scholarly journals Soil Unconfined Compressive Strength Prediction Using Random Forest (RF) Machine Learning Model

2020 ◽  
Vol 14 (1) ◽  
pp. 278-285
Author(s):  
Hai-Bang Ly ◽  
Binh Thai Pham
Keyword(s):  
Machine Learning ◽  
Compressive Strength ◽  
Random Forest ◽  
Specific Gravity ◽  
Moisture Content ◽  
Unconfined Compressive Strength ◽  
Void Ratio ◽  
Clay Content ◽  
Liquid Limit ◽  
Plastic Limit

Aims: Understanding the mechanical performance and applicability of soils is crucial in geotechnical engineering applications. This study investigated the possibility of application of the Random Forest (RF) algorithm – a popular machine learning method to predict the soil unconfined compressive strength (UCS), which is one of the most important mechanical properties of soils. Methods: A total number of 118 samples collected and their tests derived from the laboratorial experiments carried out under the Long Phu 1 power plant project, Vietnam. Data used for modeling includes clay content, moisture content, specific gravity, void ratio, liquid limit and plastic limit as input variables, whereas the target is the UCS. Several assessment criteria were used for evaluating the RF model, namely the correlation coefficient (R), root mean squared error (RMSE) and mean absolute error (MAE). Results: The results showed that RF exhibited a strong capability to predict the UCS, with the R value of 0.914 and 0.848 for the training and testing datasets, respectively. Finally, a sensitivity analysis was conducted to reveal the importance of input parameters to the prediction of UCS using RF. The specific gravity was found as the most affecting variable, following by clay content, liquid limit, plastic limit, moisture content and void ratio. Conclusion: This study might help in the accurate and quick prediction of the UCS for practice purpose.

scholarly journals Estimation of Soil Cohesion Using Machine Learning Method: A Random Forest Approach

2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Hai-Bang Ly ◽  
Thuy-Anh Nguyen ◽  
Binh Thai Pham
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Soil Properties ◽  
Clay Content ◽  
Absolute Error ◽  
Experimental Methods ◽  
Liquid Limit ◽  
Support Vector ◽  
Data Set ◽  
Soil Cohesion

Soil cohesion (C) is one of the critical soil properties and is closely related to basic soil properties such as particle size distribution, pore size, and shear strength. Hence, it is mainly determined by experimental methods. However, the experimental methods are often time-consuming and costly. Therefore, developing an alternative approach based on machine learning (ML) techniques to solve this problem is highly recommended. In this study, machine learning models, namely, support vector machine (SVM), Gaussian regression process (GPR), and random forest (RF), were built based on a data set of 145 soil samples collected from the Da Nang-Quang Ngai expressway project, Vietnam. The database also includes six input parameters, that is, clay content, moisture content, liquid limit, plastic limit, specific gravity, and void ratio. The performance of the model was assessed by three statistical criteria, namely, the correlation coefficient (R), mean absolute error (MAE), and root mean square error (RMSE). The results demonstrated that the proposed RF model could accurately predict soil cohesion with high accuracy (R = 0.891) and low error (RMSE = 3.323 and MAE = 2.511), and its predictive capability is better than SVM and GPR. Therefore, the RF model can be used as a cost-effective approach in predicting soil cohesion forces used in the design and inspection of constructions.

scholarly journals The Effect of Oil Contamination on some Geotechnical Properties of West Qurna Oilfield Soils at Basrah, Southern Iraq

2021 ◽  
Vol 54 (2B) ◽  
pp. 76-84
Author(s):  
Ahmed K. Al-Nimah
Keyword(s):  
Compressive Strength ◽  
Moisture Content ◽  
Crude Oil ◽  
Unconfined Compressive Strength ◽  
Liquid Limit ◽  
Geotechnical Properties ◽  
Oil Contamination ◽  
The West ◽  
Coefficient Of Consolidation ◽  
Clayey Silt

Oil contamination in soils causes several geotechnical problems that must be considered during construction. The contamination occurs due to oil seepage which could happen during oil explorations and production processes or oil transportation. The site of West Qurna oilfield in Basrah was selected for this study because it has witnessed oil seepages many times. In order to study the significant impact on geotechnical properties of soils in the West Qurna site, as uncontaminated bulk soil sample was taken at a depth of 1 m, and crude oil was added at weight ratios of 2, 4, 6, 8, and 10 %. Laboratory tests were performed on all samples; these tests included particle size distribution, moisture content, Atterberg’s limits, consolidation, unconfined compressive strength, and water absorption. The results show that soil at the West Qurna site is clayey silt with little sand and the moisture content is 29.21%. The values of liquid limit and plasticity index were gradually decreased, while the plastic limit was increased with increasing of crude oil in the soil of study. There was an increase in consolidation coefficients [compressive index, swelling index, pre-consolidation pressure, and coefficient of consolidation] with an increase in the percentages of crude oil in the soil. The results also show that there was a decrease in the values of unconfined compressive strength and absorption of water as the crude oil was increased in the soil.

