Mahir Dursun - Academia.edu (original) (raw)

Papers by Mahir Dursun

Research paper thumbnail of COVID-19 detection by X-Ray images and Deep Learning

2021 6th International Conference on Mathematics and Artificial Intelligence

The new Coronavirus, also called COVID-19, appeared in Wuhan, China in December 2019 Spread out b... more The new Coronavirus, also called COVID-19, appeared in Wuhan, China in December 2019 Spread out big. The virus has now made the COVID-19 disease a worldwide epidemic. This virus has affected the global health, economy, and the daily lives of individuals. The timely diagnosis of COVID-19 is a crucial task as it reduces the risk of pandemic spread. Chest x-rays play an important role in the testing and diagnosis of COVID-19 disease in the recent pandemic. Advanced Artificial Intelligence techniques such as Deep Learning have shown high efficiency in detecting patterns such as those that can be found in diseased tissue, so convolution neural networks were evaluated for their ability to detect infected patients from chest X-ray images. The key component of deep learning research is the availability of training data sets. In this study, a parallel Convolutional Neural Networks (CNNs) model has been proposed for the detection of covid-19 infected patients using chest X-ray radiographs. This paper also presents how to evaluate the effectiveness of the state-of-the-art CNN proposed by the scientific community about their expertise in the automatic diagnosis of Covid-19 from thoracic x-rays. To validate results, we trained the CNN model network by using 500 Covid-19 Positive images and 1600 Covid-19 Negative images. It was given a classification accuracy of 90% and validation accuracy of 88%.

Research paper thumbnail of Preface: ICEEE 2023

2023 10th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of Effect of Electrical Vehicles Participation Rate on Time Delayed Load Frequency Control

2023 10th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of İnsansız Hava Aracı (İHA) Veri Linki İçin Raspberry Pi 3 ile https Üzerinden Güvenli Video Transferi Uygulaması

DergiPark (Istanbul University), Jan 31, 2018

Research paper thumbnail of Low Cost Water Level Control with Radio Frequency

DergiPark (Istanbul University), Dec 1, 2010

Research paper thumbnail of Convolutional Neural Networks Based Turkish Text Classification

INTERNATIONAL JOURNAL OF INFORMATICS TECHNOLOGIES, Jan 31, 2023

Research paper thumbnail of Expandable Optimum Avionics Architecture Solution with Advantage in terms of Cabling Weight for a Helicopter

Europan journal of science and technology, Dec 17, 2022

Research paper thumbnail of Uygun Yenilenebilir Hibrit Çiftlerinin Seçilmesi: Bir Vaka Çalışması

Journal of Polytechnic

The renewable hybrid energy system may be quite an energy that uses the same energy infrastructur... more The renewable hybrid energy system may be quite an energy that uses the same energy infrastructure. Although hybrid energy structures are fed from different types of renewable energy sources, they are systems that give energy to the grid over the same busbar and energy transmission line. The amount of energy that hybrid energy systems can deliver to the grid is limited by the installed capacity of the main source. A well-designed hybrid system can produce more stable electricity throughout the year. The advanced technology networks of various energy sources guarantee the very best efficiency in energy production. Hybrid energy systems combine centralized and non-local storage with intelligent and low-loss control and energy consumption. Also, energy management is provided owing to this integrated renewable hybrid system. During this study, a brand-new algorithm was designed for renewable hybrid selection. Algorithm and integrated renewable hybrid systems are clarified by a case stud...

Research paper thumbnail of Deadbeat control of a DC servo motor at low speed

2018 4th International Conference on Control, Automation and Robotics (ICCAR), 2018

In a robot system, it is very important that electrical motor acceleration time, motors run at th... more In a robot system, it is very important that electrical motor acceleration time, motors run at the desired stability, overshoot and state speed this stability. Especially at low speed of motor control is more difficult because of its nonlinear electromagnetic characteristic. In this study, the parameters of a servo motor were determined, and its dynamic equations were derived. The motor dynamic parameters and speed were determined. It was realized with deadbeat in the PID and MATLAB/Simulink environment, and the performance was compared. It was observed from the obtained results that the rise time of the motor speed controlled by the deadbeat control system was 9% shorter, its initial overflow was 7% smaller and its placement time was 8% shorter. Moreover, the oscillation after the initial overflow did not occur in the PID control method. It was concluded from the obtained results that the deadbeat controlled DC motor could be used in industrial and robotic systems.

