Models For Unconfined Compressive Strength Research Articles

Random forest, CART, and MLR-based predictive model for unconfined compressive strength of cement reinforced clayey soil: a comparative analysis

Random forest, CART, and MLR-based predictive model for unconfined compressive strength of cement reinforced clayey soil: a comparative analysis

Developing interpretable machine learning-Shapley additive explanations model for unconfined compressive strength of cohesive soils stabilized with geopolymer.

This paper seeks to develop an interpretable Machine Learning (ML) model for predicting the unconfined compressive strength (UCS) of cohesive soils stabilized with geopolymer at 28 days. Four models including Random Forest (RF), Artificial Neuron Network (ANN), Extreme Gradient Boosting (XGB), and Gradient Boosting (GB) are built. The database consists of 282 samples collected from the literature with three different types of cohesive soil stabilized with three geopolymer categories including Slag-based geopolymer cement, alkali-activated fly ash geopolymer and slag/fly ash-based geopolymer cement. The optimal model is selected by comparing their performances with each other. The values of hyperparameters are tuned by Particle Swarm Optimization (PSO) algorithm and K-Fold Cross Validation. Statistical indicators show the superior performance of the ANN model with three metrics performance such as coefficient of determination R2 = 0.9808, Root Mean Square Error RMSE = 0.8808 MPa and Mean Absolute Error MAE = 0.6344 MPa. In addition, a sensitivity analysis was performed to determine the influence of different input parameters on the UCS of cohesive soils stabilized with geopolymer. The order of feature effect can be ordered in descending order using the Shapley additive explanations (SHAP) value as follows: Ground granulated blast slag content (GGBFS) > Liquid limit (LL) > Alkali/Binder ratio (A/B) > Molarity (M) > Fly ash content (FA) > Na/Al > Si/Al. The ANN model can obtain the best accuracy using these seven inputs. LL has a negative correlation with the growth of unconfined compressive strength, whereas GGBFS has a positive correlation.

Investigation on stabilization/solidification characteristics of lead-contaminated soil using innovative composite model of cement and soda residue

The cement solidification/stabilization method of heavy metal contaminated soils has been promoted in engineering practice and applied on a large scale for site remediation, but it still reveals some scientific problems in the current complex and variable global extreme climate. To solve these problems, this study used waste soda residue as a composite additive, and established an innovative composite model (cement-soda) to remediate lead-contaminated soil. The innovative composite model for unconfined compressive strength (UCS) and toxic leaching properties under different conditions were investigated As well as curing and leaching mechanisms are discussed, and future industrial practice was evaluated. The results showed that the addition of soda residue improved the early (20 days) UCS of the composite curing agent for lead-contaminated soil by an average of 23.1% MPa. The UCS strength was 0.96 MPa, which was the maximum when the percentage of soda residue composite was 40%. The concentration of Pb2+ in the leachate of the cement-soda residue (CSH) composite curing agent was greatly reduced (average 3.28 times) compared with that of a single cement in the same situation, with an average leached Pb2+ concentration of 1.87 mg·L−1. This finding indicates that the addition of alkali residue improved the curing effect. The curing mechanism was divided into four steps, mainly a complex physicochemical reaction between the CSH composite and soil particles. The leaching mechanism of CSH to aqueous solution is mainly the consumption of acid ions by alkaline substances. Collectively, this study will provide scientific data to support potential lead-containing soil in site remediation technologies and future large-scale engineering applications.

Performance Prediction of Cement Stabilized Soil Incorporating Solid Waste and Propylene Fiber.

