Investigation of SPT-CPT Relationship for Various Soil Types with Artificial Intelligence Support

Arda Burak Ekmen

Article Details

Investigation of SPT-CPT Relationship for Various Soil Types with Artificial Intelligence Support

Authors:

Arda Burak Ekmen

Publisher:

Erciyes University

Main Topics:

Mühendislik

Journal:

Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.39 Issue 2, pp.204-216

Document Type:

Article

Publication Date:

8/2023

Document Language:

Turkish

Document ID (Org No.):

2001297831

pISSN:

1012-2354

Keywords:

Abstract:

 Standard Penetration Test SPT and Cone Penetration Test CPT are among the most commonly used methods in soil investigations. Many soil parameters are associated with SPT andor CPT. It is essential to obtain a reliable correlation between SPT and CPT to use one of these tests individually. In this study, the tip resistance qc values obtained with CPT corresponding to SPT-N data for various soil types were reached utilizing the literature. Analysis of variance was performed to ascertain whether there was a significant relationship between SPT-N values and tip resistance qc data. Separate functions with high-order correlation values were obtained for each soil type by creating simulatable networks with artificial neural networks for SPT-CPT correlation. The obtained results were compared with experimental data and equations in the literature. As a result, functions in good correlation values created by the support of artificial intelligence for various soil types have provided the opportunity to correlate SPT and CPT outcomes significantly. It has been concluded that using a large number of data from various locations will enable in an effective way for generating SPT-CPT correlation with artificial neural networks.

 Standard Penetration Test SPT and Cone Penetration Test CPT are among the most commonly used methods in soil investigations. Many soil parameters are associated with SPT andor CPT. It is essential to obtain a reliable correlation between SPT and CPT to use one of these tests individually. In this study, the tip resistance qc values obtained with CPT corresponding to SPT-N data for various soil types were reached utilizing the literature. Analysis of variance was performed to ascertain whether there was a significant relationship between SPT-N values and tip resistance qc data. Separate functions with high-order correlation values were obtained for each soil type by creating simulatable networks with artificial neural networks for SPT-CPT correlation. The obtained results were compared with experimental data and equations in the literature. As a result, functions in good correlation values created by the support of artificial intelligence for various soil types have provided the opportunity to correlate SPT and CPT outcomes significantly. It has been concluded that using a large number of data from various locations will enable in an effective way for generating SPT-CPT correlation with artificial neural networks.

 Standard Penetration Test SPT and Cone Penetration Test CPT are among the most commonly used methods in soil investigations. Many soil parameters are associated with SPT andor CPT. It is essential to obtain a reliable correlation between SPT and CPT to use one of these tests individually. In this study, the tip resistance qc values obtained with CPT corresponding to SPT-N data for various soil types were reached utilizing the literature. Analysis of variance was performed to ascertain whether there was a significant relationship between SPT-N values and tip resistance qc data. Separate functions with high-order correlation values were obtained for each soil type by creating simulatable networks with artificial neural networks for SPT-CPT correlation. The obtained results were compared with experimental data and equations in the literature. As a result, functions in good correlation values created by the support of artificial intelligence for various soil types have provided the opportunity to correlate SPT and CPT outcomes significantly. It has been concluded that using a large number of data from various locations will enable in an effective way for generating SPT-CPT correlation with artificial neural networks.

 Standart Penetrasyon Testi SPT ve Koni Penetrasyon Testi CPT zemin araştırmalarında en sık kullanılan yöntemler arasında yer almaktadır. Birçok zemin parametresi SPT veveya CPT ile ilişkilendirilmiştir. Bu testlerden herhangi birinin yokluğunda bir diğerinin kullanılabilmesi için SPT-CPT arasında güvenilir bir korelasyonun elde edilmesi önem arz etmektedir. Bu çalışmada literatürden yararlanılarak çeşitli zemin tipleri için SPT-N verilerine karşılık gelen CPT ile elde edilmiş uç direnci qc değerlerine ulaşılmıştır. SPT-N değerleri ile uç direnci qc verileri arasında anlamlı bir ilişkinin olup olmadığını belirlemek için varyans analizi gerçekleştirilmiştir. SPT-CPT korelasyonu için yapay sinir ağları ile simüle edilebilir ağlar oluşturularak her zemin tipi için yüksek dereceli korelasyon değerlerine sahip ayrı fonksiyonlar elde edilmiştir. Ulaşılan sonuçlar deneysel verilerle ve literatürdeki denklemlerle karşılaştırılmıştır. Böylece farklı zemin tipleri için yapay zeka desteğiyle oluşturulmuş iyi derecede korelasyon değerlerine sahip fonksiyonların yardımıyla SPT ile elde edilen sonuçların CPT ile anlamlı olarak ilişkilendirilmesine olanak sağlanmıştır. Farklı bölgelere ait çok sayıda verinin kullanılması durumunda yapay sinir ağları ile SPT-CPT korelasyonu oluşturmanın başarılı bir yöntem olacağı sonucuna varılmıştır.

Article Details

Investigation of SPT-CPT Relationship for Various Soil Types with Artificial Intelligence Support

Authors:

Arda Burak Ekmen

Publisher:

Erciyes University

Main Topics:

Mühendislik

Journal:

Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.39 Issue 2, pp.204-216

Document Type:

Article

Publication Date:

8/2023

Document Language:

Turkish

Document ID (Org No.):

2001297831

pISSN:

1012-2354

Cone Penetration Test

Standard Penetration Test

Artificial intelligence

SPT-CPT correlationKoni Penetrasyon Testi

Standart Penetrasyon Testi

Yapay Zeka

SPT-CPT korelasyonu

Issues

Article Details

Authors: Arda Burak Ekmen

Publisher: Erciyes University

Main Topics: Mühendislik

Journal: Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.39 Issue 2, pp.204-216

Document Type: Article

Publication Date: 8/2023

Document Language: Turkish

Document ID (Org No.): 2001297831

pISSN: 1012-2354

Issues

Authors:

Arda Burak Ekmen

Publisher:

Erciyes University

Main Topics:

Mühendislik

Journal:

Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.39 Issue 2, pp.204-216

Document Type:

Article

Publication Date:

8/2023

Document Language:

Turkish

Document ID (Org No.):

2001297831

pISSN:

1012-2354