scholarly journals Stabilization of Marine Clay Soil Using Polyurethane

2018 ◽  
Vol 250 ◽  
pp. 01004 ◽  
Author(s):  
Samaila Saleh ◽  
Nur Zurairahetty Mohd Yunus ◽  
Kamarudin Ahmad ◽  
Nazri Ali
Keyword(s):  
Compressive Strength ◽  
Moisture Content ◽  
Unconfined Compressive Strength ◽  
Clay Soil ◽  
Organic Matter Content ◽  
Road Construction ◽  
Liquid Limit ◽  
Matter Content ◽  
Marine Clay ◽  
The Road

Many chemicals stabilisation techniques are being employed all over the world to improve the engineering and physical properties of the problematic soils and reduce the potential damages caused by them. Out of those chemical stabilisation technics, application of Polyurethane to improve the strength of marine clay was investigated in the laboratory. Characterization of the soil geotechnical properties was carried out by conducting laboratory test that includes natural moisture content, Atterberg limits, grains sizes analyses, specific gravity, moisture-density relationship, unconfined compressive strength (UCS), organic matter content and PH tests. Unconfined compressive strength test at optimum moisture content with varying the dose of the Polyurethane content was conducted to test the effectiveness of Polyurethane as a chemical stabiliser. The result of the preliminary tests of the sample shows that the soil has a liquid limit of 65%, plastic limit of 26% and plasticity index of 53%. The percentages of gravel, sand and fines in the marine clay sample were 0 %, 1.32 % and 98.68 % respectively %. The results of the UCS test also revealed that Polyurethane stabilisation improved the strength of marine clay by 230%. Thus, the improvement in strength of stabilised marine clay soil can significantly reduce the overall thickness of the pavement and total cost of the road construction in future.

scholarly journals Experimental Study of the Effects of Some Selected Geotechnical Indices on the Unconfined Compressive Strength of Lateritic Soil

2019 ◽  
Vol 2 (Issue 1) ◽  
Author(s):  
G.O Adunoye ◽  
O.C Onah ◽  
F.O Ajibade
Keyword(s):  
Experimental Study ◽  
Compressive Strength ◽  
Specific Gravity ◽  
Moisture Content ◽  
Unconfined Compressive Strength ◽  
Soil Samples ◽  
Atterberg Limits ◽  
Effective Size ◽  
Natural Moisture Content ◽  
Coefficient Of Uniformity

This study undertook an experimental study of the comparative effects of Atterberg limits, particles size and compaction parameters on the unconfined compressive strength of selected soils. This was with a view to ascertaining which of the combinations of the soil properties will produce a good prediction of the unconfined compressive strength. To achieve this aim, soil samples were obtained from selected locations within Ife Central Local Government Area, Osun State, Nigeria. The following tests were conducted on the soil samples, following standard procedures: natural moisture content determination, specific gravity, Atterberg limits, compaction and unconfined compressive strength. Using Regression tool, the results obtained from the laboratory tests were used to develop the relationships between each of the index properties and unconfined compressive strength. Results showed that the natural moisture content of soil samples ranges between 18.48 % and 25.03 %; specific gravity ranges between 2.35 and 2.69; liquid limit ranges between 39.95 % and 83.98 %; plastic limit ranges between 29.32 % and 51.18 %; and plasticity index is between 8.74 % and 33.56 %. The maximum dry density ranges between 15.30kN/m3 and 19.30kN/m3 with their optimum moisture contents ranging between 13.80 % and 35.50 % while unconfined compressive strength is between 36.00 kN/m2 and 97.14 kN/m2. The results of regression analysis showed that effective size and coefficient of uniformity have the greatest effect (R2 = 0.82) on unconfined compressive strength of the tested soil samples. Therefore, the study concluded that effective size and coefficient of uniformity could be used to estimate the unconfined compressive strength of the soils.

scholarly journals Geotechnical investigation of road pavement failure along arigidi / oke - agbe akoko, southwestern, Nigeria

2020 ◽  
Vol 8 (2) ◽  
pp. 35
Author(s):  
Thompson Henry Tolulope Ogunribido ◽  
Tunde Ezekiel Fadairo
Keyword(s):  
Compressive Strength ◽  
Specific Gravity ◽  
Moisture Content ◽  
Unconfined Compressive Strength ◽  
Soil Samples ◽  
California Bearing Ratio ◽  
Dry Density ◽  
Geological Condition ◽  
Road Pavement ◽  
Natural Moisture Content

