Evaluation of water quality indexes with novel machine learning and SHapley Additive ExPlanation (SHAP) approaches

Journal of Water Process Engineering, Год журнала: 2024, Номер 58, С. 104789 - 104789

Опубликована: Янв. 17, 2024

Язык: Английский

---

Forecasting the strength of graphene nanoparticles-reinforced cementitious composites using ensemble learning algorithms

Results in Engineering, Год журнала: 2024, Номер 21, С. 101837 - 101837

Опубликована: Фев. 6, 2024

Contemporary infrastructure requires structural elements with enhanced mechanical strength and durability. Integrating nanomaterials into concrete is a promising solution to improve However, the intricacies of such nanoscale cementitious composites are highly complex. Traditional regression models encounter limitations in capturing these intricate compositions provide accurate reliable estimations. This study focuses on developing robust prediction for compressive (CS) graphene nanoparticle-reinforced (GrNCC) through machine learning (ML) algorithms. Three ML models, bagging regressor (BR), decision tree (DT), AdaBoost (AR), were employed predict CS based comprehensive dataset 172 experimental values. Seven input parameters, including graphite nanoparticle (GrN) diameter, water-to-cement ratio (wc), GrN content (GC), ultrasonication (US), sand (SC), curing age (CA), thickness (GT), considered. The trained 70 % data, remaining 30 data was used testing models. Statistical metrics as mean absolute error (MAE), root square (RMSE) correlation coefficient (R) assess predictive accuracy DT AR demonstrated exceptional accuracy, yielding high coefficients 0.983 0.979 training, 0.873 0.822 testing, respectively. Shapley Additive exPlanation (SHAP) analysis highlighted influential role positively impacting CS, while an increased (w/c) negatively affected CS. showcases efficacy techniques accurately predicting nanoparticle-modified concrete, offering swift cost-effective approach assessing nanomaterial impact reducing reliance time-consuming expensive experiments.

Язык: Английский

---

Metaheuristic optimization algorithms-based prediction modeling for titanium dioxide-Assisted photocatalytic degradation of air contaminants

Results in Engineering, Год журнала: 2024, Номер 23, С. 102637 - 102637

Опубликована: Июль 29, 2024

Airborne contaminants pose significant environmental and health challenges. Titanium dioxide (TiO2) has emerged as a leading photocatalyst in the degradation of air compared to other photocatalysts due its inherent inertness, cost-effectiveness, photostability. To assess effectiveness, laboratory examinations are frequently employed measure photocatalytic rate TiO2. However, this approach involves time-consuming requirements, labor-intensive tasks, high costs. In literature, ensemble or standalone models commonly used for assessing performance TiO2 water contaminants. Nonetheless, application metaheuristic hybrid potential be more effective predictive accuracy efficiency. Accordingly, research utilized machine learning (ML) algorithms estimate photo-degradation constants organic pollutants using nanoparticles exposure ultraviolet light. Six metaheuristics optimization algorithms, namely, nuclear reaction (NRO), differential evolution algorithm (DEA), human felicity (HFA), lightning search (LSA), Harris hawks (HHA), tunicate swarm (TSA) were combined with random forest (RF) technique establish models. A database 200 data points was acquired from experimental studies model training testing. Furthermore, multiple statistical indicators 10-fold cross-validation examine established model's robustness. The TSA-RF demonstrated superior prediction among six suggested models, achieving an impressive correlation (R) 0.90 lower root mean square error (RMSE) 0.25. contrast, HFA-RF, HHA-RF, NRO-RF exhibited slightly R-value 0.88, RMSE scores 0.32. DEA-RF LSA-RF while effective, showed marginally 0.85, values 0.45 0.44, respectively. Moreover, SHapley Additive exPlanation (SHAP) results indicated that rates through photocatalysis most notably influenced by factors such reactor sizes, dosage, humidity, intensity.

Язык: Английский

---

Predicting Penetration Depth in Ultra-High-Performance Concrete Targets under Ballistic Impact: An Interpretable Machine Learning Approach Augmented by Deep Generative Adversarial Network

Results in Engineering, Год журнала: 2025, Номер unknown, С. 103909 - 103909

Опубликована: Янв. 1, 2025

Язык: Английский

---

Concrete 3D printing technology for sustainable construction: A review on raw material, concrete type and performance

Developments in the Built Environment, Год журнала: 2024, Номер 17, С. 100378 - 100378

Опубликована: Фев. 23, 2024

This paper reviews recent developments and proposes perspectives for future research on three-dimensional printing concrete (3DPC). review originally analyses the 3DP applications combined with types that are classified into three groups: functional concrete, sustainable special concrete. The technique shows different effects due to various modification methods (e.g., nano-additive, fibre addition, chemical reagent) challenging requirements anisotropy exploit defect). Summarily, oriented of 3DPC is a double-edged sword, asking optimal structural design engineered cementitious composite (ECC), ultra-high-performance (UHPC), most fibre-improved not propitious all types, such as foam because additional pressure in process poses huge disadvantage stability. also protentional from view features, which represents contribution advanced technology development direction.