Research paper thumbnail of Velocity Control of Linear Switched Reluctance Motor for Prototype Elevator Load

Przegląd Elektrotechniczny, 2011

In this study, a double-sided, 12/8 poled, 3 phased Linear Switched Reluctance Motor (LSRM) desig... more In this study, a double-sided, 12/8 poled, 3 phased Linear Switched Reluctance Motor (LSRM) designed, manufactured and its velocity simulated by fuzzy logic control. Motor power is about 900 W. LSRM drive applied to a prototype elevator. LSRM motion distance is 2000 mm and it having about 500 N propulsion force used in locations such as elevators where linearly moving accurate position, easy control and rapid response are requested. Its operation and switching sections as motor and generator are determined and the simulation of fuzzy logic and PI speed and position control are simulated. Reference velocity is constant during the accelerating and decelerating. As a different, reference velocity is variable depended to position and time. The velocity of translator was controlled versus to position by Fuzzy Logic Control (FLC) and PI control techniques. Velocity and velocity error signals of translator separately were compared the simulation results PI and FLC techniques. It is conclud...

Research paper thumbnail of Matrix Converter Application

DergiPark (Istanbul University), Sep 1, 2008

Research paper thumbnail of Fuzzy Logic Controlled Switched Reluctance Motor Driver Designing for a Lift System

DergiPark (Istanbul University), Feb 1, 2006

Research paper thumbnail of Reservoir Effect on the Hybrid Solar-Hydroelectric (SHE) System

Journal of Polytechnic

The solar-hydroelectric (SHE) energy system is a renewable hybrid energy system consisting of sol... more The solar-hydroelectric (SHE) energy system is a renewable hybrid energy system consisting of solar and hydroelectric energy. An optimization algorithm has been designed to work out the installed power size of the SHE hybrid system, which is planned to be integrated into the existing hydroelectric power systems. This designed algorithm provides the optimum installed power with the benefit/cost approach. The value of the hydro cost and also the energy generation is taken from the actual values since it's an existing facility, and also the electricity production and price of the solar power are obtained from the algorithm that works iteratively. This study aims to indicate that more electricity will be produced by regulating water flows due to the reservoir of hydroelectric power plants. Hydro energy enables energy management to be administrated more effectively with the reservoir, which could be a natural enclosure, without using the other energy storage equipment/method. As a re...

Research paper thumbnail of Switched Reluctance Motor Drive with Pic Microcontroller

DergiPark (Istanbul University), Dec 1, 2003

Research paper thumbnail of Analysis of Fuel Cell Based Time Delayed Load Frequency Control Having Energy Storage Systems

2022 9th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of Determination Of Geometric Dimensions Of A Double Sided Linear Switched Reluctance Motor

In this study, a double-sided linear switched reluctance motor (LSRM) drive was investigated as a... more In this study, a double-sided linear switched reluctance motor (LSRM) drive was investigated as an alternative actuator for vertical linear transportation applications such as a linear elevator door, hospital and subway doors which move linearly and where accurate position control and rapid response is requested. A prototype sliding elevator door that is focused on a home elevator with LSRMs is designed. The motor has 6/4 poles, 3 phases, 8A, 24V, 250 W and 250 N pull forces. Air gap between rotor and translator poles of the designed motor and phase coil-s ideal inductance profile are obtained in compliance with the geometric dimensions. Operation and switching sections as motor and generator has been determined from the inductance profile.

Research paper thumbnail of A HPA Applicatiop for Greenhouse Temperature Control System

Research paper thumbnail of Simulation of Spatial Variability of Organic Matter on the Vineyard Area Using the Model of Artificial Neural Networks

Crop Modeling and Decision Support, 2009

... Soil Sci. 63:251 - 263. Yıldırım S, Sukkar MF, Demirci R, Aslantas V (1996) 'Design of a... more ... Soil Sci. 63:251 - 263. Yıldırım S, Sukkar MF, Demirci R, Aslantas V (1996) 'Design of adaptive NNs-Robust-PID controller for a robot control' Proceedings of IEEE Int. Symposium on Intelligent Control, Michigan, USA, pp.508 - 513. ...

