Yazar "Beskirli, Mehmet" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    A comparative study of Improved Bat Algorithm and Bat Algorithm on numerical benchmarks
    (IEEE, 2015) Beskirli, Mehmet; Koc, Ismail
    Optimization is employed in solutions of many problems today. Optimization is described as finding the most suitable alternative among many others under the given constraints. Meta-heuristic algorithms used in solutions of the problems are developed upon the behaviors of living creatures in the nature. One of these is Bat Algorithm (BA), an optimization method based on swarm intelligence. BA is a numerical optimization technique developed in recent times. In this paper, it is aimed at improving Bat Algorithm (IBA) by using Differential Evolution Algorithm population strategy instead of population generation method of BA. IBA was tested on 17 benchmark functions with different characteristics. Suggested method has been seen to exhibit better results compared to the original BA.
  • Küçük Resim Yok
    Öğe
    The energy demand estimation for Turkey using differential evolution algorithm
    (SPRINGER INDIA, 2017) Beskirli, Mehmet; Hakli, Huseyin; Kodaz, Halife
    The energy demand estimation commands great importance for both developing and developed countries in terms of the economy and country resources. In this study, the differential evolution algorithm ( DE) was used to forecast the long-term energy demand in Turkey. In addition to being employed for solving regular optimization problems, DE is also a global, meta-heuristic algorithm that enables fast, reliable and operative stochastic searches based on population. Considering the correlation between the increase in certain economic indicators in Turkey and the increase of energy consumption, two equations were used-one applying the linear form and the other the quadratic form. Turkey's long-term energy demand from 2012 to 2031 was estimated through the DE method in three different scenarios and in terms of the gross domestic product, import, export and population. To prove the success of the DE method in addressing the energy demand problem, the DE method was compared to other methods found in the literature. Results showed that the proposed DE method was more successful than the other methods. Furthermore, the future projections of energy demand obtained using the proposed method were compared to the indicators of energy demand estimated and observed by the Ministry of Energy and Natural Resources.
  • Küçük Resim Yok
    Öğe
    A new optimization algorithm for solving wind turbine placement problem: Binary artificial algae algorithm
    (PERGAMON-ELSEVIER SCIENCE LTD, 2018) Beskirli, Mehmet; Koc, Ismail; Hakli, Huseyin; Kodaz, Halife
    The wind turbine has grown out to be one of the most common renewable energy sources around the world in recent years. As wind energy becomes more important, the significance of wind turbine placement also increases. This study was intended to position the wind turbines on a wind farm to achieve the highest performance possible. The turbine placement operation was designed for a 2 km x 2 km area. The surface of the area was calculated by dividing it into a 10 x 10 grid and a 20 x 20 grid with the use of binary coding. The calculation revealed ten different new binary algorithms using ten different transfer functions of the Artificial Algae Algorithm (AAA) that has been successfully applied to solve continuous optimization problems. These algorithms were applied to the turbine placement problem, and the algorithm that obtained the best result was called the Binary Artificial Algorithm (BinAAA). The results of the proposed algorithm for the binary turbine placement optimization problem were compared with those of other well-known algorithms in the relevant literature. The algorithm that was proposed in the study is an efficient algorithm for the placement problem of wind turbines since it optimized the binary search space and achieved the most successful result (C) 2017 Elsevier Ltd. All rights reserved.