Predicting compressive strength using the texture coefficient with soft computing techniques for rocks
Küçük Resim Yok
Tarih
2022
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Rock strength plays one of the most dominant roles for mining, geology, and civil engineering in terms of planning, excavation, and safety. Compressive strength (fc), which is the most used strength type, requires time, cost, and standard size specimens are needed to find it in the laboratory. In this study, Regression Analysis (RA), Neural Networks (NNs), Gene-Expression Programming (GEP), and Adaptive Network-based Fuzzy Inference System (ANFIS) were used for predicting using both textural and mechanical properties which are detected with a dimensionless sample or directly in the field. For this purpose, a data set consists of 136 data value (46 magmatic, 77 sedimentary and 13 metamorphic rocks) was used, and three different feature sets were constructed. The comparison of the estimated results with each other was performed by training, testing, and checking of these models. The comparisons and results of the statistical analyses indicate that soft computing techniques represent significantly effective methods to calculate fc even in situations when input and output values are not related to each other, and it is possible to create statistically suitable and valid mathematical models by everyone using GEP.
Açıklama
Anahtar Kelimeler
Jeoloji, Neural networks, Texture coefficient, Compression strength, Adaptive network-based fuzzy inference system, Gene-expression programming.
Kaynak
Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi
WoS Q Değeri
Scopus Q Değeri
Cilt
11
Sayı
4
