Yazar "Kahraman, Hamdi Tolga" seçeneğine göre listele
Listeleniyor 1 - 20 / 20
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Analysis, test and management of the metaheuristic searching process: an experimental study on SOS(2020) Kahraman, Hamdi Tolga; Aras, Sefa; Sönmez, Yusuf; Güvenç, Uğur; Gedikli, EyüpIn a search process, getting trapped in a local minimum or jumping the global minimum problems are also one of the biggestproblems of meta-heuristic algorithms as in artificial intelligence methods. In this paper, causes of these problems are investigatedand novel solution methods are developed. For this purpose, a novel framework has been developed to test and analyze the metaheuristic algorithms. Additionally, analysis and test studies have been carried out for Symbiotic Organisms Search (SOS)Algorithm. The aim of the study is to measure the mimicking a natural ecosystem success of symbiotic operators. Thus, problemsin the search process have been discovered and operators' design mistakes have been revealed as a case study of the developedtesting and analyzing method. Moreover, ways of realizing a precise neighborhood search (intensification) and getting rid of thelocal minimum (increasing diversification) have been explored. Important information that enhances the performance of operatorsin the search process has been achieved through experimental studies. Additionally, it is expected that the new experimental testmethods developed and presented in this paper contributes to meta-heuristic algorithms studies for designing and testing.Öğe Classification of Human and Vehicles with The Deep Learning Based on Transfer Learning Method(2021) Cengiz, Enes; Kahraman, Hamdi Tolga; Yılmaz, CemalThere has been a significant increase in the use of deep learning algorithms in recent years. Convolutional neural network (CNN), one of the deep learning models, is frequently used in applications to distinguish important objects such as humans and vehicles from other objects, especially in image processing. With the development of image processing hardware, the image processing process is significantly reduced. Thanks to these developments, the performance of studies on deep learning is increasing. In this study, a system based on deep learning has been developed to detect and classify objects (human, car and motorcycle / bicycle) from images captured by drones. Two datasets, the image set of Stanford University and the drone image set created at Afyon Kocatepe University (AKÜ), are used to train and test the deep neural network with the transfer learning method. The precision, recall and f1 score values are evaluated according to the process of determining and classifying human, car and motorcycle / bicycle classes using GoogleNet, VggNet and ResNet50 deep learning algorithms. According to this evaluation result, high performance results are obtained with 0.916 precision, 0.895 recall and 0.906 f1 score value in the ResNet50 model.Öğe Combined heat and power economic emission dispatch using dynamic switched crowding based multi-objective symbiotic organism search algorithm(Elsevier, 2024) Ozkaya, Burcin; Kahraman, Hamdi Tolga; Duman, Serhat; Guvenc, Ugur; Akbel, MustafaCombined heat and power economic emission dispatch (CHPEED) problem is a highly complex, non-linear, non -convex multiobjective optimization problem due to two conflicting objectives and various operational constraints such as valve-point loading effect, power transmission loss, prohibited operating zone, and the feasible operating region of combined heat and power unit. In order to overcome these challenges, it is necessary to design an algorithm that exhibits a search behavior, which is suitable for the characteristics of objective and constraint space of the CHPEED problem. For these reasons, a dynamic switched crowding based multi-objective symbiotic organism search (DSC-MOSOS) algorithm was designed to meet the requirements and geometric space of the CHPEED problem. By applying the DSC method in the MOSOS algorithm, it was aimed to improve the exploration ability, to strengthen exploitation-exploration balance, and to prevent the catching into local solution traps. A comprehensive experimental study was carried out to prove the performance of the proposed al-gorithm on IEEE CEC 2020 multi-modal multi-objective problems (MMOPs) and CHPEED problem. In the experimental study conducted among eleven versions of MOSOS variations created with DSC-method and the base MOSOS algorithm on IEEE CEC 2020 MMOPs, according to Friedman scores based on the four performance metrics, the base MOSOS algorithm ranked the last. In other experimental study, the best DSC-MOSOS variant was applied to solve the CHPEED problem, where 5-, 7-, 10-and 14-unit test