ÖNGERİLMELİ I KESİTLİ SARGILI BETONARME KİRİŞLERİN KESME KAPASİTELERİNİN TAHMİNİNE YÖNELİK BİR ARAŞTIRMA
Yükleniyor...
Dosyalar
Tarih
2017
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Selçuk Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Betonarme öngerilmeli beton kirişler, genellikle kullanım yükleri altında eğilme ve kesme kuvveti etkisindedir. Bazı durumlarda kiriş, eğilme kapasitesine ulaşmadan kesme etkisi altında da göçmektedir. Kirişlerin kesme etkisinden zarar görmemesi ve kirişin eğilme kapasitesine tam olarak ulaşabilmesi için kesme kapasitesinin doğru olarak belirlenmesi önemlidir. Bu çalışmada öngerilmeli betonarme kirişlerin kesme kapasitelerinin belirlenmesi için akıllı sistem tabanlı analitik bir çalışma gerçekleştirilmiştir. Çalışmada ilk olarak öngerilmeli betonarme kirişlerin kesme kapasitelerinin teorik hesaplarından bahsedilmiştir. Daha sonra ise bu konuda yapılan deneysel çalışmalar derlenerek deneyler hakkında açıklamalar ve deneylerin gruplandırmaları yapılmıştır. Kesme kapasitesinin belirlenmesinde oldukça fazla parametrenin etkisinin olması bu yorumlamaları zorlaştırmış olsa da çalışmanın bir sonraki ayağı olan akıllı sistem modellemesi ile kesme kapasitesinin belirlenmesi için önemli bir deneysel ve analitik altyapı oluşturulmuştur. Akıllı bir sistem türü olan yapay sinir ağları (YSA) hakkında bilgi verilerek analiz yöntemi tanıtılmış ve derlenen deneysel veri setinde yer alan betonarme ve öngerilmeli beton kirişlerin MATLAB paket programında YSA modellemeleri yapılmıştır. Modellemelerde kullanılan kirişlere ait parametreler doğrultusunda mevcut yönetmeliklere göre kiriş kesme kapasiteleri de ayrıca belirlenmiştir. Oluşturulan YSA modelleri ile mevcut tasarım ve yapım yönetmeliklerinin kiriş kesme kapasitesinin belirlenmesindeki performansları karşılaştırılmalı olarak değerlendirilmiştir. Yapılan karşılaştırmalarda YSA’nın yönetmelik yaklaşımlara göre oldukça yüksek bir tahmin oranına sahip olduğu görülmüştür.
Prestressed reinforced concrete beams are generally bending and shear under serfice loads. In some cases, the beam falls under the effect of shear without reaching the bending capacity. Accurate determination of the shear capacity becomes important so that the beams are not damaged by the shear effect. In addition, the shear capacity of the beam must be correctly determined so that the beam can fully reach its bending moment capacity. In this study, an intelligent system based analytical study was carried out to determine the shear capacities of prestressed reinforced concrete beams. In the study, the theoretical calculations of the shear capacities of prestressed concrete beams were first mentioned and then the experimental works carried out in this subject were compiled and groupings of explanations and experiments were made about the experiments. It is examined how the shear capacity changes according to each parameter according to the present experimental data. Although a lot of parametric effects have been made in determining the shear capacity, these interpretations have been difficult, but an important experimental and analytical infrastructure has been established for the determination of shear capacity by intelligent system modeling, which is the next step of the study. The analysis method is introduced by giving information about artificial neural networks (ANN) which is an intelligent system type and ANN models are made in the MATLAB package program of reinforced concrete and prestressed concrete beams in the complied experimental data set. Shear capacities of the beams are also determined according to the present regulations in line with the parameters of beams used in the models. The performance of the ANN models generated and the determination of the beam shear capacity of the present regulations have been evaluated comparatively. In the comparisons made, it was seen that ANN has a fairly high prediction ratio in regulatory approaches. Moreover, the success of ANN in predicting reinforced concrete beams is higher than the regulations norms.
Prestressed reinforced concrete beams are generally bending and shear under serfice loads. In some cases, the beam falls under the effect of shear without reaching the bending capacity. Accurate determination of the shear capacity becomes important so that the beams are not damaged by the shear effect. In addition, the shear capacity of the beam must be correctly determined so that the beam can fully reach its bending moment capacity. In this study, an intelligent system based analytical study was carried out to determine the shear capacities of prestressed reinforced concrete beams. In the study, the theoretical calculations of the shear capacities of prestressed concrete beams were first mentioned and then the experimental works carried out in this subject were compiled and groupings of explanations and experiments were made about the experiments. It is examined how the shear capacity changes according to each parameter according to the present experimental data. Although a lot of parametric effects have been made in determining the shear capacity, these interpretations have been difficult, but an important experimental and analytical infrastructure has been established for the determination of shear capacity by intelligent system modeling, which is the next step of the study. The analysis method is introduced by giving information about artificial neural networks (ANN) which is an intelligent system type and ANN models are made in the MATLAB package program of reinforced concrete and prestressed concrete beams in the complied experimental data set. Shear capacities of the beams are also determined according to the present regulations in line with the parameters of beams used in the models. The performance of the ANN models generated and the determination of the beam shear capacity of the present regulations have been evaluated comparatively. In the comparisons made, it was seen that ANN has a fairly high prediction ratio in regulatory approaches. Moreover, the success of ANN in predicting reinforced concrete beams is higher than the regulations norms.
Açıklama
Anahtar Kelimeler
Betonarme kiriş, kiriş kesme kapasitesi, öngerilmeli beton, yapay sinir ağları, Artificial neural networks, prestressed concrete, reinforced concrete beam, shear capacity of beam
Kaynak
Selcuk University Journal of Engineering Sciences
WoS Q Değeri
Scopus Q Değeri
Cilt
16
Sayı
3
Künye
Sertkaya, M. S., Arslan, M. H., (2017). ÖNGERİLMELİ I KESİTLİ SARGILI BETONARME KİRİŞLERİN KESME
KAPASİTELERİNİN TAHMİNİNE YÖNELİK BİR ARAŞTIRMA. Selcuk University Journal of Engineering Sciences, 16 (3) 104-122.
