Search

Over the years, increasing energy demands with the growth of the population and the development of technology have caused more fossil fuel consumption. Besides, environmental pollution and climate change, which are vital importance for humanity, are encountered. In order to avoid these dangerous situations, people have started to turn to clean and renewable energy sources such as wind energy. Due to the rapid development of such situations, it is very important to obtain information on the determination of the regions where wind energy facility will be installed and the characteristics of the wind speed. Wind power estimation can be made through various statistical distributions used to explain the characteristics of wind speed data. Rayleigh, Weibull, Nakagami, Gamma, Logistic, Loglogistic, Lognormal and Burr Type XII distributions, which are frequently used in the wind energy literature, are discussed in this study and the performances of the specified distributions are compared through the data sets obtained from the stations in Istanbul from Marmara region. One of the most preferred methods in estimation problems is the maximum likelihood method, and a differential evolution algorithm is proposed for ML estimation of the parameters of the distributions examined in the study. In addition, various model selection criteria are also utilized to determine the distribution that best fits the wind speed data. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Polytechnic is the property of Journal of Polytechnic and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Polytechnic is the property of Journal of Polytechnic and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Weibull distribution is widely used in various fields such as failure rates and system reliability. Because of its wide application, it is very important to determine the best method of parameter estimation for the Weibull distribution. Although the most known parameter estimation methods are least square (LS), weighted least square (WLS) and maximum likelihood (ML), these methods have different properties and can give different estimation results. In this study, in which it is aimed to obtain better estimation results by evaluating these methods together, it is recommended to use the multi-objective programming approach. The non-dominated sorting genetic algorithm II method, which is a multi-objective heuristic approach, was used to solve the created multi-objective programming estimation model. The Kevlar 49 / Epoxy dataset was used to demonstrate the applicability of the proposed approach. According to the results, the best parameter estimation results were given in cases where the ML method was evaluated only and together, LS-ML and WLS-ML multi-objective parameter estimation models. [ABSTRACT FROM AUTHOR]

The ranking of the decision making units (DMUs) is an essential problem in data envelopment analysis (DEA). Numerous approaches have been proposed for fully ranking of units. Majority of these methods consider DMUs with optimistic approach, whereas their weaknesses are ignored. In this study, for fully ranking of the units, a modified optimistic–pessimistic approach, which is based on game cross efficiency idea is proposed. The proposed game like iterative optimistic-pessimistic DEA procedure calculates the efficiency scores according to weaknesses and strengths of units and is based on non-cooperative game. This study extends the optimistic-pessimistic DEA approach to obtain robust rank values for DMUs. The proposed approach yields Nash equilibrium solution, thus overcomes the problem of non-uniqueness of the DEA optimal weights that can possibly reduce the usefulness of cross efficiency. Finally, in order to verify the validity of the proposed model and to show the practicability of algorithm, we apply a real-world example for selection of industrial R&D projects. The proposed model can increase the discriminating power of DMUs and can fully rank the DMUs. [ABSTRACT FROM AUTHOR]

Backpropagation algorithm is a classical technique used in the training of the artificial neural networks. Since this algorithm has many disadvantages, the training of the neural networks has been implemented with various optimization methods. In this paper, a hybrid intelligent model, i.e., hybridGSA (hybrid Genetic Algorithm and Simulated Annealing), is developed for training artificial neural networks (ANN) and undertaking data classification problems. The hybrid intelligent system aims to exploit the advantages of genetic and simulated annealing algorithms and, at the same time, alleviate their limitations. To evaluate the effectiveness of the hybridGSA method, three benchmark data sets, i.e., Breast Cancer Wisconsin, Pima Indians Diabetes, and Liver Disorders from the UCI Repository of Machine Learning, and a simulation experiment are used for evaluation. A comparative analysis on the real data sets and simulation data show that the hybridGSA algorithm may offer efficient alternative to traditional training methods for the classification problem. [ABSTRACT FROM AUTHOR]

The cross-efficiency evaluation (CEE) method, which was developed as a contribution to classical Data Envelopment Analysis (DEA), has been successively used to solve problems involving unrealistic weight distribution as well as problems that do not require prior information for ranking the decision making units (DMUs). Originally, the CEE method included the efficiency evaluations that were obtained for a DMU by the classical DEA for the reuse of optimal weights in the other DMUs. As the optimal weights in the classical DEA solutions usually have multiple solutions, this reduces the usefulness of the CEE method. Hence, this study suggests a new technique that could be used in the second stage of the CEE method for removing the problem of multiple optimal weights and for determining the reasonable ranks of DMUs. The performance of the proposed model is examined on real data set relative to the efficiencies of Turkey cities. [ABSTRACT FROM AUTHOR]

In this study, a multi-criteria data envelopment analysis (MCDEA) model, used in the literature to moderate the homogeneity of weights dispersion, is solved using pre-emptive goal programming. The MCDEA model solved using pre-emptive goal programming gives the same relative efficiency as the classical DEA model while it improves the homogeneity of input-output weights. This conclusion is confirmed by the computational results obtained when the two models are applied to a real data set relative to the socio-economic performances of European countries and to randomly generated instances with various numbers of decision making units, inputs and outputs. [ABSTRACT FROM AUTHOR]

In this study, a multi-criteria data envelopment analysis (MCDEA) model, used in the literature to moderate the homogeneity of weights dispersion, is solved using pre-emptive goal programming. The MCDEA model solved using pre-emptive goal programming gives the same relative efficiency as the classical DEA model while it improves the homogeneity of input-output weights. This conclusion is confirmed by the computational results obtained when the two models are applied to a real data set relative to the socio-economic performances of European countries and to randomly generated instances with various numbers of decision making units, inputs and outputs. [ABSTRACT FROM AUTHOR]