Cement stabilized soil (CSS) yields wide application as a routine cementitious material due to cost-effectiveness. However, the mechanical strength of CSS impedes development. This research assesses the feasible combined enhancement of unconfined compressive strength (UCS) and flexural strength (FS) of construction and demolition (C&D) waste, polypropylene fiber, and sodium sulfate. Moreover, machine learning (ML) techniques including Back Propagation Neural Network (BPNN) and Random Forest (FR) were applied to estimate UCS and FS based on the comprehensive dataset. The laboratory tests were conducted at 7-, 14-, and 28-day curing age, indicating the positive effect of cement, C&D waste, and sodium sulfate. The improvement caused by polypropylene fiber on FS was also evaluated from the 81 experimental results. In addition, the beetle antennae search (BAS) approach and 10-fold cross-validation were employed to automatically tune the hyperparameters, avoiding tedious effort. The consequent correlation coefficients (R) ranged from 0.9295 to 0.9717 for BPNN, and 0.9262 to 0.9877 for RF, respectively, indicating the accuracy and reliability of the prediction. K-Nearest Neighbor (KNN), logistic regression (LR), and multiple linear regression (MLR) were conducted to validate the BPNN and RF algorithms. Furthermore, box and Taylor diagrams proved the BAS-BPNN and BAS-RF as the best-performed model for UCS and FS prediction, respectively. The optimal mixture design was proposed as 30% cement, 20% C&D waste, 4% fiber, and 0.8% sodium sulfate based on the importance score for each variable.

Evaluation of Cement and Bitumen Emulsion on Strength Characteristics of Native South African Granular Soil as Base Course Material

Pavement rehabilitation where the material in the existing pavement is recycled in-situ with bitumen will sustain the environment with conservation of natural aggregates, reduction in noise, dust emission and traffic disruption. This study investigate the effects of a native South African granular material stabilized with cement and bitumen emulsion as a base layer in pavement construction. The material stabilized with cement-bitumen emulsion (2-3%) was subjected to Unconfined Compressive Strength (UCS) and Indirect Tensile Strength (ITS) tests for 1, 4, 7 and 28 days curing. The UCS and ITS requirement was evaluated with respect to a base layer for design traffic application of less than six million equivalent single axles. The results of UCS and ITS tests for the stabilized material showed improved strength and have the potential for use as a base course material for the design traffic. The result revealed that 2.5% cement and bitumen emulsion meets the minimum strength characteristics for the base layer. Relative to 2% cement and 2% bitumen emulsion, ITS obtained for 4 and 7 days of curing increased approximately by 24%, 41% and 24%, 53% respectively. Models for UCS in terms of ITS was developed for cement and bitumen emulsion which will make one test among the two sufficient to indicate the strength of cement and bitumen emulsion stabilized materials at the mix design level. Bitumen stabilization is a quick construction method, with lower cost than reconstruction and good for rehabilitation. Keywords— bitumen emulsion, cement, granular, indirect tensile strength unconfined compressive strength. 

Determination of strength and modulus of elasticity of heterogenous sedimentary rocks: An ANFIS predictive technique

The properties of sedimentary rocks, in term of strength and deformation, are significant for various purposes in different fields such as in mine planning and design, reservoir stability evaluation, surface and sub-surface structure designs. This study proposes suitable predictive models for estimation of unconfined compressive strength (UCS) and modulus of elasticity (E) of sedimentary rocks using easy and economically determinable key geomechanical properties. A total of 45 dataset of geomechanical properties including density, porosity, P-wave velocity (Vp), UCS and E were determined through rigorous laboratory tests on samples from Umrer sandstone, Singrauli sandstone and Kutch limestone. In proposed predictive models, density, porosity and Vp were considered as input parameters for prediction of output parameters (UCS and E) using adaptive neuro-fuzzy inference system (ANFIS). The prediction performance of ANFIS models checked through evaluating different membership functions. The suitability of models and their robustness examined by the coefficient of determination (R2), and performance indices such as the mean absolute percentage error (MAPE), variance account for (VAF) and root mean square error (RMSE). Among the five proposed models for UCS, the model 3 was best suited with performance indices MAPE, VAF, and RMSE calculated as 16.53%, 95.60%, and 4.69 respectively and R2 equal to 0.935. Model 3 was best suited for the proposed models for E with MAPE, VAF, and RMSE calculated as 17.84%, 95.43%, and 2.26 respectively and R2 equal to 0.955. The high-performance indices and R2 of proposed models for UCS and E of sedimentary rocks using ANFIS can be confidently used in the field of geotechnical engineering.

