Person: VATANSEVER, FAHRİ
Loading...
Email Address
Birth Date
Research Projects
Organizational Units
Job Title
Last Name
VATANSEVER
First Name
FAHRİ
Name
8 results
Search Results
Now showing 1 - 8 of 8
Publication Gozde: A novel metaheuristic algorithm for global optimization(Elsevier, 2022-11-01) Kuyu, Yiğit Çağatay; Vatansever, Fahri; KUYU, YİĞİT ÇAĞATAY; VATANSEVER, FAHRİ; Elektrik Elektronik Mühendisliği Bölümü; 0000-0002-3885-8622; AAG-8425-2021; AAC-6923-2021This study proposes a new metaheuristic algorithm, called "Geometric Octal Zones Distance Estimation "(GOZDE) algorithm to solve global optimization problems. The presented GOZDE employs a search scheme with the information sharing between the zones considering the distance of the zones utilizing median values. The whole population represents the eight zones that are the combination of different search strategies to guide knowledge dissemination from one zone to others in the search space. To demonstrate the effectiveness of the proposed optimizer, it is compared with two classes of metaheuristics, which are (1) GA, PSO, DE, CS and HS as the classical metaheuristics and (2) BWO, SSA, MVO, HHO, ChOA, AOA and EBOwithCMAR as the up-to-date metaheuristics. The search capability of the proposed algorithm is tested on two different numerical benchmark sets including low and high dimensional problems. The developed algorithm is also adapted to ten real world applications to handle constraint optimization problems. In addition, to further analyse the results of the proposed algorithm, three well-known statistical metrics, Friedman, Wilcoxon rank sum and Whisker-Box statistical tests are conducted. The experimental results statistically show that GOZDE is significantly better than, or at least comparable to the twelve metaheuristic algorithms with outstanding performance in solving numerical functions and real-world optimization problems. (C) 2022 Elsevier B.V. All rights reserved.Publication Haar-MUSİC: A new hybrid method for frequency estimation(Springer, 2022-11-30) Yalçın, Nedim Aktan; Vatansever, Fahri; YALÇIN, NEDİM AKTAN; VATANSEVER, FAHRİ; Mühendislik Fakültesi; Elektrik Elektronik Mühendisliği Bölümü; 0000-0002-0049-7841; 0000-0002-3885-8622; AAH-1474-2021; AAG-8425-2021Parameter estimation is very important in signal analysis. In this study, a new hybrid method based on implementation of Multiple Signal Classification (MUSIC) method with Discrete Haar transform (DHT) coefficients for frequency estimation of signals is proposed. This method decreases the input data size and sampling frequency and limits noise subspace correlation matrix according to Nyquist criteria. The realized simulations and real test data show that the proposed method converges to signals' frequencies faster than the classical MUSIC algorithm and gives accurate results even under high noise.Publication Comparison of the evolutionary algorithm's performances on power flow analysis(Pamukkale Univ, 2018-01-01) Kuyu, Yiğit Çağatay; KUYU, YİĞİT ÇAĞATAY; Erdem, Nergis; ERDEM, NERGİS; Vatansever, Fahri; VATANSEVER, FAHRİ; Yılmaz, Güneş; YILMAZ, GÜNEŞ; Mühendislik Fakültesi; Elektrik ve Elektronik Mühendisliği Bölümü; 0000-0002-3885-8622; 0000-0001-8972-1952; AAG-8425-2021; AAC-6923-2021; AAH-4182-2021; AAH-4017-2021Power flow in energy systems is one of the major problems. Several classical analysis methods are utilized for solving this problem. However, power generation limits, valve loading effects of units also makes the power flow problem become much harder to solve in the system. In this case, it is possible to achieve the most appropriate solutions with evolutionary algorithms. In this study, optimal power flow problems are solved under same beginning conditions, comprehensively performed with evolutionary algorithms which are recently used and associated algorithm performance is analyzed in IEEE 30-bus test system for two cases. Energy gains of algorithms are obtained; the best, worst and mean values found from optimization are evaluated; convergence analyses are performed comparatively. Thus the effectiveness and efficiency of evolutionary algorithms are clearly demonstrated on solution of optimal power flow problems.Publication A new hybrid method for signal estimation based on haar transform and prony analysis(IEEE-inst Electrical Electronics Engineers Inc, 2021-01-01) Yalçın, Nedim Aktan; Vatansever, Fahri; YALÇIN, NEDİM AKTAN; VATANSEVER, FAHRİ; Elektrik Elektronik Mühendisliği Bölümü; 0000-0002-0049-7841; 0000-0002-3885-8622; AAH-1474-2021; AAG-8425-2021The signal estimation is very important in electrical and electronic engineering. In this study, it is shown that signal parameters' (frequency, amplitude, and phase) estimation can be realized with the implementation of Prony method on Haar transform coefficients. In order to accomplish this, mathematical relationship between roots of Prony polynomial which are found with original signal values and roots which are calculated with Haar approximation/detail coefficients is constructed. Frequency components of signal are estimated with this relationship. Next, the second part of Prony algorithm which constructs the matrix equation between roots and signal values in order to find the amplitude and phase values is implemented with Haar coefficients. In other words, a new matrix equation is derived for finding amplitudes and phases with the found roots in the first step and Haar coefficients. Thus, implementations of the first and second steps give signal parameters. Derived equations are valid for all degrees of Haar coefficients not just the first one. The use of Haar coefficients decreases the data size and increases the speed and accuracy. The proposed method is also more robust of selection of different Prony polynomial coefficient sizes.Publication A conceptual investigation of the effect of random numbers over the performance of metaheuristic algorithms(Springer, 2023-03-31) Kuyu, Yiğit Çağatay; Vatansever, Fahri; VATANSEVER, FAHRİ; Mühendislik Fakültesi; Elektrik Elektronik Mühendisliği Bölümü; 0000-0002-3885-8622; AAG-8425-2021A lot of research studies focus on the development of a