Prediction of footing settlements with geogrid reinforcement and eccentricity

Abstract

This study presents settlement predictions for footings with geogrid reinforcement and biaxial eccentricity using multi-linear regression (MLR) and artificial neural network (ANN) methods. The effects of central, uniaxial and biaxial eccentric loading conditions on embedded and non-embedded square footings in unreinforced and reinforced soils were investigated with laboratory model tests given in the first part of the study. Variations in the bearing capacity were determined through vertical load versus settlement curves drawn after each test. In the second part of this study, MLR and ANN models used to predict settlement were improved using independent variables related with the footings and geogrid. The results showed that fluctuations in the datasets of the settlement were very well reflected by the ANN methods.

Pull‐out capacity prediction of sustainable cementitious composites with artificial intelligence and statistical methods

“…Regression analysis is a method used to determine the reason-result relationship between variables. 36 MLR is used to modeling with multiple independent variables. The effect of independent variables on dependent variables is expressed by beta coefficients (β).…”

“…Thus, network learning is realized by improving the network with the reassigned weights in each layer. 36 In this study, ANN models with single hidden layer were created. Gradient descent backpropagation (GD), gradient descent with momentum backpropagation (GDM), Levenberg-Marquardt backpropagation (LM), and resilient backpropagation (RP) algorithms were used to train ANN models.…”

Yapay Zekâ ve İstatistiksel Yöntemler ile Küresel Ticarette Rekabet Ölçütü Olan Lojistik Performans İndeksine (LPI) Etken Parametrelerin Ülke Bazlı İncelenmesi ve Tahmin Modellerinin Geliştirilmesi

“…İstatistiksel bir yöntem olan regresyon analizi değişkenler arasındaki sebep-sonuç ilişkisinin belirlenmesinde kullanılmaktadır [22]. Bağımsız değişkenlerin bağımlı değişkenler üzerindeki etkisi katsayılar ile ifade edilir.…”

“…Bu işlemler ağdaki katman sayısına bağlı olarak tekrarlanmaktadır. Her katmanda ağ modeli geliştirmek için yeni ağırlıklar atanarak ağın öğrenmesi gerçekleştirilmektedir [22]. YSA yönteminin avantajlarının yanı sıra dezavantajları da bulunmaktadır.…”

Artificial Neural Networks Method for Prediction of Rainfall-Runoff Relation: Regional Practice

“…Thus, the basin, which will be measured, is examined with some equations with these accumulation systems and other effective parameters. Recently, artificial intelligence method is a Black Box model that is frequently used in modeling the groundwater level (Demirci et al (2017(Demirci et al ( , 2018a, Üneş et al (2018a), Kaya et al (2018)), suspended sediment (Demirci and Baltaci (2013), , Tasar et al ( 2017)), rainfallrunoff relationship (Ünes et al (2018b), Tasar et al ( 2019)), dam reservoir and lake level (Ünes and Demirci (2015), Ünes et al (2015, 2019a), Demirci and Kaya (2019)), density flow plunging (Üneş (2010)), dam reservoir volume (Unes et al (2017), Demirci et al (2018b), Üneş et al (2019b)), sand bar crest ), evaporation (Üneş et al (2018c(Üneş et al ( ), Tasar et al (2018, Kaya and Tasar (2019)), many different disciplines-areas (Cansiz et al (2009(Cansiz et al ( , 2017a(Cansiz et al ( , 2017b(Cansiz et al ( , 2017c, Dogan et al (2017), Cansiz (2007Cansiz ( , 2011, Cansiz and Easa (2011), Cansız and Askar (2018), Dal et al (2019)).…”