Search Results (92)

Growing concerns about global climate change and its negative consequences for communities have put immense pressure on the building industry, which is one of the primary sources of greenhouse gas emissions. Due to the environmental issues associated with the manufacture of sustainable construction materials, alkali-activated concrete (AAC) has emerged as a competitive alternative to cement. To predict the compressive strength (CS) of AAC, four machine learning (ML) models, namely, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), were employed in this study using 193 data points. The input variables include Precursor “P” (kg/m³), Blast Furnace Slag “BFS ratio”, Sodium hydroxide “Na” (kg/m³), silicate modulus “Ms”, water content “W” (kg/m³), fine aggregate “FA” (kg/m³), coarse aggregate “A” (kg/m³), and curing time “CT” (day), with CS (MPa) as the output variable. The dataset was checked for stationarity and then normalized to decrease data redundancy and increase integrity. Furthermore, three model combinations were developed based on the relationship between the input and target variables. The XGB-M3 model outperformed all other models with a high degree of accuracy, according to the study’s findings. Specifically, the Pearson correlation coefficient (PCC) was 0.9577, and the mean absolute percentage error (MAPE) was 14.95% during the calibration phase. SHAP, an explainable AI approach that provides interpretable insights into complex AI systems by assigning feature importance to model predictions, was employed. Results suggest the higher predictions from the XGB-M3 and RF-M3 models were largely driven by curing time (CT). Full article

This paper investigates the formation mechanism and key influencing factors of freckle defects that arise during the directional solidification of a novel third-generation nickel-based single crystal superalloy turbine blade. A combined experimental and multi-physics numerical simulation approach was adopted. The results indicate that freckle formation primarily originates from solutal convection, which subsequently triggers a cascade of processes, including the development of convection-induced segregation channels, flow-driven dendrite fragmentation, and the migration and aggregation of dendrite fragments. The severity of freckling is closely dependent on both the casting’s position within the furnace and its local geometric characteristics. Castings located in regions with poorer heating conditions exhibit lower temperature gradients and slower solidification rates, significantly increasing their susceptibility to freckle formation. Similarly, on a given casting, the side subjected to less favorable heating is more prone to freckle initiation. The freckle number varies non-monotonically along the blade height, increasing from 3 to a maximum of 16, with a temporary decrease near the platform and a final reduction near the top. This trend is mainly attributed to thickness-dependent channel segregation, as well as freckle propagation into the interior and coalescence at higher positions. This study provides a crucial theoretical basis for understanding the formation mechanism of freckle defects in nickel-based single crystal superalloys and offers valuable guidance for optimizing blade manufacturing processes, reducing solidification defects, and enhancing blade quality and service performance. Full article

A hybrid modelling framework is proposed to predict endpoint sulphur content in the ladle furnace (LF) refining process by embedding metallurgical expert knowledge into interpretable machine learning (ML). Industrial process data were extracted from the Level-2 (L2) system of a steel plant, and a desulphurisation dataset comprising 5169 heats with 29 process variables was constructed using a knowledge-guided time window from the joint satisfaction of refining conditions to the final argon-blowing stage. After data cleaning, normalisation and correlation-based feature selection, four algorithms—Random Forest (RF), Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM) and Artificial Neural Network (ANN)—were trained and compared on a representative cluster of steel grades identified by K-means. The ANN model achieved a coefficient of determination (R²) of 0.7752, a root mean square error (RMSE) of 0.0027 wt%, a mean absolute error (MAE) of 0.0017 wt% and a hit rate (HR, ±0.0025 wt% for S) of 76.40% on the test set. SHapley Additive exPlanations (SHAP) indicate that limestone addition, slag basicity, argon flow rate, refining time and initial sulphur content dominantly govern sulphur removal. The expert-knowledge-guided, interpretable framework provides quantitative support for specification-conforming endpoint sulphur control while mitigating over-desulphurisation and reagent consumption. Full article

This paper builds a one-dimensional transient numerical model of mixed fuel of woody and non-woody biomass to simulate the multistage conversion process of biomass in a moving grate-fired bed, including drying, pyrolysis, gasification, and char combustion. Based on time and space discretization, the model comprehensively considers the conservation of mass, momentum, and energy. It also introduces reaction kinetics and freeboard radiation coupling effects to more accurately describe the bed temperature distribution and reaction process. The analysis focuses on the effects of different non-woody biomass mixing ratios and moisture content. This provides references for optimization of the design of future furnaces and operating parameters and mixed fuel composition. The simulation results show that, for pure woody biomass, the surface temperature reaches approximately 200 °C in the first zone, followed by char reactions with peak temperatures up to 592 °C. The whole conversion process takes about 62% of the grate length. Increasing the pepper mixing ratio leads to lower bed temperatures due to the higher moisture content. The maximum bed temperature in the first zone decreases from 592 °C for pure wood to 551 °C at 30 wt.% pepper, with delayed pyrolysis and a thinner char reaction zone. When the pepper mixing ratio is below 20 wt.%, the combustion process maintains a stable temperature gradient and a continuous reaction front, compared to the mixing ratio of 30% pepper case. This confirms the feasibility of non-woody biomass application to combustion technology. Although a higher pepper mixing ratio leads to a slight temperature decrease, the reaction remains stable along the grate, indicating reliable combustion performance. Full article