Язык: Английский

---

Evaluation of machine learning models for predicting TiO2 photocatalytic degradation of air contaminants

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Июнь 13, 2024

Abstract The escalation of global urbanization and industrial expansion has resulted in an increase the emission harmful substances into atmosphere. Evaluating effectiveness titanium dioxide (TiO 2 ) photocatalytic degradation through traditional methods is resource-intensive complex due to detailed photocatalyst structures wide range contaminants. Therefore this study, recent advancements machine learning (ML) are used offer data-driven approach using thirteen techniques namely XG Boost (XGB), decision tree (DT), lasso Regression (LR2), support vector regression (SVR), adaBoost (AB), voting Regressor (VR), CatBoost (CB), K-Nearest Neighbors (KNN), gradient boost (GB), random Forest (RF), artificial neural network (ANN), ridge (RR), linear (LR1) address problem estimation TiO rate air models developed literature data different methodical tools evaluate ML models. XGB, DT LR2 have high R values 0.93, 0.926 training 0.936, 0.924 test phase. While ANN, RR LR lowest 0.70, 0.56 0.40 0.62, 0.63 0.31 phase respectively. low MAE RMSE 0.450 min -1 /cm , 0.494 0.49 for 0.263 0.285 0.29 stage. DT, 93% percent errors within 20% error XGB 92% 94% with remained highest performing most robust effective predictions. Feature importances reveal role input parameters prediction made by Dosage, humidity, UV light intensity remain important experimental factors. This study will impact positively providing efficient estimate contaminants .

Язык: Английский

---

Predicting the properties of concrete incorporating graphene nano platelets by experimental and machine learning approaches

Case Studies in Construction Materials, Год журнала: 2024, Номер 20, С. e03018 - e03018

Опубликована: Март 2, 2024

Modern infrastructure requirements necessitate structural components with improved durability and strength properties. The incorporation of nanomaterials (NMs) into concrete emerges as a viable strategy to enhance both the concrete. Nevertheless, complexities inherent in these nanoscale cementitious composites are notably intricate. Traditional regression models face constraints comprehensively capturing intricate compositions. Thus, posing challenges delivering precise dependable estimations. Therefore, current study utilized three machine learning (ML) methods, including artificial neural network (ANN), gene expression programming (GEP), adaptive neuro-fuzzy inference system (ANFIS), conjunction experimental investigation effect integration graphene nanoplatelets (GNPs) on electrical resistivity (ER) compressive (CS) containing GNPs. Concrete GNPs demonstrated an fractional change (FCR) strength. measures depict that enhancement was notable at GNP concentrations 0.05% 0.1%, showcasing increases 13.23% 16.58%, respectively. Simultaneously, highest observed FCR reached -12.19% -13%, prediction efficacy proved be outstanding forecasting characteristics For CS, GEP, ANN, ANFIS impressive correlation coefficient (R) values 0.974, 0.963, 0.954, resistivity, exhibited high R-values 0.999, 0.995, 0.987, comparative analysis revealed GEP model delivered predictions for ER CS. mean absolute error (MAE) GEP-CS 14.51% reduction compared ANN-CS substantial 48.15% improvement over ANFIS-CS model. Similarly, displayed MAE 38.14% lower Moreover, GEP-ER 56.80% 82.47% Shapley Additive explanation (SHAP) provided curing age SHAP score. indicating its predominant contribution CS prediction. In predicting ER, content score, signifying estimation. This highlights ML's accuracy properties nanoplatelets, offering fast cost-effective alternative time-consuming experiments.

Язык: Английский

---

A Review of Concrete Carbonation Depth Evaluation Models

Coatings, Год журнала: 2024, Номер 14(4), С. 386 - 386

Опубликована: Март 26, 2024

Carbonation is one of the critical issues affecting durability reinforced concrete. Evaluating depth concrete carbonation great significance for ensuring quality and safety construction projects. In recent years, various prediction algorithms have been developed evaluating depth. This article provides a detailed overview existing models According to data processing methods used in model, can be divided into mathematical curve machine learning models. The further following categories: artificial neural network decision tree support vector combined basic idea model directly establish relationship between age by using certain function curves. advantage that only small amount experimental needed fitting, which very convenient engineering applications. limitation it consider influence some factors on concrete, accuracy cannot guaranteed. predict many considered at same time. When there are sufficient data, trained give more accurate results than model. main defect needs lot as training samples, so not A future research direction may combine with evaluate accurately.

Язык: Английский

---

Utilizing contemporary machine learning techniques for determining soilcrete properties

Earth Science Informatics, Год журнала: 2025, Номер 18(1)

Опубликована: Янв. 1, 2025

Abstract Soilcrete is an innovative construction material made by combining naturally occurring earth materials with cement. It can be effectively used in areas where other are not readily available due to financial or environmental reasons since soilcrete from natural clay. also help cut down the greenhouse gas emissions industry encouraging use of resources that locally available. Thus, it imperative reliably predict different properties accurate determination these crucial for widespread materials. However, laboratory subjected significant time and resource constraints. As a result, this research was undertaken provide empirical prediction models density, shrinkage, strain mixes using two machine learning algorithms: Gene Expression Programming (GEP) Extreme Gradient Boosting (XGB). The analysis revealed XGB-based predictions correlated more real-life values than GEP having training $${\text{R}}^{2}=0.999$$ R 2 = 0.999 both density shrinkage $${\text{R}}^{2}=0.944$$ 0.944 prediction. Moreover, several explanatory analyses including individual conditional expectation (ICE) shapely were done on XGB model which showed water-to-binder ratio, metakaolin content, modulus elasticity some most important variables forecasting properties. Furthermore, interactive graphical user interface (GUI) has been developed effective utilization civil engineering forecast

Язык: Английский

---

Development and optimization of geopolymer concrete with compressive strength prediction using particle swarm-optimized extreme gradient boosting

Applied Soft Computing, Год журнала: 2025, Номер unknown, С. 113149 - 113149

Опубликована: Апрель 1, 2025

Язык: Английский

---