systems and eight case studies were considered. The important points of this study were that 10-unit and 14-unit test systems were presented to the literature, and the prohibited operating zone was considered in CHPEED problem for the first time. According to the results obtained from eight case studies obtained from the DSC-MOSOS and fourteen competitor algorithms, while the improvement in cost was between 0.2% and 16.55%, the reduction of the emission value was between 0.2 kg and 42.97 kg compared to the competitor algorithms. On the other hand, the stability of the DSC-MOSOS and the base MOSOS was evaluated using stability analysis. While the MOSOS algorithms was not able to perform a success in any case study, the DSC-MOSOS was achieved an average success rate with 91.16%. Thus, the performance of the DSC-MOSOS over the MOSOS was verified by the results of experimental studies and analysis.Öğe Developing of Decision Support System for Land Mine Classification by Meta-heuristic Classifier(Ieee, 2016) Yılmaz, Cemal; Kahraman, Hamdi Tolga; Söyler, Salih; Sönmez, Yusuf; Güvenç, UğurIn this study, a decision support system has been developed for land mine detection and classification. Data obtained from detector based magnetic anomaly have been used to classify the land mines. With this classification, it is decided that whether obtained data belongs to a land mine or not, and the type of mine. The meta-heuristic k-NN classifier (HKC) has been used in developed decision support system. Consequently, it is seen that decision support system detects the presence of mines and decides the type of mine with 100% success for measurements in a certain range, and the proposed classifying method shows much higher performance than traditional instance-based classification method.Öğe Dynamic FDB selection method and its application: modeling and optimizing of directional overcurrent relays coordination(Springer, 2021) Kahraman, Hamdi Tolga; Bakir, Huseyin; Duman, Serhat; Kati, Mehmet; Aras, Sefa; Guvenc, UgurThis article has four main objectives. These are: to develop the dynamic fitness-distance balance (dFDB) selection method for meta-heuristic search algorithms, to develop a strong optimization algorithm using the dFDB method, to create an optimization model of the coordination of directional overcurrent relays (DOCRs) problem, and to optimize the DOCRs problem using the developed algorithm, respectively. A comprehensive experimental study was conducted to analyze the performance of the developed dFDB selection method and to evaluate the optimization results of the DOCRs problem. Experimental studies were carried out in two steps. In the first step, to test the performance of the developed dFDB method and optimization algorithm, studies were conducted on three different benchmark test suites consisting of different problem types and dimensions. The data obtained from the experimental studies were analyzed using non-parametric statistical methods and the most effective among the developed optimization algorithms was determined. In the second step, the DOCRs problem was optimized using the developed algorithm. The performance of the proposed method for the solution to the DOCRs coordination problem was evaluated on five test systems including the IEEE 3-bus, the IEEE 4-bus, the 8-bus, the 9-bus, and the IEEE 30-bus test systems. The numerical results of the developed algorithm were compared with previously proposed algorithms available in the literature. Simulation results showed the effectiveness of the proposed method in minimizing the relay operating time for the optimal coordination of DOCRs.Öğe Exploring the Effect of Distribution Methods on Meta-Heuristic Searching process(Ieee, 2017) Kahraman, Hamdi Tolga; Aras, Sefa; Güvenç, Uğur; Sönmez, YusufIn this study, the effect of distributions of solution candidates on the problem space in the meta-heuristic search process and the performance of algorithms has been investigated. For this purpose, solution candidates have been created with random and gauss (normal) distributions. Search performance is measured separately for both types of distribution of algorithms. The performances of the algorithms have been tested on the most popular and widely used benchmark problems. Experimental studies have been conducted on the most recent meta-heuristic search algorithms. It has been seen that the search performance of algorithms varies considerably depending on the method of distribution. In fact, better results were obtained than the distribution methods used in the original versions of the algorithms. Algorithms have revealed their abilities in terms of neighborhoods searching, getting rid of local minimum traps and speeding up searches.