Estimation of shear strength parameter of silty sand from SPT-N60 using machine learning models

Hossain, Altaf; Alam, Tahsina; Barua, Suman; Rahman, Manas

doi:10.1080/17486025.2021.1975048

Cited by 3 publications

(2 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their findings revealed a significant correlation between database size and prediction accuracy, indicating that larger databases yield improved results. Hossain et al (2021) conducted a study in which they employed multiple-regression analysis techniques, including Multiple Linear Regression (MLR), Artificial Neural Network (ANN) and Support Vector Machine (SVM), to estimate the internal frictional angle (Φ) of soil. SVM, like ANN, is a machine-learning algorithm capable of effectively predicting a target variable.…”

Section: Introductionmentioning

confidence: 99%

“…The models developed in Larbi et al (2019), Hossain et al (2021), Tabarsa et al (2021) and Saadat & Bayat (2022) studies were not used in comparison with the current study. This is because Larbi et al's (2019) study was to estimate the compressive strength of concretes, whereas Hossain et al (2021) built a model to predict the internal frictional angle (Φ) of soil. Although Tabarsa et al (2021) and Saadat & Bayat (2022) aimed to predict qu, their models used different input variables compared to the present study.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Regression Analysis for Predicting Soil Strength in Bangladesh

Rahman Sabab,

Md. Shahin,

Muhim Bondhon

et al. 2023

JJCE

View full text Add to dashboard Cite

This study focuses on establishing a robust relationship between Standard Penetration Test-N values (SPT-N), geotechnical parameters and unconfined compressive strength (qu) using regression analysis. The proposed relationship offers a reliable method for estimating qu based on SPT-N values. A comprehensive dataset comprising approximately 200 soil samples collected from various boreholes across Dhaka city was utilized. Multiple Linear Regression (MLR), Rando-forest Regression (RFR) and AdaBoost Regression techniques were employed to develop a unified correlation model. Evaluation metrics including R-squared (R2), Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE), along with Trend-behavior Analysis were employed to assess and compare the performances of the models. Additionally, sensitivity analysis was carried out on the selected model in order to assess the importance of each parameter used to predict qu. Finally, the selected model was compared against the existing empirical models that were published in previous studies. In terms of evaluation metrics and Trend-behavior Analysis, the results showed that the RFR model performed better than the others. Additionally, the selected model outperformed the others, demonstrating the highest R2 score, the smallest RMSE and MAE values and lower residuals compared to the previous models. Hence, the proposed model provides accurate predictions of qu for clayey soil in Bangladesh. Its implementation could ensure more efficient geotechnical designs, specifically adjusted to the geological conditions of the Dhaka region. While previous studies have established regional equations for various parts of the world, our model uniquely has incorporated the Plasticity Index (PI) as a predictor for qu and is specifically calibrated for the geological characteristics of Dhaka city. The findings of this study highlight the effectiveness and applicability of regression analysis in predicting qu for Dhaka's soil properties, thus introducing a valuable tool for enhancing the accuracy and effectiveness of geotechnical assessments and design in the region. KEYWORDS: Unconfined compressive strength, Standard penetration test-N values, Plasticity index, Multiple linear regression, Random-forest regression, AdaBoost regression, Evaluation metrics, Trend-behavior analysis, Sensitivity analysis

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Regression Analysis for Predicting Soil Strength in Bangladesh

Rahman Sabab,

Md. Shahin,

Muhim Bondhon

et al. 2023

JJCE

View full text Add to dashboard Cite

show abstract

Evaluation of SPT-N values and internal friction angle correlation using artificial intelligence methods in granular soils

Ekmen

2023

Soil Res.

View full text Add to dashboard Cite

Context Artificial neural networks (ANNs) and genetic algorithms (GAs) have become widely used in various engineering fields due to their ability to solve complicated issues directly. Aims In this study, internal friction angle (ϕ) values for granular soils were calculated using ANNs, GAs, and empirical methods based on standard penetration test (SPT) data to designate the system that produced the best statistical outcomes. Methods Utilising the literature, experimentally determined internal friction angle (Eϕ) values were obtained for a significant quantity of standard penetration test data. Analysis of variance was performed to ascertain whether there was a significant correlation between SPT-N60 values and Eϕ. A simulated network was created with ANNs, and a function was obtained with GAs for SPT-N60–ϕ correlation. The outcomes obtained with ANNs and GAs were compared with empirical equations and experimental results. Optimisation analysis was conducted with the novel Improved Goal Attainment method to minimise the margin of error. Key results Compared to the GAs and empirical equations, the ANN has been determined to have a reasonable correlation with experimental results. Conclusions It was determined that by utilising ANNs, the current empirical equations indicating the relationship between different soil parameters and the data of tests such as SPT and cone penetration test (CPT) could be produced in improved correlations by employing a large number of data sets obtained from different regions. Implications Effective predictions can be achieved instead of present methods.

show abstract

Relationship of Resistivity Value with the Ground Material Information Obtained from Borehole Data: Case study

Bakar,

Talib,

Arifin

et al. 2024

Preprint

View full text Add to dashboard Cite

In typical site investigations, borehole data offers localized information at specific depths within the borehole. It offers high vertical resolution but limited spatial coverage since it represents a point measurement. Electrical Resistivity Tomography (ERT) data, on the other hand, provides a continuous profile of subsurface resistivity distribution along a survey line or grid. It offers relatively high spatial coverage, capturing information between measurement locations. The study aims to understand on reliability of resistivity value with the soil or rock material information obtained from conventional borehole data located in Melaka, Malaysia. A resistivity lines were conducted ERT method with the Gradient XL protocol. Based on the results, high resistivity values were observed at first 5 meters, indicating the presence of Gravelly SAND and Silty SAND layers ranging from 5 to 6 SPT N-value. The resistivity value then decreases below 100 Ωm starting from a depth of 5meters and continuing until 9meters which indicate highly saturated layer. Then from a depth of 9 to 15 meters, the resistivity value increases, indicating the soil layer is less saturated with water due to its high resistivity value. It can be concluded that ERT data is reliable to represent subsurface data when comparing to borehole data. However, not all data can be fully understood or interpreted based on solely resistivity value especially around weathered materials. For instance, in area of immersed soil or rock, resistivity sometimes still gives higher value that is almost the same as in an unsaturated area. It was observed that the relationship of SPT-N and resistivity value is more consistent when the materials are in range stiff to hard layer.

show abstract

Estimation of shear strength parameter of silty sand from SPT-N60 using machine learning models

Cited by 3 publications

References 18 publications

Regression Analysis for Predicting Soil Strength in Bangladesh

Regression Analysis for Predicting Soil Strength in Bangladesh

Evaluation of SPT-N values and internal friction angle correlation using artificial intelligence methods in granular soils

Relationship of Resistivity Value with the Ground Material Information Obtained from Borehole Data: Case study

Contact Info

Product

Resources

About