Browsing by Author "Almali, Mehmet Nuri"

Now showing 1 - 7 of 7

Design and Control of a Novel Topology for Multilevel Inverters Using High Frequency Link
(Elsevier Science Sa, 2023) Hatas, Hasan; Almali, Mehmet Nuri
The requirement of more than one source in multilevel inverters is an issue to be solved for applications with a single DC source. One solution to this problem is to obtain the required voltages using High Frequency Link (HFL). With the traditional use structure of HFL, only one DC source is generated from each diode bridge unit. In this study, voltage generation is diversified as full and half by adding only one switching device and one diode to this circuit. With the proposed modified HFL circuit topology, 35 levels were obtained from 15 level generating circuits. It was compared with some recently published papers using HFL and a satisfactory result was observed in terms of total standing voltage (TSV), although the use of switching devices was reasonable. Proposed to-pology with different frequency and amplitude outputs has been verified both in simulation environment and experimentally using FPGA board. In terms of the economic dependability of the proposed inverter, calculations for the TSV, cost per level factor, and efficiency have been performed. According to the results obtained, it has been observed that the total harmonic distortion of the output waveform of the proposed topology is 2.5%.
Design and Control of Bypass Diode Multilevel Inverter Using a Single Dc Source
(Elsevier Science Sa, 2023) Hatas, Hasan; Almali, Mehmet Nuri
The main benefit of using asymmetrical sources in multilevel inverters is that the levels can be created with a minimum number of power sources. However, for most applications it is desirable to use multilevel inverters with a single DC source. In this study, a new topology is proposed to use a single DC source and achieve a low THD. In this topology, bypass diode technique is used and 31 levels are obtained. A small and inexpensive high -frequency link (HFL) is used to obtain the input voltages from a DC source. Nearest Level Control (NLC) and SPWM methods are applied to simulate the output waveform to the reference signal. The power distribution of the HFL voltage levels is designed by calculating according to the NLC method. The proposed topology is implemented in a simulation environment on a resistive load and filter. Experimental results and simulation results were confirmed. In the experimental study, FPGA was used to generate the signals. With the proposed 31-level topology, the output waveform THD value is obtained as 2.7% without filter.
Detection of Parkinson's Disease by Shifted One Dimensional Local Binary Patterns From Gait
(Pergamon-elsevier Science Ltd, 2016) Ertugrul, Omer Faruk; Kaya, Yilmaz; Tekin, Ramazan; Almali, Mehmet Nuri
The Parkinson's disease (PD) is one of the most common diseases, especially in elderly people. Although the previous studies showed that the PD can be diagnosed by expert systems through its cardinal symptoms such as the tremor, muscular rigidity, disorders of movements and voice, it was reported that the presented approaches, which utilize simple motor tasks, were limited and lack of standardization. To achieve a standard approach in PD detection, an approach, which is built on shifted one-dimensional local binary patterns (Shifted 1D-LBP) and machine learning methods, was proposed. Shifted 1D-LBP is built on 1D-LBP, which is sensitive to local changes in a signal. In 1D-LBP the positions of neighbors around center data are constant and therefore, the number of patterns that can be exacted by it is limited. This drawback was solved by Shifted 1D-LBP by changeable positions of neighbors. In evaluation and validation stages, the Gait in Parkinson's Disease (gaitpdb) dataset, which consists of three gait datasets that were recorded in different tasks or experiment protocols, were employed. Statistical features were exacted from formed histograms of gait signals transformed by Shifted 1D-LBP. Whole features and selected features were classified by machine learning methods. Obtained results were compared with statistical features exacted from signals in both time and frequency domains and results reported in the literature. Achieved results showed that the proposed approach can be successfully employed in PD detection from gait. This work is not only an attempt to develop a PD detection method, but also a general-purpose approach that is based on detecting local changes in time ordered signals. (C) 2016 Elsevier Ltd. All rights reserved.
