Yazar "Bilgil, A." seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • Küçük Resim Yok
    Öğe
    Investigation of flow resistance in smooth open channels using artificial neural networks
    (ELSEVIER SCI LTD, 2008) Bilgil, A.; Altun, H.
    An accurate prediction of the friction coefficient is very important in hydraulic engineering since it directly affects the design of water structures, the calculation of velocity distribution, and an accurate determination of energy losses. However, conventional approaches that are profoundly based on empirical methods lack in providing high accuracy for the prediction of the friction coefficient. Consequently, new and accurate techniques are still highly demanded. This study introduces an efficient approach to estimate the friction coefficient via an artificial neural network, which is a promising computational tool in civil engineering. The estimated value of the friction coefficient is used in Manning Equation to predict the open channel flows in order to carry out a comparison between the proposed neural networks based approach and the conventional ones. Results show that the proposed approach is in good agreement with the experimental results when compared to the conventional ones. (C) 2008 Elsevier Ltd. All rights reserved.
  • Küçük Resim Yok
    Öğe
    The effect of alkali activation solutions with different water glass/NaOH solution ratios on geopolymer composite materials
    (Iop Publishing Ltd, 2019) Dogan-Saglamtimur, N.; Oz, H. Oznur; Bilgil, A.; Vural, T.; Suzgec, E.
    In this study, geopolymer materials were produced from fly ash (FA) supplied from Isken Sugozu Thermal Power Plant located in Adana, Turkey. FA and Rilem Cembureau Standard Sand were used together with the ratio of 0.50. At first, two different alkaline solution/material ratio (FA+standard sand) (L/M) were selected as 0.20 and 0.40 for the design parameters. In the production of geopolymer composite material, sodium silicate (Na2SiO3) and sodium hydroxide (12 M NaOH) were used together within the ratio of 1, 1.5, 2, 2.5 and 3 by weight, respectively. A totally of 20 mixes were cured at 70 and 100 C for 24 hrs, respectively and thereafter kept in room temperature until testing age. Physical characteristics of hardened mortar were determined via the bulk density, water absorption and porosity values at 28 days while the strength of geopolymers was obtained on the results of compressive strength and flexural strength tests conducted at 7, 28 and 90 days. Considering the testing parameters, geopolymer material with the highest compressive strength was found as 76.0 MPa (28-days) on the mixture produced with L/M ratio of 20% by weight; the alkaline solution consisted of a mixture of Na2SiO3 and 12 M NaOH in weight ratio of 2 by curing at 70 degrees C for 24 hrs. However, test results showed that there was an optimum limit for the alkaline solution ratio, such that exceeding this limit gave the reverse effect for the strength characteristics of the geopolymer material.
  • Küçük Resim Yok
    Öğe
    Treatment of multi-dimensional data to enhance neural network estimators in regression problems
    (PERGAMON-ELSEVIER SCIENCE LTD, 2007) Altun, H.; Bilgil, A.; Fidan, B. C.
    This paper proposes and explains a data treatment technique to improve the accuracy of a neural network estimator in regression problems, where multi-dimensional input data set is highly skewed and non-normally distributed. The proposed treatment modifies the distribution characteristics of the data set. The prediction of the suspended sediment, which is an important problem in river engineering applications, will be considered as a case study. Conventional approaches lack in providing high accuracy due to the inherently employed simplicity in order to obtain empirical formulae. On the other hand, artificial neural networks are able to model the non-linear characteristics of the mechanism of the sediment transport and have a growing body of applications in diverse applications in civil engineering. It will be shown that a significant enhancement and superior score in accuracy, compared with the classical approaches, are obtainable when the proposed treatment is employed. The proposed technique is an extension to the understanding of the practical aspects of neural computing applications. Therefore the outcome of the present study is important as it is applicable to any scenario where neural network approaches are involved. (C) 2006 Elsevier Ltd. All rights reserved.
  • Küçük Resim Yok
    Öğe
    Treatment of skewed multi-dimensional training data to facilitate the task of engineering neural models
    (PERGAMON-ELSEVIER SCIENCE LTD, 2007) Altun, H.; Bilgil, A.; Fidan, B. C.
    Successful application of neural network models relies heavily on problem-dependent internal parameters. As the theory does not facilitate the choice of the optimal parameters of neural models, these can solely be obtained through a tedious trial-and-error process. The process requires performing multiple training simulations with various network parameters, until satisfactory performance criteria of a neural model are met. In literature, it has been shown that neural models are not consistently good in prediction under highly skewed data. Consequently, the cost of engineering neural models rises in such circumstance to seek for appropriate internal parameters. In this paper the aim is to show that a recently proposed treatment of highly skewed data eases the task of practitioners in engineering neural network models to meet satisfactory performance criteria. As the applications of neural models grows dramatically in diverse engineering domains, the understanding of the treatment show indispensable practical values. (c) 2006 Elsevier Ltd. All rights reserved.