Research paper thumbnail of Dempster Shafer Evidence Theory Application for Fault Diagnosis of Power Transformers

2022 9th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of A HPA Application for Greenhouse Temperature Control System

Research paper thumbnail of COVID-19 detection by X-Ray images and Deep Learning

2021 6th International Conference on Mathematics and Artificial Intelligence

The new Coronavirus, also called COVID-19, appeared in Wuhan, China in December 2019 Spread out b... more The new Coronavirus, also called COVID-19, appeared in Wuhan, China in December 2019 Spread out big. The virus has now made the COVID-19 disease a worldwide epidemic. This virus has affected the global health, economy, and the daily lives of individuals. The timely diagnosis of COVID-19 is a crucial task as it reduces the risk of pandemic spread. Chest x-rays play an important role in the testing and diagnosis of COVID-19 disease in the recent pandemic. Advanced Artificial Intelligence techniques such as Deep Learning have shown high efficiency in detecting patterns such as those that can be found in diseased tissue, so convolution neural networks were evaluated for their ability to detect infected patients from chest X-ray images. The key component of deep learning research is the availability of training data sets. In this study, a parallel Convolutional Neural Networks (CNNs) model has been proposed for the detection of covid-19 infected patients using chest X-ray radiographs. This paper also presents how to evaluate the effectiveness of the state-of-the-art CNN proposed by the scientific community about their expertise in the automatic diagnosis of Covid-19 from thoracic x-rays. To validate results, we trained the CNN model network by using 500 Covid-19 Positive images and 1600 Covid-19 Negative images. It was given a classification accuracy of 90% and validation accuracy of 88%.

Research paper thumbnail of Preface: ICEEE 2023

2023 10th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of Effect of Electrical Vehicles Participation Rate on Time Delayed Load Frequency Control

2023 10th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of İnsansız Hava Aracı (İHA) Veri Linki İçin Raspberry Pi 3 ile https Üzerinden Güvenli Video Transferi Uygulaması

DergiPark (Istanbul University), Jan 31, 2018

Research paper thumbnail of Low Cost Water Level Control with Radio Frequency

DergiPark (Istanbul University), Dec 1, 2010

Research paper thumbnail of Convolutional Neural Networks Based Turkish Text Classification

INTERNATIONAL JOURNAL OF INFORMATICS TECHNOLOGIES, Jan 31, 2023

Research paper thumbnail of Expandable Optimum Avionics Architecture Solution with Advantage in terms of Cabling Weight for a Helicopter

Europan journal of science and technology, Dec 17, 2022

Research paper thumbnail of Uygun Yenilenebilir Hibrit Çiftlerinin Seçilmesi: Bir Vaka Çalışması

Journal of Polytechnic

The renewable hybrid energy system may be quite an energy that uses the same energy infrastructur... more The renewable hybrid energy system may be quite an energy that uses the same energy infrastructure. Although hybrid energy structures are fed from different types of renewable energy sources, they are systems that give energy to the grid over the same busbar and energy transmission line. The amount of energy that hybrid energy systems can deliver to the grid is limited by the installed capacity of the main source. A well-designed hybrid system can produce more stable electricity throughout the year. The advanced technology networks of various energy sources guarantee the very best efficiency in energy production. Hybrid energy systems combine centralized and non-local storage with intelligent and low-loss control and energy consumption. Also, energy management is provided owing to this integrated renewable hybrid system. During this study, a brand-new algorithm was designed for renewable hybrid selection. Algorithm and integrated renewable hybrid systems are clarified by a case stud...

Research paper thumbnail of Deadbeat control of a DC servo motor at low speed

2018 4th International Conference on Control, Automation and Robotics (ICCAR), 2018

In a robot system, it is very important that electrical motor acceleration time, motors run at th... more In a robot system, it is very important that electrical motor acceleration time, motors run at the desired stability, overshoot and state speed this stability. Especially at low speed of motor control is more difficult because of its nonlinear electromagnetic characteristic. In this study, the parameters of a servo motor were determined, and its dynamic equations were derived. The motor dynamic parameters and speed were determined. It was realized with deadbeat in the PID and MATLAB/Simulink environment, and the performance was compared. It was observed from the obtained results that the rise time of the motor speed controlled by the deadbeat control system was 9% shorter, its initial overflow was 7% smaller and its placement time was 8% shorter. Moreover, the oscillation after the initial overflow did not occur in the PID control method. It was concluded from the obtained results that the deadbeat controlled DC motor could be used in industrial and robotic systems.

