Compressive strength prediction of nano-modified concrete: A comparative study of advanced machine learning techniques DOI Creative Commons
X.M. Tao

AIP Advances, Journal Year: 2024, Volume and Issue: 14(7)

Published: July 1, 2024

This study aims to develop predictive models for accurately forecasting the uniaxial compressive strength of concrete enhanced with nanomaterials. Various machine learning algorithms were employed, including backpropagation neural network (BPNN), random forest (RF), extreme gradient boosting (XGB), and a hybrid ensemble stacking method (HEStack). A comprehensive dataset containing 94 data points nano-modified was collected, eight input parameters: water-to-cement ratio, carbon nanotubes, nano-silica, nano-clay, nano-aluminum, cement, coarse aggregates, fine aggregates. To evaluate performance these models, tenfold cross-validation case prediction conducted. It has been shown that HEStack model is most effective approach precisely predicting properties concrete. During cross-validation, found have superior accuracy resilience against overfitting compared stand-alone models. underscores potential algorithm in enhancing performance. In study, predicted results assessed using metrics such as coefficient determination (R2), mean absolute percentage error (MAPE), root square (RMSE), ratio RMSE standard deviation observations (RSR), normalized bias (NMBE). The achieved lowest MAPE 2.84%, 1.6495, RSR 0.0874, NMBE 0.0064. addition, it attained remarkable R2 value 0.9924, surpassing scores 0.9356 0.9706 0.9884 indicating its exceptional generalization capability.

Language: Английский

Development of machine learning models for forecasting the strength of resilient modulus of subgrade soil: genetic and artificial neural network approaches DOI Creative Commons

Laiba Khawaja,

Usama Asif, Kennedy C. Onyelowe

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Aug. 6, 2024

Accurately predicting the Modulus of Resilience (MR) subgrade soils, which exhibit non-linear stress–strain behaviors, is crucial for effective soil assessment. Traditional laboratory techniques determining MR are often costly and time-consuming. This study explores efficacy Genetic Programming (GEP), Multi-Expression (MEP), Artificial Neural Networks (ANN) in forecasting using 2813 data records while considering six key parameters. Several Statistical assessments were utilized to evaluate model accuracy. The results indicate that GEP consistently outperforms MEP ANN models, demonstrating lowest error metrics highest correlation indices (R2). During training, achieved an R2 value 0.996, surpassing (R2 = 0.97) 0.95) models. Sensitivity SHAP (SHapley Additive exPlanations) analysis also performed gain insights into input parameter significance. revealed confining stress (21.6%) dry density (26.89%) most influential parameters MR. corroborated these findings, highlighting critical impact on predictions. underscores reliability as a robust tool precise prediction applications, providing valuable performance significance across various machine-learning (ML) approaches.

Language: Английский

Citations

9
Enhancing earthquakes and quarry blasts discrimination using machine learning based on three seismic parameters DOI Creative Commons
Mohamed S. Abdalzaher, Moez Krichen, Mostafa M. Fouda

et al.

Ain Shams Engineering Journal, Journal Year: 2024, Volume and Issue: 15(9), P. 102925 - 102925

Published: July 1, 2024

Explosions and other artificial seismic sources remain a major risk to human survival. Seismicity catalogs often suffer from contamination, which hinders the differentiation of tectonic non-tectonic events. To address this issue, an automated control system is developed employing machine learning (ML) techniques discriminate between earthquakes quarry blasts (QBs). By using ML approaches, such as probabilistic statistical techniques, QBs can be differentiated natural earthquakes. The proposed method utilizes latitude, longitude, magnitude information improve performance. Evaluation measures, including R2, F1-score, MCC score, others, are employed assess algorithm's effectiveness. Experimental results demonstrate superiority suggested method, achieving success rate 97.21%. algorithm has significant potential for enhancing hazard assessment, supporting urban development planning, promoting safer communities by accurately discriminating man-made earthquake

Language: Английский

Citations

5
Sustainable foundation design: Hybrid TLBO-XGB model with confidence interval enhanced load–displacement prediction for PGPN piles DOI

Tram Bui-Ngoc,

Duy-Khuong Ly, Tan Nguyen

et al.