Öğe Fitness-Distance Balance based adaptive guided differential evolution algorithm for security-constrained optimal power flow problem incorporating renewable energy sources(Elsevier, 2021) Guvenc, Ugur; Duman, Serhat; Kahraman, Hamdi Tolga; Aras, Sefa; Kati, MehmetOne of the most difficult types of problems computationally is the security-constrained optimal power flow (SCOPF), a non-convex, nonlinear, large-scale, nondeterministic polynomial time optimization problem. With the use of renewable energy sources in the SCOPF process, the uncertainties of operating conditions and stress on power systems have increased even more. Thus, finding a feasible solution for the problem has become a still greater challenge. Even modern powerful optimization algorithms have been unable to find realistic solutions for the problem. In order to solve this kind of difficult problem, an optimization algorithm needs to have an unusual exploration ability as well as exploitation-exploration balance. In this study, we have presented an optimization model of the SCOPF problem involving wind and solar energy systems. This model has one problem space and innumerable local solution traps, plus a high level of complexity and discrete and continuous variables. To enable the optimization model to find the solution effectively, the adaptive guided differential evolution (AGDE) algorithm was improved by using the Fitness-Distance Balance (FDB) method with its balanced searching and high-powered diversity abilities. By using the FDB method, solution candidates guiding the search process in the AGDE algorithm could be selected more effectively as in nature. In this way, AGDE's exploration and balanced search capabilities were improved. To solve the SCOPF problem involving wind and solar energy systems, the developed algorithm was tested on an IEEE 30-bus test system under different operational conditionals. The simulation results obtained from the proposed algorithm were effective in finding the optimal solution compared to the results of the metaheuristics algorithms and reported in the literature. (C) 2021 Elsevier B.V. All rights reserved.Öğe Fitness-Distance-Constraint (FDC) based guide selection method for constrained optimization problems(Elsevier, 2023) Ozkaya, Burcin; Kahraman, Hamdi Tolga; Duman, Serhat; Guvenc, UgurIn the optimization of constrained type problems, the main difficulty is the elimination of the constraint violations in the evolutionary search process. Evolutionary algorithms are designed by default according to the requirements of unconstrained and continuous global optimization problems. Since there are no constraint functions in these type of problems, the constraint violations are not considered in the design of the guiding mechanism of evolutionary algorithms. In this study, two new methods were introduced to redesign the evolutionary algorithms in accordance with the requirements of constrained optimization problems. These were (i) constraint space-based, called Fitness-Distance -Constraint (FDC), selection method and (ii) dynamic guiding mechanism. Firstly, thanks to the FDC guide selection method, the constraint violation values of the individuals in the population were converted into score values and the individuals who increase the diversity in the search process were selected as guide. On the other hand, in dynamic guiding mechanism, the FDC method was applied in case of constraint violation, otherwise the default guide selection method was used The proposed methods were used to redesign the guiding mechanism of adaptive guided differential evolution (AGDE), a current evolutionary algorithm, and the FDC-AGDE algorithm was designed. The performance of the FDC-AGDE was tested on eleven different constrained real-world optimization problems. The results of the FDC-AGDE and AGDE were evaluated using the Friedman and Wilcoxon test methods. According to Wilcoxon pairwise results, the FDC-AGDE showed better performance than the AGDE in nine of the eleven problems and equal performance in two of the eleven problems. Moreover, the proposed algorithm was compared with the competitive and up-to-date MHS algorithms in terms of the results of Friedman test, Wilcoxon test, feasibility rate, and success rate. According to Friedman test results, the first three algorithms were the FDC-AGDE, LSHADE-SPACMA, and AGDE algorithms with the score of 2.69, 4.05, and 4.34, respectively. According to the mean values of the success rates obtained from the eleven problems, the FDC-AGDE, LSHADE-SPACMA, and AGDE algorithms ranked in the first three with the success rates of 67%, 48% and 28%, respectively. Consequently, the FDC-AGDE algorithm showed a superior performance comparing with the competing MHS algorithms. According to the results, it is expected that the proposed methods will be widely used in the constrained optimization problems in the future.