Twenty soil samples collected from the failed portions in the study area were air dried for two weeks before analyses. Each soil samples were subjected to eight engineering tests which include: natural moisture content, atterberg limit, specific gravity, compaction, unconfined compressive strength, California bearing ratio, grain size and hydrometer analysis. Results showed that the natural moisture content ranged from 17.7% to 37.8%, liquid limit from 48.5% to 62.4%, plastic limit from 18.3% to 26.8%, plasticity index from 25.7% to 37.7%, shrinkage limit from 5.8%-12.5%, optimum moisture content from 14.2% to 32.4%, maximum dry density from 1301 Kg/rn3 to 2002 Kg/rn3. Soaked California bearing ratio ranged from 5% to 17%, unsoaked from 15% to 38%, specific gravity from 2.5 to 2.68, unconfined compressive strength r from 112.8 Kpa to 259.7 Kpa, shear strength from 56.4 Kpa to 129.9 Kpa and hydrometer analysis from 48.5% to 72.1%. Based on the Federal Government specifications for pavement construction, for the soil to be suitable, stabilization with bitumen, Portland cement, lime, coal fly ash, and saw dust should be done. Road pavement failure along Arigidi – Oke Agbe road was due to poor engineering geological condition of the sub-grade soils and poor drainage systems.  

Experimental Study on the Stabilization Effects of Dredged Sludge by Fly Ash or Phosphogypsum

2013 ◽  
Vol 689 ◽  
pp. 342-347 ◽  
Author(s):  
Zhi Hua Yu ◽  
Yue Gui ◽  
Qing Zhang ◽  
Xiang Yun Kong
Keyword(s):  
Compressive Strength ◽  
Fly Ash ◽  
Unconfined Compressive Strength ◽  
Liquid Limit ◽  
Industrial Wastes ◽  
Plastic Limit ◽  
Dredged Sludge ◽  
Ratio Limit ◽  
Optimal Effect ◽  
Curing Age

It is very essential to explore a more efficient and a lower-cost stabilizer based the traditional stabilizer, lime. Through the laboratory test, this article made a comparison on the stabilization effects from the water ratio limit and unconfined compressive strength of the stabilized sludge, which was processed by using two common industrial wastes as stabilizers, fly ash and phosphogypsum, with the lime. The laboratory experiment results indicate that liquid limit and plastic limit of phosphogypsum compound stabilizers have a significant increase compared with the single lime added solidified sludge, but little change in the plasticity index over a curing age of 7 days or 28 days; while that of by adding fly ash has almost no change in liquid limit and plastic limit compared with the single lime added solidified sludge. Meanwhile the solidified sludge by adding the previous two stabilizers have an increase in unconfined compressive strength compared with the single lime added solidified sludge. Comprehensively compared above, the waste phosphogypsum as the extra additive stabilizer of the lime makes the optimal effect.

scholarly journals Utilization of dolerite waste powder for improving geotechnical parameters of compacted clay soil

2021 ◽  
Vol 13 (1) ◽  
pp. 1523-1535
Author(s):  
Syed Husnain Ali Shah ◽  
Mohammad Arif ◽  
Qasim ur Rehman ◽  
Fawaz Manzoor
Keyword(s):  
Compressive Strength ◽  
Moisture Content ◽  
Unconfined Compressive Strength ◽  
Void Ratio ◽  
Clay Soil ◽  
Atterberg Limits ◽  
Degree Of Saturation ◽  
Dry Density ◽  
Compacted Clay ◽  
Grain Sizes

Abstract This study explores how dolerite cutting waste could be utilized for improving the quality of compacted clay soils. Different proportions of dolerite waste powder with varying grain sizes were used as admixtures and their impact on clay soil properties investigated. Ten samples were prepared by mixing clay soil with different proportions of dolerite waste powder having grain sizes of 0.210, 0.297, and 0.420 mm. The resulting samples were subjected to Proctor compaction, and their maximum dry density and optimum moisture content were measured. Next, all the compacted samples were subjected to geotechnical testing, including the determination of Atterberg limits, California bearing ratio (CBR), unconfined compressive strength, and specific gravity (Gs). The values of compaction parameters, Atterberg limits, and Gs were utilized for finding the porosity, void ratio, saturation potential, liquidity index (LI), and consistency index (CI). The results demonstrate that the addition of dolerite powder produces a substantial improvement in the plasticity index, compaction parameters, CBR, unconfined compressive strength, Gs, porosity, void ratio, degree of saturation, LI, and CI. The foremost reason for this improvement is the presence of denser and less water-adoring minerals in the added dolerite relative to pristine clay soil. Furthermore, the observed positive impact on the soils’ geotechnical comportment is comparatively higher with coarser than finer dolerite because of the decrease in surface area that causes a reduction in the moisture content and porosity but an increase in the density of soil.