new algorithm or the techniques which improve the performance of the original algorithm. Very few studies conduct the research on the effect of the initial population on the solution quality of algorithms. However, in these studies, one or two algorithms have been used, and a limited number of problems have been handled. To fill in the gap in the literature, this study presents a comprehensive analysis of the five algorithms on the effect of the initial population on their final results including both the numerical and real-world problems along with a wide variety of types of distributions. The study consisted of three rounds and followed the strategy for determining the candidate algorithms to be participated in the next rounds, supported by the statistical tests. Rather than using popular random numbers, fourteen different distributions are used to imitate the random numbers in the initial population generation mechanisms of the algorithms. Two different numerical benchmark sets along with nine real-world problems are used to evaluate the performance of the algorithms. The results are compared with the original ones and other distribution-integrated algorithms. Since knowledge of the appropriate random number source is not available a priori, this study could be a good foundation for future studies not only on the matter of the effect of several distributions on the performances of the algorithms but also introducing an alternative way in generating an initial population.Publication Modified forensic-based investigation algorithm for global optimization(Springer, 2021-02-26) Kuyu, Yiğit Çağatay; Vatansever, Fahri; KUYU, YİĞİT ÇAĞATAY; VATANSEVER, FAHRİ; Mühendislik Fakültesi; Elektrik Elektronik Mühendisliği Bölümü; 0000-0002-3885-8622; 0000-0002-7054-3102; AAG-8425-2021; AAC-6923-2021Forensic-based investigation (FBI) is recently developed metaheuristic algorithm inspired by the suspect investigation-location-pursuit operations of police officers. This study focuses on the search processes of the FBI algorithm, called Step A and Step B, to improve and increase its performance. For this purpose, opposition-based learning is adopted to Step A to enhance diversity, while Cauchy-based mutation is integrated with Step B to guide the search to different regions and to jump out of local minima. To show the effectiveness of these improvements, the proposed algorithm has been tested with two different benchmark sets. To verify the performance of the new modified algorithm, the statistical test is carried out on numerical functions. This study also investigates the application of the proposed algorithm to a set of six real-world problems. The proposed and adapted/integrated methods appear to have a significant impact on the FBI algorithm, which augments its performance, resulting in better solutions than the compared algorithms in most of the functions and real-world problems.Publication Advanced metaheuristic algorithms on solving multimodal functions: Experimental analyses and performance evaluations(Springer, 2021-12) Kuyu, Yiğit Çağatay; Vatansever, Fahri; KUYU, YİĞİT ÇAĞATAY; VATANSEVER, FAHRİ; Elektrik Mühendisliği Bölümü; 0000-0002-3885-8622; 0000-0002-7054-3102; AAC-6923-2021; AAG-8425-2021Optimization problems encountered in real-world have multiple local minimums. Multimodal functions can well represent many real-world applications as they include two or more local minimum points in nature. Numerous metaheuristic algorithms aim to find the best balance between exploration and exploitation, and better algorithms have been developed during the search for such a balance. Therefore, it becomes necessary to answer the question: Which metaheuristic algorithm is the best-suited algorithm among the metaheuristics that have been developed? This study presents a comprehensive and fair investigation of the seven metaheuristic algorithms developed in the last five years on twenty multimodal functions with a wide range of dimensions commonly used in literature. Each is subject to the same initial conditions but with three different performance criteria. The strengths and weaknesses of the each algorithm were demonstrated for each criterion and the experimental results were analyzed statistically by using the Friedman test. Furthermore, to the best of our knowledge, this is the first attempt to address these challenging problems, in combination with these algorithms and performance metrics, which can also give a further insight to the researchers for choosing appropriate algorithms in the context of global optimization.Publication A comparative study of the state-of-the-art algorithms on multi-objective problems using performance metrics(Ieee, 2019-01-01) Kuyu, Yigit Çağatay; KUYU, YİĞİT ÇAĞATAY; Vatansever, Fahri; VATANSEVER, FAHRİ; Mühendislik Fakültesi; Elektrik ve Elektronik Mühendisliği Bölümü; 0000-0002-3885-8622; AAC-6923-2021; AAG-8425-2021Most real-world optimization problems have several objectives which can require to satisfy simultaneously. As the complexity of the problems increases, the problems become more challenging to solve for the algorithms. Over the last few years, many state-of-the-art multi-objective optimization algorithms (MOOAs) have been developed but, to the best knowledge of the authors, there has been no comprehensive comparative study in the literature together with these algorithms. Moreover, there are many parameters that affect the performances of algorithms, such as the population size, iteration number, and initial population. These values may also differ from one study to another, making comparisons more difficult to perform fairly. In this study, four MOOAs which were not previously used together for the comparisons of performance are applied to find optimal solutions of various mathematical optimization problems with the same initial conditions. The performances of the algorithms are evaluated by using two different performance metrics. In addition, Pareto fronts found by the algorithms are given comparatively. Experimental results give an overview for the performance of each algorithm on the different type of the problems by making fair comparison.