& COPY; 2023 Elsevier B.V. All rights reserved.Öğe Improved adaptive gaining-sharing knowledge algorithm with FDB-based guiding mechanism for optimization of optimal reactive power flow problem(Springer, 2023) Bakir, Huseyin; Duman, Serhat; Guvenc, Ugur; Kahraman, Hamdi TolgaOptimal reactive power flow (ORPF) is of great importance for the electrical reliability and economic operation of modern power systems. The integration of distributed generations (DGs) and two-terminal high voltage direct current (HVDC) systems into electrical networks has further complicated the ORPF problem. Due to the high computational complexity of the ORPF problem, a powerful and robust optimization algorithm is required to solve it. This paper proposes a powerful metaheuristic algorithm namely fitness-distance balance-based adaptive gaining-sharing knowledge (FDBAGSK). In the performance evaluation, 39 IEEE CEC benchmark functions are used to compare FDBAGSK with the original AGSK algorithm. Moreover, the proposed algorithm is applied to perform the ORPF task in modified IEEE 30- and IEEE 57-bus test systems. The effectiveness of the FDBAGSK method was tested for the optimization of three non-convex objectives: active power loss, voltage deviation and voltage stability index. The ORPF results obtained from the FDBAGSK algorithm are compared with other optimization algorithms in the literature. Given that all results are together, it has been observed that FDBAGSK is an effective method that can be used in solving global optimization and constrained real-world engineering problems.Öğe Improved L ' evy flight distribution algorithm with FDB-based guiding mechanism for AVR system optimal design(Pergamon-Elsevier Science Ltd, 2022) Bakır, Hüseyin; Güvenç, Uğur; Kahraman, Hamdi Tolga; Duman, SerhatThis paper presents the improved version of the Le ' vy Flight Distribution (LFD) algorithm for solving real-valued numerical optimization problems. In the proposed algorithm, the Fitness-Distance Balance (FDB) selection method was used to determine the search agents that well know the migration routes and guide the herd. Thus, the FDB-LFD algorithm, which has a much stronger search performance, was developed. The performance of the proposed algorithm was tested and verified on CEC17 and CEC20 benchmark problems for low-, middle-and high-dimensional search spaces. Results of the FDB-LFD was compared to the performance of 11 other powerful and up-to-date metaheuristic search algorithms. According to Friedman statistical test results, the proposed FDBLFD algorithm ranked first, whereas the LFD was ranked eleventh. This result demonstrated that the changes in the design of the LFD algorithm had been successful. Moreover, using the proposed algorithm, optimum solutions were found for one of the popular industrial engineering applications: the automatic voltage regulator (AVR) system design. The simulation results revealed that the FDB-LFD is an effective algorithm for solving both unconstrained benchmark and constrained industrial engineering design problems.Öğe Improved Runge Kutta Optimizer with Fitness Distance BalanceBased Guiding Mechanism for Global Optimization of HighDimensional Problems1(2021) Suiçmez, Çağrı; Kahraman, Hamdi Tolga; Cengiz, Enes; Yılmaz, CemalRunge Kutta (RUN) is an up-to-date and well-founded metaheuristic algorithm. The RUN algorithm aims to find the global best in solving problems by going beyond the traps of metaphors. For this purpose, enhanced solution quality mechanism is used to avoid local optimum solutions and increase the convergence speed. Although the RUN algorithm offers promising solutions, it is seen that this algorithm has shortcomings, especially in solving high dimensional multimodal problems. In this study, the solution candidates that guide the search process in the RUN algorithm are developed using the Fitness-Distance Balance (FDB) method. Thus, using the FDB-based RUN algorithm, the global optimum value of many optimization problems will be obtained in the future. CEC 2020 which has current benchmark problems was used to test the performance of the developed FDB-RUN algorithm. 10 different unconstrained benchmark problems taken from CEC 2020 were designed by arranging them in 30/50/100 dimensions. Experimental studies were carried out using the designed benchmark problems and analyzed with Friedman and Wilcoxon statistical test methods. According to the results of the analysis, it was seen that the FDB-RUN variations showed a superior performance compared to the base algorithm (RUN) in all experimental studies. In particular, it has been shown to provide more effective results for the continuous optimization of high-dimensional problems.