scholarly journals Experimental Study on Improvement of Marine Soft Soil with Alkali Residue

2018 ◽  
Vol 53 ◽  
pp. 04021
Author(s):  
SHAO Yong ◽  
LIU Xiao-li ◽  
ZHU Jin-jun
Keyword(s):  
Experimental Study ◽  
Compressive Strength ◽  
Unconfined Compressive Strength ◽  
Void Ratio ◽  
Soft Soil ◽  
Compressive Modulus ◽  
Engineering Properties ◽  
Test Results ◽  
Use Of Resources ◽  
One Year

Industrial alkali slag is the discharge waste in the process of alkali production. About one million tons of alkali slag is discharged in China in one year. It is a burden on the environment, whether it is directly stacked or discharged into the sea. If we can realize the use of resources, it is a multi-pronged move, so alkali slag is used to improve solidified marine soft soil in this paper. The test results show that the alkali residue can effectively improve the engineering properties of marine soft soil. Among them, the unconfined compressive strength and compressive modulus are increased by about 10 times, and the void ratio and plasticity index can all reach the level of general clay. It shows that alkali slag has the potential to improve marine soft soil and can be popularized in engineering.

Selected Properties of Parallel Strand Lumber Made from Southern Pine and Yellow-Poplar

Holzforschung ◽  
2003 ◽  
Vol 57 (2) ◽  
pp. 207-212 ◽  
Author(s):  
Y. Liu ◽  
A.W.C. Lee
Keyword(s):  
Compressive Strength ◽  
Shear Strength ◽  
Specific Gravity ◽  
Moisture Content ◽  
Dimensional Stability ◽  
Modulus Of Rupture ◽  
Southern Pine ◽  
Yellow Poplar ◽  
Average Modulus ◽  
Parallel Strand Lumber

Summary This study was conducted to explore basic physical and mechanical properties of parallel strand lumber (PSL) made exclusively from southern pine and yellow-poplar, respectively, and to examine their relationships using statistical analysis. Small specimens were prepared from commercial southern pine PSL and yellow-poplar PSL billets and tested for specific gravity, moisture content, dimensional stability, bending properties, shear strength and compressive strength. Results indicate average specific gravity of southern pine PSL is higher than that of yellow-poplar PSL, while their average moisture content and dimensional stability are very similar. Southern pine PSL has higher average modulus of elasticity but lower average modulus of rupture than yellow-poplar PSL. While average longitudinal shear strength does not exhibit differences between southern pine PSL and yellow-poplar PSL, average compressive strength of southern pine PSL is higher than that of yellow-poplar PSL. There are positive correlations among modulus of elasticity, modulus of rupture and specific gravity. PSL improves some properties of solid wood from which PSL is made.

scholarly journals Prediction of Shear Strength of Soil Using Direct Shear Test and Support Vector Machine Model

2020 ◽  
Vol 14 (1) ◽  
pp. 41-50 ◽  
Author(s):  
Hai-Bang Ly ◽  
Binh Thai Pham
Keyword(s):  
Support Vector Machine ◽  
Shear Strength ◽  
Moisture Content ◽  
Mean Squared Error ◽  
Learning Algorithm ◽  
Liquid Limit ◽  
Support Vector ◽  
Plastic Limit ◽  
Svm Model ◽  
Soil Shear Strength

Background: Shear strength of soil, the magnitude of shear stress that a soil can maintain, is an important factor in geotechnical engineering. Objective: The main objective of this study is dedicated to the development of a machine learning algorithm, namely Support Vector Machine (SVM) to predict the shear strength of soil based on 6 input variables such as clay content, moisture content, specific gravity, void ratio, liquid limit and plastic limit. Methods: An important number of experimental measurements, including more than 500 samples was gathered from the Long Phu 1 power plant project’s technical reports. The accuracy of the proposed SVM was evaluated using statistical indicators such as the coefficient of correlation (R), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) over a number of 200 simulations taking into account the random sampling effect. Finally, the most accurate SVM model was used to interpret the prediction results due to Partial Dependence Plots (PDP). Results: Validation results showed that SVM model performed well for prediction of soil shear strength (R = 0.9 to 0.95), and the moisture content, liquid limit and plastic limit were found as the three most affecting features to the prediction of soil shear strength. Conclusion: This study might help in quick and accurate prediction of soil shear strength for practical purposes in civil engineering.

Sign in / Sign up

Export Citation Format

Share Document