New prediction models for unconfined compressive strength of geopolymer stabilized soil using multi-gen genetic programming

This study presents new models for the prediction of unconfined compressive strength (UCS) of geopolymer stabilized clayey soils using a modified branch of genetic programming, called multi-gen genetic programming (MGGP). The proposed MGGP models incorporate several parameters affecting the behavior of the UCS of the clayey stabilized soil. UCS is formulated in terms of percentages of fly ash, ground granulated blast furnace slag, liquid limit, plastic limit, plasticity index, molar concentration, alkali to binder ratio, and ratios of sodium and silicon to aluminum. The importance of each predictor variable is measured through a sensitivity analysis. The validity of the models and the trend of the results are verified by performing parametric study. The parametric study results are also in good agreement with previous studies. The results indicate that the proposed equations are capable of evaluating UCS accurately.

On the Multiple Linear Regression and Artificial Neural Networks for Strength Prediction of Soil-Based Controlled Low-Strength Material

This paper presents two approaches, multiple linear regression (MLR) and artificial neural network (ANN), to develop predictive models for unconfined compressive strength of soil-based controlled low-strength material (CLSM). Our obtained laboratory data conducting on the soil-based CLSM were employed for analysis. Two strength prediction models were proposed: (1) strength is assumed to be a function of mix proportion and curing period; and (2) it is estimated from measured ultrasonic pulse velocity combined with effect of mixture parameters and curing ages. In each model, three predicted formulas were developed; one from MLR and two from ANN. It was showed that all the proposed equations have a well-predicted capacity.

Application of Generalized Regression Neural Networks in Predicting the Unconfined Compressive Strength of Carbonate Rocks

Measuring unconfined compressive strength (UCS) using standard laboratory tests is a difficult, expensive, and time-consuming task, especially with highly fractured, highly porous, weak rock. This study aims to establish predictive models for the UCS of carbonate rocks formed in various facies and exposed in Tasonu Quarry, northeast Turkey. The objective is to effectively select the explanatory variables from among a subset of the dataset containing total porosity, effective porosity, slake durability index, and P-wave velocity in dry samples and in the solid part of samples. This was based on the adjusted determination coefficient and root-mean-square error values of different linear regression analysis combinations using all possible regression methods. A prediction model for UCS was prepared using generalized regression neural networks (GRNNs). GRNNs were preferred over feed-forward back-propagation algorithm-based neural networks because there is no problem of local minimums in GRNNs. In this study, as a result of all possible regression analyses, alternative combinations involving one, two, and three inputs were used. Through comparison of GRNN performance with that of feed-forward back-propagation algorithm-based neural networks, it is demonstrated that GRNN is a good potential candidate for prediction of the unconfined compressive strength of carbonate rocks. From an examination of other applications of UCS prediction models, it is apparent that the GRNN technique has not been used thus far in this field. This study provides a clear and practical summary of the possible impact of alternative neural network types in UCS prediction.

Chemical activation in view of MSWI bottom ash recycling in cement-based systems

In the present study, the feasibility of recycling incinerator bottom ash in cementitious systems by means of chemical activation was investigated. Different Na-, K- and Ca-based hydroxides and salts were selected for the experiments on the basis of their recognized effects on activation of typical pozzolanic materials. The evolution of mechanical properties of bottom ash/Portland cement mixtures and the leaching of trace metals from the materials were a matter of major concern. The experiments were arranged according to a full factorial design, which also allowed to derive a predictive model for unconfined compressive strength as affected by bottom ash content as well as activator type and dosage. Among the activators tested, calcium chloride was found to affect mechanical strength far more positively than the other species used, at the same time ensuring low metal release from the material. On the other hand, the use of potassium sulfate was observed to cause a significant increase in metal leaching at pH < 12, which was probably associated to the release of contaminants initially immobilized within the structure of ettringite as soon as it converted into monosulfate over time.