Öğe Improved Slime-Mould-Algorithm with Fitness Distance Balancebased Guiding Mechanism for Global Optimization Problems(2021) Işık, Mehmet Fatih; Yılmaz, Cemal; Suiçmez, Çağrı; Cengiz, Enes; Kahraman, Hamdi TolgaIn this study, the performance of Slime-Mould-Algorithm (SMA), a current Meta-Heuristic Search algorithm, is improved. In order to model the search process lifecycle process more effectively in the SMA algorithm, the solution candidates guiding the search process were determined using the fitness-distance balance (FDB) method. Although the performance of the SMA algorithm is accepted, it is seen that the performance of the FDB-SMA algorithm developed thanks to the applied FDB method is much better. CEC 2020, which has current benchmark problems, was used to test the performance of the developed FDB-SMA algorithm. 10 different unconstrained comparison problems taken from CEC 2020 are designed by arranging them in 30-50-100 dimensions. Experimental studies were carried out using the designed comparison problems and analyzed with Friedman and Wilcoxon statistical test methods. According to the results of the analysis, it has been seen that the FDB-SMA variations outperform the basic algorithm (SMA) in all experimental studies.Öğe A novel optimal power flow model for efficient operation of hybrid power networks(Pergamon-Elsevier Science Ltd, 2023) Bakir, Huseyin; Duman, Serhat; Guvenc, Ugur; Kahraman, Hamdi TolgaIn the power industry, the design of an efficient optimal power flow (OPF) model is one of the important research challenges. This study presents the formulation and solution of the OPF problem in the presence of RESs, VSC-MTDC transmission lines, and FACTS devices, simultaneously. Fuel cost, voltage deviation, and power loss were selected as OPF objectives and optimized with state-of-the-art metaheuristic algorithms such as MFO, BSA, COA, MRFO, TLABC, and FDB-TLABC. Based on the optimization results, FDB-TLABC has obtained the best fuel cost results of 785.5850 $/h, 815.1251 $/h, and 820.3022 $/h on the IEEE 30-bus power network. Besides, the algorithm reduced voltage deviation and active power loss by 11.76% and 0.52% compared to TLABC which second most successful algorithm, respectively. The experimental results are statistically analyzed using Wilcoxon signed-rank test. The analysis results show that the FDB-TLABC is a robust and powerful method to solve the introduced OPF problem.Öğe Optimal operation and planning of hybrid AC/DC power systems using multi-objective grasshopper optimization algorithm(Springer London Ltd, 2022) Bakır, Hüseyin; Güvenç, Uğur; Kahraman, Hamdi TolgaOptimal power flow (OPF) in a hybrid alternating current and multi-terminal high-voltage direct current (AC-MTHVDC) grid is currently one of the most popular optimization problems in modern power systems. The critical necessity of addressing global warming and reducing generation costs is encouraging the integration of eco-friendly renewable energy sources (RESs) into the OPF problem. In this direction, the present research has centred on the formulation and solution of the multi-objective (MO) AC-MTHVDC-OPF problem incorporating RESs such as wind, solar, small-hydro, and tidal power. The available power of RESs is calculated by means of the Weibull, lognormal, and Gumbel probability density functions. The proposed MO-OPF optimizes the double and triple configurations of various objective functions, including total cost, the total cost with the valve-point effect, the total cost with emission and carbon tax, voltage deviation, and power loss. Multi-objective grasshopper optimization algorithm (MOGOA) is applied to find non-dominated Pareto-optimal solutions of the non-convex, nonlinear and high-dimensional MO/AC-MTHVDC-OPF problem. The obtained results are compared with the results of MSSA, MODA, MOALO, and MO_Ring_PSO_SCD algorithms. The comparison of results gives that MOGOA outperforms competitive optimizers with respect to the quality of Pareto-optimal solutions and their distribution.