Yazar "Cinar, Ahmet Cevahir" seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • Küçük Resim Yok
    Öğe
    APPLICATION OF ARTIFICIAL NEURAL NETWORK FORECASTING OF DAILY MAXIMUM TEMPERATURE IN KONYA
    (BRNO UNIV TECHNOLOGY VUT PRESS, 2011) Tasdemir, Sakir; Cinar, Ahmet Cevahir
    Weather forecast is one of the most effective factors on human beings and other living creatures. Maximum air temperature is one of the most important parameters to be estimated for meteorology, because the maximum and minimum temperature data is the outlook of the institution and the most interesting aspect of weather forecast presentations. Many meteorological variables play an important role in estimating the lowest and highest temperature of the day. Today, numerical models are mainly used in weather forecasting. The incredible success of Artificial Neural Networks (ANN) in classification and estimation makes it necessary to use this approach in the area of meteorology. Apart from known methods, ANN, which is an artificial intelligence technique, was used to forecast maximum temperature, which is the modeling of a non-linear process. In this study, the data for the years 2008 and 2009 was used that were obtained from the Turkish Meteorological Data Archive System and The Directorate of Konya Airfield Meteorology Station, which are the institutions of the General Directorate of Turkish State Meteorological Services. The developed ANN has 6 inputs and 1 output. The six input variables were respectively the temperature at 850 hpa level (t(850)-degrees C), daily average actual pressure (P-mb), daily minimum temperature (t(min)-degrees C), daily mean temperature (t(mean)-degrees C), daily average relative humidity (H-%) and daily sunshine duration (SD-hour). The output parameter value was the daily maximum temperature (t(max)-degrees C). Feed-forward back-propagation ANN model was used in this study. Levenberg-Marquardt (trainlm) training algorithm and Hyperbolic Tangent Sigmoid (tansig) and Logarithmic Sigmoid (logsig) transfer function were tried in the software developed in MATLAB and the results were obtained. The study put forth that accuracy rates and mean absolute error (MSE) obtained from training and test operations can be used in determining the maximum air temperature in the generated model.
  • Küçük Resim Yok
    Öğe
    Boundary Conditions in Tree-Seed Algorithm Analysis of the success of search space limitation techniques in Tree-Seed Algorithm
    (IEEE, 2017) Cinar, Ahmet Cevahir; Kiran, Mustafa Servet
    Swarm intelligence or evolutionary computation algorithms search possible solutions in a predetermined search space of an optimization problem. In some cases candidate solutions go out of search space during the search. In such situations, search space limitation techniques or boundary conditions are used for sway up this outcast individual into search space. The boundary conditions are classified two main categories whose names are restricted and unrestricted techniques. Restricted boundary conditions forces outcast individual into search space but unrestricted boundary conditions does not force. In this work we use four restricted boundary conditions whose names are Absorbing, Reflecting, Damping and Randomly. Additionally, three unrestricted boundary conditions (Invisible, Invisible Reflecting and Invisible Damping) are used in the study. These boundary conditions are applied in Tree-Seed Algorithm (TSA). The test material is five standard benchmark functions and these arc Sphere, Rastrigin, Rosenbrock, Griewank and Ackley. The main idea of this study is to investigate whether there is a significant difference among the limitation methods in TSA. Experimental results show that there is no significant difference among the boundary conditions methods for TSA.
  • Küçük Resim Yok
    Öğe
    A modification of tree-seed algorithm using Deb's rules for constrained optimization
    (ELSEVIER SCIENCE BV, 2018) Babalik, Ahmet; Cinar, Ahmet Cevahir; Kiran, Mustafa Servet
    This study focuses on the modification of Tree-Seed Algorithm (TSA) to solve constrained optimization problem. TSA, which is one of the population-based iterative search algorithms, has been developed by inspiration of the relations between trees and seeds grown on a land, and the basic version of TSA has been first used to solve unconstrained optimization problems. In this study, the basic algorithmic process of TSA is modified by using Deb's rules to solve constrained optimization problems. Deb's rules are based on the objective function and violation of constraints and it is used to select the trees and seeds that will survive in next iterations. The performance of the algorithm is analyzed under different conditions of control parameters of the proposed algorithm, CTSA for short, and well-known 13 constrained maximization or minimization standard benchmark functions and engineering design optimization problems are employed. The results obtained by the CTSA are compared with the results of particle swarm optimization (PSO), artificial bee colony algorithm (ABC), genetic algorithm (GA) and differential evolution (DE) algorithm on the standard benchmark problems. The results of state-of-art methods are also compared with the proposed algorithm on engineering design optimization problems. The experimental analysis and results show that the proposed method produces promising and comparable results for the constrained optimization benchmark set in terms of solution quality and robustness. (C) 2017 Elsevier B.V. All rights reserved.
  • Küçük Resim Yok
    Öğe
    Similarity and Logic Gate-Based Tree-Seed Algorithms for Binary Optimization
    (PERGAMON-ELSEVIER SCIENCE LTD, 2018) Cinar, Ahmet Cevahir; Kiran, Mustafa Servet
    This paper focuses on solving binary optimization problems by using Tree-Seed Algorithm, TSA for short. While TSA is firstly proposed for solving optimization problems with continuously-structured solution space, TSA is modified to solve binary optimization problems, which is a subfield of discrete optimization, by using logic gates (LogicTSA) and similarity measurement techniques (SimTSA). In order to improve performance of these methods, a hybrid variant (SimLogicTSA) is also proposed. The performance of the proposed algorithms is investigated on uncapacitated facility location problems (UFLPs), which are pure binary optimization problems. The experimental results on 15 test instances are compared with each other and state-of-art algorithms. The comparisons demonstrate that hybrid variant of the algorithm is better than the other variants of the algorithm and state-of-art algorithms in terms of solution quality and robustness.