Advanced Engineering Informatics, Journal Year: 2025, Volume and Issue: 65, P. 103288 - 103288

Published: April 6, 2025

Language: Английский

Citations

0
Predicting Standard Penetration Test N-value from Cone Penetration Test Data Using Gene Expression Programming DOI

Mehtab Alam,

Jianfeng Chen,

Muhammad Umar

et al.

Geotechnical and Geological Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: June 14, 2024

Language: Английский

Citations

2
Prediction Method of Rock Uniaxial Compressive Strength Based on Feature Optimization and SSA-XGBoost DOI Open Access

Huihui Xie,

Peng Lin,

Jintao Kang

et al.

Sustainability, Journal Year: 2024, Volume and Issue: 16(19), P. 8460 - 8460

Published: Sept. 28, 2024

In order to establish an optimal model for reasonably predicting the uniaxial compressive strength (UCS) of rocks, a method based on feature optimization and SSA-XGBoost was proposed. Firstly, UCS predictor system considering petrographic physical parameters, determined systematic discussion factors affecting rocks. Then, selection combining RReliefF algorithm Pearson correlation coefficient proposed further determine optional input features. The XGBoost used prediction rock UCS. process training, Sparrow Search Algorithm (SSA) optimize hyperparameters. Finally, evaluation carried out test performance model. applied validated in granitic tunnel. results show that can effectively predict Compared with simply adopting or parameters as features model, characteristics improve generalization ability effectively. this study is reasonable provide some reference establishing universal accurately quickly

Language: Английский

Citations

0
Modeling of the effect of gradation and compaction characteristics on the california bearing ratio of granular materials for subbase and landfill liner construction DOI Creative Commons
Majed Alzara, Kennedy C. Onyelowe, Ahmed M. Ebid

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Oct. 9, 2024

The California bearing ratio (CBR) of a granular materials are influence by the soil particle distribution indices such as D10, D30, D50, and D60 also compaction properties maximum dry density (MDD) optimum moisture content (OMC). For this reason, packing compactibility play big role in design construction subbases landfills. In research paper, experimental data entries have been collected reflecting CBR behavior used to construct landfill subbase. database was utilized 78-22% predict considering artificial neural network (ANN), evolutionary polynomial regression (EPR), genetic programming (GP), Extreme Gradient Boosting (XGBoost), Random Forest (RF) response surface methodology (RSM) intelligent learning symbolic abilities. relative importance values for each input parameter were carried out, which indicated that value depends mainly on average size (D30, 50 & 60). They showed combined index 66% considered parameters model exercise. This further shows structural particles within D50 range material consistency purposes. Performance study ability models. ANN best performance with accuracy 88%, then GP, EPR RF almost same accuracies 85% lastly XGBoost 81%. Also, RSM produced an R2 0.9464 p-value less than 0.0001. These show decisive superior forecast subbase waste compacted earth liner material. results optimal depended subgrade, subbase, purposes during monitoring phase constructed flexible pavement foundations liners.

Language: Английский

Citations

0
Compressive strength prediction of nano-modified concrete: A comparative study of advanced machine learning techniques DOI Creative Commons
X.M. Tao

AIP Advances, Journal Year: 2024, Volume and Issue: 14(7)

Published: July 1, 2024

This study aims to develop predictive models for accurately forecasting the uniaxial compressive strength of concrete enhanced with nanomaterials. Various machine learning algorithms were employed, including backpropagation neural network (BPNN), random forest (RF), extreme gradient boosting (XGB), and a hybrid ensemble stacking method (HEStack). A comprehensive dataset containing 94 data points nano-modified was collected, eight input parameters: water-to-cement ratio, carbon nanotubes, nano-silica, nano-clay, nano-aluminum, cement, coarse aggregates, fine aggregates. To evaluate performance these models, tenfold cross-validation case prediction conducted. It has been shown that HEStack model is most effective approach precisely predicting properties concrete. During cross-validation, found have superior accuracy resilience against overfitting compared stand-alone models. underscores potential algorithm in enhancing performance. In study, predicted results assessed using metrics such as coefficient determination (R2), mean absolute percentage error (MAPE), root square (RMSE), ratio RMSE standard deviation observations (RSR), normalized bias (NMBE). The achieved lowest MAPE 2.84%, 1.6495, RSR 0.0874, NMBE 0.0064. addition, it attained remarkable R2 value 0.9924, surpassing scores 0.9356 0.9706 0.9884 indicating its exceptional generalization capability.

Language: Английский

Citations

0