Öğe Optimal Power Flow for Hybrid AC/DC Electrical Networks Configured With VSC-MTDC Transmission Lines and Renewable Energy Sources(Ieee-Inst Electrical Electronics Engineers Inc, 2023) Bakir, Huseyin; Guvenc, Ugur; Duman, Serhat; Kahraman, Hamdi TolgaThis article presents the single objective optimal power flow (OPF) formulation incorporating both renewable energy sources, and voltage source converter-based multiterminal direct current transmission lines, simultaneously. To solve the formulated OPF problem, powerful metaheuristic optimization algorithms including adaptive guided differential evolution, marine predators algorithm, atom search optimization, stochastic fractal search (SFS), and fitness-distance balance-based SFS (FDB-SFS) are employed. The performance of the algorithms is tested for the minimization of fuel cost, pollutant emissions of thermal generators, voltage deviation, and active power loss in a modified IEEE 30-bus power network. The simulation results give that FDB-SFS achieved the best results on the fuel cost (786.5361 $/h), the fuel cost with valve point effect (815.6644 $/h), and the fuel cost with emission-carbon tax (820.5991 $/h). In addition, FDB-SFS reduced voltage deviation and active power loss values by 14.2587% and 6.7438% compared to SFS. The nonparametric Wilcoxon and Friedman statistical test results confirmed that FDB-SFS is an effective and robust algorithm that can be used in the optimization of the introduced OPF problem.Öğe Optimal Scheduling of Short-term Hydrothermal Generation Using Symbiotic Organisms Search Algorithm(Ieee, 2016) Kahraman, Hamdi Tolga; Döşoğlu, Mehmet Kenan; Güvenç, Uğur; Duman, Serhat; Sönmez, YusufIn this study, the Symbiotic Organisms Search (SOS) algorithm is proposed to solve the short-term hydrothermal generation scheduling (STHGS) problem. This problem aims to optimize the power generation strategy produced by hydroelectric and thermal plants by minimizing the total fuel cost function while satisfying some operational constraints. In order to evaluate the effectiveness of the SOS, it has been tested on a system having a hydro plant with four-cascaded reservoir and a thermal plant. Results have been compared other meta-heuristic methods. Results obtained from the experiment show that the proposed algorithm produces better results than the other methods and shows a good convergence.Öğe Optimal solution of the combined heat and power economic dispatch problem by adaptive fitness-distance balance based artificial rabbits optimization algorithm(Pergamon-Elsevier Science Ltd, 2024) Ozkaya, Burcin; Duman, Serhat; Kahraman, Hamdi Tolga; Guvenc, UgurCombined heat and power economic dispatch (CHPED) problem is one of the most widely handled, optimization problem by researchers in modern power systems. CHPED problem is a complicated, non-continuous, and nonconvex optimization problem due to the constraints. Moreover, considering the valve-point loading effect (VPLE), transmission losses (TLs), and prohibited operating zones (POZs) of power-only units as constraints, the complexity of CHPED problem increases. Therefore, a powerful optimization algorithm needs to be introduced to find global solution that meets all constraints. In this paper, a novel adaptive fitness-distance balance based artificial rabbits optimization (AFDB-ARO) is developed to solve CHPED problems. AFDB-based guiding mechanism was implemented to enhance the exploration capability of ARO and to strengthen exploitation-exploration balance. A comprehensive experimental study was realized to prove the performance of the proposed algorithm on the CHPED and benchmark problems. In experimental study between AFDB-ARO variants and ARO on 40 benchmark problems, according to Wilcoxon analysis results, all AFDB-ARO variants outperformed the base ARO, and the best AFDB-ARO variant won victory in 20 of 40 problem and achieved similar results in other 20 problem. In other experimental study, AFDB-ARO algorithm was implemented on the CHPED systems with 4-, 5-, 7-, 24-, 48-, 96-, and 192-units, and fifteen case studies were considered using these systems, VPLE, TLs, and POZs. One of the important points of this study was that POZs were considered for the first time in 96-and 192 -units system. The results show that AFDB-ARO achieved the best optimal solution in ten of fifteen cases, was same in one case, and obtained almost same results in four cases compared to the literature. Moreover, the stability of the AFDB-ARO and base ARO algorithms in solving the CHPED problem were tested by performing stability analysis. While the mean success rate, mean iteration number, and mean search time were obtained 87.62%, 353.63, and 2.91 sec of AFDB-ARO, respectively, ARO managed to find the optimal solution in two cases. Thus, the superior performance of AFDB-ARO algorithm is confirmed by experimental studies and analysis against ARO algorithm. The source codes of the AFDB-ARO algorithm (proposed method) can be accessed at this link: https://www.mathworks.com/matlabcentral/fileexchange/136846-afdb-aro-an-improved-aro-algorithm-for-optimization-problem.