Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi
Yazarlar: Recep Koray KIYILDI
Konular:İnşaat Mühendisliği
DOI:10.28948/ngumuh.866566
Anahtar Kelimeler:Asphalt concrete,Marshall test,Artificial neural networks
Özet: In this article, Marshall stability tests carried out in Niğde-Adana highway construction site were used. The results of the experiments performed on the core sample were modelled by Artificial Neural Networks (ANN). In Marshall stability test performed on the specimen of asphalt concrete mixture; the Marshall stability values, which are determined by the percentage of bitumen according to the weight of aggregate, the percentage by weight of the bitumen in the mixture, the volume specific gravity and the void values, were estimated by using the ANN method. When the prediction model obtained as a result of the training was inspected with the previously separated experimental results, it was determined that there was a good relationship between the Marshall stability values estimated by the ANN model and the experimentally obtained values.