Research paper thumbnail of Velocity Control of Linear Switched Reluctance Motor for Prototype Elevator Load

Przegląd Elektrotechniczny, 2011

In this study, a double-sided, 12/8 poled, 3 phased Linear Switched Reluctance Motor (LSRM) desig... more In this study, a double-sided, 12/8 poled, 3 phased Linear Switched Reluctance Motor (LSRM) designed, manufactured and its velocity simulated by fuzzy logic control. Motor power is about 900 W. LSRM drive applied to a prototype elevator. LSRM motion distance is 2000 mm and it having about 500 N propulsion force used in locations such as elevators where linearly moving accurate position, easy control and rapid response are requested. Its operation and switching sections as motor and generator are determined and the simulation of fuzzy logic and PI speed and position control are simulated. Reference velocity is constant during the accelerating and decelerating. As a different, reference velocity is variable depended to position and time. The velocity of translator was controlled versus to position by Fuzzy Logic Control (FLC) and PI control techniques. Velocity and velocity error signals of translator separately were compared the simulation results PI and FLC techniques. It is conclud...

Research paper thumbnail of Matrix Converter Application

DergiPark (Istanbul University), Sep 1, 2008

Research paper thumbnail of Fuzzy Logic Controlled Switched Reluctance Motor Driver Designing for a Lift System

DergiPark (Istanbul University), Feb 1, 2006

Research paper thumbnail of Reservoir Effect on the Hybrid Solar-Hydroelectric (SHE) System

Journal of Polytechnic

The solar-hydroelectric (SHE) energy system is a renewable hybrid energy system consisting of sol... more The solar-hydroelectric (SHE) energy system is a renewable hybrid energy system consisting of solar and hydroelectric energy. An optimization algorithm has been designed to work out the installed power size of the SHE hybrid system, which is planned to be integrated into the existing hydroelectric power systems. This designed algorithm provides the optimum installed power with the benefit/cost approach. The value of the hydro cost and also the energy generation is taken from the actual values since it's an existing facility, and also the electricity production and price of the solar power are obtained from the algorithm that works iteratively. This study aims to indicate that more electricity will be produced by regulating water flows due to the reservoir of hydroelectric power plants. Hydro energy enables energy management to be administrated more effectively with the reservoir, which could be a natural enclosure, without using the other energy storage equipment/method. As a re...

Research paper thumbnail of Switched Reluctance Motor Drive with Pic Microcontroller

DergiPark (Istanbul University), Dec 1, 2003

Research paper thumbnail of Analysis of Fuel Cell Based Time Delayed Load Frequency Control Having Energy Storage Systems

2022 9th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of Determination Of Geometric Dimensions Of A Double Sided Linear Switched Reluctance Motor

In this study, a double-sided linear switched reluctance motor (LSRM) drive was investigated as a... more In this study, a double-sided linear switched reluctance motor (LSRM) drive was investigated as an alternative actuator for vertical linear transportation applications such as a linear elevator door, hospital and subway doors which move linearly and where accurate position control and rapid response is requested. A prototype sliding elevator door that is focused on a home elevator with LSRMs is designed. The motor has 6/4 poles, 3 phases, 8A, 24V, 250 W and 250 N pull forces. Air gap between rotor and translator poles of the designed motor and phase coil-s ideal inductance profile are obtained in compliance with the geometric dimensions. Operation and switching sections as motor and generator has been determined from the inductance profile.

Research paper thumbnail of A HPA Applicatiop for Greenhouse Temperature Control System

Research paper thumbnail of Simulation of Spatial Variability of Organic Matter on the Vineyard Area Using the Model of Artificial Neural Networks

Crop Modeling and Decision Support, 2009

... Soil Sci. 63:251 - 263. Yıldırım S, Sukkar MF, Demirci R, Aslantas V (1996) 'Design of a... more ... Soil Sci. 63:251 - 263. Yıldırım S, Sukkar MF, Demirci R, Aslantas V (1996) 'Design of adaptive NNs-Robust-PID controller for a robot control' Proceedings of IEEE Int. Symposium on Intelligent Control, Michigan, USA, pp.508 - 513. ...

Research paper thumbnail of Dempster Shafer Evidence Theory Application for Fault Diagnosis of Power Transformers

2022 9th International Conference on Electrical and Electronics Engineering (ICEEE)

Research paper thumbnail of A HPA Application for Greenhouse Temperature Control System