Öğe A powerful meta-heuristic search algorithm for solving global optimization and real-world solar photovoltaic parameter estimation problems(Pergamon-Elsevier Science Ltd, 2022) Duman, Serhat; Kahraman, Hamdi Tolga; Sönmez, Yusuf; Güvenç, Uğur; Kati, Mehmet; Aras, SefaThe teaching-learning-based artificial bee colony (TLABC) is a new hybrid swarm-based metaheuristic search algorithm. It combines the exploitation of the teaching learning-based optimization (TLBO) with the exploration of the artificial bee colony (ABC). With the hybridization of these two nature-inspired swarm intelligence algorithms, a robust method has been proposed to solve global optimization problems. However, as with swarm-based algorithms, with the TLABC method, it is a great challenge to effectively simulate the selection process. Fitness-distance balance (FDB) is a powerful recently developed method to effectively imitate the selection process in nature. In this study, the three search phases of the TLABC algorithm were redesigned using the FDB method. In this way, the FDB-TLABC algorithm, which imitates nature more effectively and has a robust search performance, was developed. To investigate the exploitation, exploration, and balanced search capabilities of the proposed algorithm, it was tested on standard and complex benchmark suites (Classic, IEEE CEC 2014, IEEE CEC 2017, and IEEE CEC 2020). In order to verify the performance of the proposed FDB-TLABC for global optimization problems and in the photovoltaic parameter estimation problem (a constrained real-world engineering problem) a very comprehensive and qualified experimental study was carried out according to IEEE CEC standards. Statistical analysis results confirmed that the proposed FDB-TLABC provided the best optimum solution and yielded a superior performance compared to other optimization methods.Öğe Rulet Elektromanyetik Alan Optimizasyon (R-EFO) Algoritması(2020) Kahraman, Hamdi TolgaMeta-sezgisel optimizasyon algoritmalarının yerel arama performansları üzerinde etkili olan iki temel öğe seçimyöntemleri ve arama operatörleridir. Bu makale çalışmasında olasılıksal bir seçim yöntemi olan rulet tekerleğiningüncel bir meta-sezgisel arama tekniği olan elektromanyetik alan optimizasyon (electromagnetic fieldoptimization, EFO) algoritmasının yerel arama performansı üzerindeki etkisi araştırılmaktadır. Elektromanyetikoptimizasyon algoritmasında çözüm adayları topluluğu uygunluk değerlerine bağlı olarak pozitif, nötr ve negatifalanlara ayrılmaktadır. Bu üç alandan seçilen çözüm adayları ise arama sürecine rehberlik etmektedirler. Busüreçte çözüm adayları açgözlü ve rastgele seçim yöntemleri ile belirlenmektedir. Bu makale çalışmasında isenegatif alandan çözüm adaylarının seçimi için rulet tekniği kullanılmaktadır. Deneysel çalışmalarda literatürdekien güncel sürekli değer problemleri olan CEC17 test seti kullanılmıştır. Deneysel çalışma sonuçları istatistikselolarak ikili karşılaştırmalarda kullanılan wilcoxon runk sum test ile analiz edilmiştir. Analiz sonuçlarına göre ruletseçim yöntemi EFO algoritmasının arama performansını kayda değer şekilde artırmaktadır.Öğe Yapay Sinir Ağları ve K-Ortalamalar Tabanlı Büyük Veri Azaltma Algoritmasının Tasarımı ve Uygulaması(2021) Kahraman, Hamdi Tolga; Temel, SeyithanBüyük veri azaltma sürecinde karşılaşılan başlıca zorluk, veri setinin homojenliğinin ve problem uzayını temsil yeteneğinin korunmasıdır. Bu durum, büyük veri setleri üzerinde yapılan modelleme çalışmalarında hesaplama karmaşıklığının yeterince azaltılamamasına, geliştirilen modelin orijinal veri setine dayalı olarak geliştirilen modele kıyasla kararlılık ve doğruluk performansının önemli ölçüde azalmasına neden olmaktadır. Bu makale çalışmasının amacı, büyük veri setleri için kararlı ve etkili bir şekilde çalışan veri azaltma algoritması geliştirmektir. Bu amaçla, yapay sinir ağları (YSA) tabanlı problem modelleme modülü ve K-ortalamalar tabanlı veri azaltma modülünden oluşan melez bir algoritma geliştirilmiştir. Problem modelleme modülü, büyük veri seti için performans eşik değerlerini tanımlamayı sağlamaktadır. Bu sayede, orijinal veri setinin ve veri azaltma işlemi uygulanmış veri setlerinin problem uzayını temsil yetenekleri ve kararlılıkları analiz edilmektedir. Kortalamalar modülünün görevi ise, veri uzayını K-adet kümede gruplamayı ve bu grupların her biri için küme merkezini referans alarak kademeli olarak veri (gözlem) azaltma işlemini gerçekleştirmektir. Böylelikle, Kortalamalar modülü ile veri azaltma işlemi uygulanırken, azaltılmış veri setlerinin performansı ise YSA modülü ile test edilmekte ve performans eşik değerlerini karşılama durumu analiz edilmektedir. Geliştirilen melez veri azaltma algoritmasının performansını test etmek ve doğrulamak amacıyla UCI Machine Learning uluslararası veri havuzunda yer alan üç farklı veri seti kullanılmıştır. Deneysel çalışma sonuçları istatistiksel olarak analiz edilmiştir. Analiz sonuçlarına göre büyük veri setlerinde kararlılık ve performans kaybı yaşanmadan %30-%40 oranları arasında veri azaltma işlemi başarılı bir şekilde gerçekleştirilmiştir.
