scholarly journals Soil Classification System from Cone Penetration Test Data Applying Distance-Based Machine Learning Algorithms

2019 ◽  
Vol 42 (2) ◽  
pp. 167-178
Author(s):  
Lucas Orbolato Carvalho ◽  
Dimas Betioli Ribeiro
Keyword(s):  
Machine Learning ◽  
Test Data ◽  
Classification System ◽  
Learning Algorithms ◽  
Soil Classification ◽  
Cone Penetration Test ◽  
Machine Learning Algorithms ◽  
Penetration Test ◽  
Cone Penetration

Computerized Cone Penetration Test for Soil Classification

2008 ◽  
Vol 2053 (1) ◽  
pp. 47-64 ◽  
Author(s):  
Murad Y. Abu-Farsakh ◽  
Zhongjie Zhang ◽  
Mehmet Tumay ◽  
Mark Morvant
Keyword(s):  
Classification System ◽  
Soil Classification ◽  
Cone Penetration Test ◽  
Classification Systems ◽  
Fuzzy Classification ◽  
Classification Methods ◽  
Penetration Test ◽  
Cone Penetration ◽  
Cone Tip Resistance ◽  
Tip Resistance

Computerized MS-Windows Visual Basic software of a cone penetration test (CPT) for soil classification was developed as part of an extensive effort to facilitate the implementation of CPT technology in many geotechnical engineering applications. Five CPT soil engineering classification systems were implemented as a handy, user-friendly, software tool for geotechnical engineers. In the probabilistic region estimation and fuzzy classification methods, a conformal transformation is first applied to determine the profile of soil classification index (U) with depth from cone tip resistance (qc) and friction ratio (Rf). A statistical correlation was established in the probabilistic region estimation method between the U index and the compositional soil type given by the Unified Soil Classification System. Conversely, the CPT fuzzy classification emphasizes the certainty of soil behavior. The Schmertmann and Douglas and Olsen methods provide soil classification charts based on cone tip resistance and friction ratio. However, Robertson et al. proposed a three-dimensional classification system that is presented in two charts: one chart uses corrected tip resistance (qt) and friction ratio (Rf); the other chart uses qt and pore pressure parameter (Bq) as input data. Five sites in Louisiana were selected for this study. For each site, CPT tests and the corresponding soil boring results were correlated. The soil classification results obtained using the five different CPT soil classification methods were compared.

scholarly journals A multiple model machine learning approach for soil classification from cone penetration test data

2021 ◽  
Vol 44 (4) ◽  
pp. 1-14
Author(s):  
Lucas Carvalho ◽  
Dimas Ribeiro
Keyword(s):  
Machine Learning ◽  
Soil Classification ◽  
Predictive Performance ◽  
Cone Penetration Test ◽  
Machine Learning Techniques ◽  
Multiple Model ◽  
Penetration Test ◽  
Cone Penetration ◽  
Learning Techniques ◽  
Soil Age

The most popular methods for soil classification from cone penetration test (CPT) data are based on examining two-dimensional charts. In the last years, several authors have dedicated efforts on replicating and discussing these methods using machine learning techniques. Nonetheless, most of them apply few techniques, include only one dataset and do not explore more than three input features. This work circumvents these issues by: (i) comparing five different machine learning techniques, which are also combined in an ensemble; (ii) using three distinct CPT datasets, one composed of 111 soundings from different countries, one composed of 38 soundings with information of soil age and the third composed of 64 soundings taken from the city of São Paulo, Brazil; and (iii) testing combinations of five input features. Results show that, in most cases, the ensemble of multiple models achieves better predictive performance than any technique isolated. Accuracies close to the maximum were obtained in some cases without the need of pore pressure information, which is costly to measure in geotechnical practice.

Cone penetration test (CPT)-based soil behaviour type (SBT) classification system — an update

2016 ◽  
Vol 53 (12) ◽  
pp. 1910-1927 ◽  
Author(s):  
P.K. Robertson
Keyword(s):  
Classification System ◽  
Soil Classification ◽  
Cost Effective ◽  
Cone Penetration Test ◽  
Classification Systems ◽  
Penetration Test ◽  
Cone Penetration ◽  
Soil Behaviour ◽  
Behaviour Type

A soil classification system is used to group soils according to shared qualities or characteristics based on simple cost-effective tests. The most common soil classification systems used in geotechnical engineering are based on physical (textural) characteristics such as grain size and plasticity. Ideally, geotechnical engineers would also like to classify soils based on behaviour characteristics that have a strong link to fundamental in situ behaviour. However, existing textural-based classification systems have a weak link to in situ behaviour, since they are measured on disturbed and remolded samples. The cone penetration test (CPT) has been gaining in popularity for site investigations due to the cost-effective, rapid, continuous, and reliable measurements. The most common CPT-based classification systems are based on behaviour characteristics and are often referred to as a soil behaviour type (SBT) classification. However, some confusion exists, since most CPT-based SBT classification systems use textural-based descriptions, such as sand and clay. This paper presents an update of popular CPT-based SBT classification systems to use behaviour-based descriptions. The update includes a method to identify the existence of microstructure in soils, and examples are used to illustrate the advantages and limitations of such a system.

Sign in / Sign up

Export Citation Format

Share Document