A High-Efficiency Single-Stage Isolated Sepic-Flyback Ac-Dc Led Driver
(Mdpi, 2023) Gurcam, Kenan; Almali, Mehmet Nuri
Regulating LED current and voltage is critical to maintaining a constant luminous flux in AC- or DC-powered LED lighting circuits. Today, users require constant current drivers that can provide a wide range of output voltages to drive different numbers of series-connected LED arrays. This work proposes an LED driver by combining an isolated SEPIC converter operating in the continuous conduction mode (CCM) and a modified Vienna rectifier. The proposed LED driver offers a single-switch control structure by adding a Vienna rectifier to the integrated SEPIC-FLYBACK converter. This driver structure provides many advantages over traditional bridge rectifier structures. The prototype circuit was tested in an 18 W continuous current mode (CCM) to verify its feasibility. As a result of the values obtained from both simulation and prototype circuit models, it has been shown to provide many of the following advantages: 95% high efficiency, high reliability, 4% low total harmonic distortion, 97% high power factor, and 70 V low switching voltage. This work meets class C 3-2 and IEC 61000 standards.
Metaheuristic Algorithm-Based Proportional-Integrative Control of a Twin Rotor Multi Input Multi Output System
(Mdpi, 2024) Cabuker, Ali Can; Almali, Mehmet Nuri
Metaheuristic algorithms are computational techniques based on the collective behavior of swarms and the study of organisms acting in communities. These algorithms involve different types of organisms. Finding controller values for nonlinear systems is a challenging task using classical approaches. Hence, using metaheuristics to find the controller values of a twin rotor multi-input multi-output system (TRMS), one of the nonlinear systems studied in the literature, seems to be more appropriate than using classical methods. In this study, different types of metaheuristic algorithms were used to find the PID controller values for a TRMS, including a genetic algorithm (GA), a dragonfly algorithm, a cuckoo algorithm, a particle swarm optimization (PSO) algorithm, and a coronavirus optimization algorithm (COVIDOA). The obtained graphs were analyzed based on certain criteria for the main rotor and tail rotor angles to reach the reference value in the TRMS. The experimental results show that when the rise and settlement times of the TRMS are compared in terms of performance, the GA took 1.5040 s (seconds) and the COVIDOA took 9.59 s to increase the pitch angle to the reference value, with the GA taking 0.7845 s and the COVIDOA taking 2.4950 s to increase the yaw angle to the reference value. For the settling time, the GA took 11.67 s and the COVIDOA took 28.01 s for the pitch angle, while the GA took 14.97 s and the COVIDOA took 26.69 s for the yaw angle. With these values, the GA and COVIDOA emerge as the foremost algorithms in this context.
Path Following of an Unmanned Ground Vehicle With Gps Feedback Using Model Predictive Control Method
(Gazi Univ, Fac Engineering Architecture, 2023) Bayram, Atilla; Almali, Mehmet Nuri; Al-Naqshbandi, Firas Muhammad
In this paper, path following control of an unmanned ground vehicle is presented based on the feedback of position and orientation errors. The measurement unit of this autonomous vehicle prototyped for such tasks contains a real-time kinematic global positioning system (RTK-GPS), an inertial measurement unit (IMU) and an absolute encoder to accurately define the position and orientation of the car. A model predictive control was proposed for the path following of the mobile robot based on the successive linearized and discretized kinematic model. This optimal control method performs on the lowest position and orientation errors with respect to a non-holonomic virtual vehicle that is considered to move flawlessly on a given reference path and the smoother steering angle. The paths followed here are defined by rationally based splines or known geometric curves created with control points from a digital mapping program. This paper includes both simulation and real-time experimental studies. The outcomes were examined in terms of the design performance and control strategy of the vehicle. Despite the physical constraints on the vehicle prototype, it has been observed that position and orientation errors occur within satisfactory limits. In particular, the fact that the steering angle is not subjected to excessive oscillations indicates that the control method has a good performance.
The Simulation of Sound Signal Masking With Different Chaotic Oscillations and Its Circuit Application
(Tubitak Scientific & Technological Research Council Turkey, 2016) Almali, Mehmet Nuri; Dikici, Zinnur
The chaotic masking process on sound signals using the systems of Lorenz, Rossler, Sprott, Chen, and Arneodo chaotic oscillations are simulated with MATLAB/Simulink. The PID control method providing synchronization in the system is defined and the mean square errors between the times to reach synchronization and the input-output sound signals are compared for the system performance. The Sprott chaotic system is found to be the best synchronized system. Therefore, the application circuits are designed with respect to the Sprott chaotic system and its results are investigated in this study. It is seen that the system marked with the Sprott chaotic oscillator has a shorter time with respect to the time reaching synchronization, while the Lorenz chaotic oscillator gives a smaller mean square error.