learning methods for classification of colon polyp types
(SCIENDOBOGUMILA ZUGA 32A, WARSAW, MAZOVIA, POLAND, 2021) Dogan, Refika Sultan; Yilmaz, Bulent; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Dogan, Refika Sultan; Yilmaz, Bulent
Determination of polyp types requires tissue biopsy during colonoscopy and then histopathological examination of the microscopic images which tremendously time-consuming and costly. The first aim of this study was to design a computer-aided diagnosis system to classify polyp types using colonoscopy images (optical biopsy) without the need for tissue biopsy. For this purpose, two different approaches were designed based on conventional machine learning (ML) and deep learning. Firstly, classification was performed using random forest approach by means of the features obtained from the histogram of gradients descriptor. Secondly, simple convolutional neural networks (CNN) based architecture was built to train with the colonoscopy images containing colon polyps. The performances of these approaches on two (adenoma & serrated vs. hyperplastic) or three (adenoma vs. hyperplastic vs. serrated) category classifications were investigated. Furthermore, the effect of imaging modality on the classification was also examined using white-light and narrow band imaging systems. The performance of these approaches was compared with the results obtained by 3 novice and 4 expert doctors. Two-category classification results showed that conventional ML approach achieved significantly better than the simple CNN based approach did in both narrow band and white-light imaging modalities. The accuracy reached almost 95% for white-light imaging. This performance surpassed the correct classification rate of all 7 doctors. Additionally, the second task (three-category) results indicated that the simple CNN architecture outperformed both conventional ML based approaches and the doctors. This study shows the feasibility of using conventional machine learning or deep learning based approaches in automatic classification of colon types on colonoscopy images.
Control of collective bursting in small Hodgkin-Haxley neuron clusters
(Communications Faculty of Sciences University of Ankara, 2018) Borisenok, Sergey; Ünal, Zeynep; Çatmabacak, Önder; AGÜ, Mühendislik Fakültesi, Elektrik & Elektronik Mühendisliği Bölümü;
The speed gradient-based control algorithm for tracking the membrane potential of Hodgkin-Huxley neurons is applied to their small clusters modeling the basic features of an epileptiform dynamics. One of the neurons plays a role of control element detecting the temporal hyper-synchronization among its network companions and switching their bursting behavior to resting. The ‘toy’ model proposed in the paper can serve as an algorithmic basement for developing special control elements at the scale of one or few cells that may work autonomously and are able to detect and suppress epileptic behavior in the networks of real biological neurons.
Control over Cavity Assisted Charging for Dicke Quantum Battery
(ResearchGate, 2020) Sergey Borisenok; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü
We study feed forward (open-loop) control approach for driving the cavity assisted charging process in the Dicke quantum battery, in which the coupling between the cavity and quantum twolevel subsystem(s) plays a role of control parameter.The dynamics of the system is described with the Tavis - Cummings Hamiltonian. The analytical result is supported with the corresponding numerical simulations to demonst
Control over performance of qubit-based sensors
(Institute for Problems in Mechanical Engineering, Russian Academy of Sciences, 2018) Borisenok, Sergey; 0000-0002-1992-628X; AGÜ, Mühendislik Fakültesi, Elektrik - Elektronik Mühendisliği Bölümü; Borisenok, Sergey
The extreme sensitivity of quantum systems towards the external perturbations and in the same time their ability to be strongly coupled to the measured target field makes them to be stable under the environmental noise. A high quality quantum sensor can be engineered even on the platform of a single trapped qubit. There is a variety of optimal and sub-optimal algorithms for effective control over the quantum system states. Here we discuss the opportunity to improve the efficiency of the external field quantum sensor based on a single qubit via its feedback tracking.

Browse

Browsing Elektrik - Elektronik Mühendisliği Bölümü Koleksiyonu by Access Right "info:eu-repo/semantics/openAccess"