The incorporation of recycled concrete aggregates (RCAs) into self-compacting concrete (SCC) represents a critical sustainable construction strategy addressing both construction waste management and natural resource conservation. However, predicting the compressive strength of recycled aggregate self-compacting concrete (RASCC) remains challenging due to complex nonlinear interactions among mixture parameters. This study develops a robust predictive framework using ensemble machine learning algorithms to accurately estimate RASCC compressive strength across diverse mixture compositions. A comprehensive database comprising 301 experimental specimens with 18 input variables—including curing age, binder components, water-to-binder ratio, recycled aggregate properties, and supplementary cementitious materials—was systematically analyzed. Four advanced modeling approaches were evaluated: Light Gradient Boosting Machine (LightGBM), Categorical Boosting (CatBoost), Stacked Generalization with Ridge regression meta-learner, and Voting ensemble with Non-Negative Least Squares optimization. The Stacking ensemble model demonstrated superior predictive performance on the independent test set, with R² = 0.963, RMSE = 3.321 MPa, and MAE = 2.506 MPa. Rigorous residual analysis confirmed model validity through satisfaction of normality, homoscedasticity, and independence assumptions. SHAP interpretability analysis identified specimen age as the dominant predictor, followed by recycled aggregate density and water-to-binder ratio, while elucidating the complex nonlinear contributions of supplementary cementitious materials including fly ash and ground granulated blast furnace slag. The developed framework demonstrates practical applicability for predicting RASCC compressive strength across conventional to high-performance grades, facilitating sustainable mix design optimization while maintaining structural performance requirements, and advancing circular economy principles through confident integration of recycled aggregates in SCC applications. Full article

As one of China’s “New Four Furnaces”, the city of Hangzhou faces significant heat challenges exacerbated by rapid urbanization. Urban parks offer effective nature-based solutions, but optimizing their multi-dimensional cooling performance—encompassing cooling area (PCA), efficiency (PCE), intensity (PCI), and gradient (PCG)—remains a key challenge. This study quantitatively analyzed the internal and external landscape features of 18 parks in Hangzhou, revealing that park cooling performance is not simply a case of “bigger is better.” We found that parks with more complex shapes and irregular boundaries exhibited higher cooling efficiency per unit area (PCE) compared to larger parks with smooth, simple shapes, though sometimes at the expense of peak PCI. Furthermore, the surrounding built environment is critical: high building density within a 300 m buffer zone was found to significantly impede the spatial extent of the cooling effect (PCA). Based on these findings, we propose that to effectively mitigate urban heat, cities should (1) shift focus away from creating large, isolated parks with smooth boundaries; (2) prioritize a network of smaller, morphologically diverse parks with irregular edges that extend into the community; and (3) enhance each park’s cooling reach through strategies like green streets and tree-lined paths. These approaches offer tangible, actionable guidance for designing high-performance cooling green infrastructure in dense urban environments. Full article

To address the issues of parameter setting, reliance on human experience, and the limitations of traditional model-driven control methods in handling complex nonlinear dynamics in the tin smelting industrial process, this paper proposes a data-driven control approach based on improved deep reinforcement learning (RL). Aiming to reduce the tin entrainment rate in smelting slag and CO emissions in exhaust gas, we construct a data-driven environment model with an 8-dimensional state space (including furnace temperature, pressure, gas composition, etc.) and an 8-dimensional action space (including lance parameters such as material flow, oxygen content, backpressure, etc.). We innovatively design a Dual-Action Discriminative Deep Deterministic Policy Gradient (DADDPG) algorithm. This method employs an online Actor network to simultaneously generate deterministic and exploratory random actions, with the Critic network selecting high-value actions for execution, consistently enhancing policy exploration efficiency. Combined with a composite reward function (integrating real-time Sn/CO content, their variations, and continuous penalty mechanisms for safety constraints), the approach achieves multi-objective dynamic optimization. Experiments based on real tin smelting production line data validate the environment model, with results demonstrating that the tin content in slag is reduced to between 3.5% and 4%, and CO content in exhaust gas is decreased to between 2000 and 2700 ppm. Full article

Biochar, a carbon-rich material produced through the pyrolysis of wood residues and agricultural byproducts, has carbon storage capacity and potential as a low-carbon construction material. This study predicts the compressive strength of cementitious composites in which cement is partially replaced with biochar using machine learning models. A total of 716 data samples were analyzed, including 480 experimental measurements and 236 literature-derived values. Input variables included the water-to-cement ratio (W/C), biochar content, cement, sand, aggregate, silica fume, blast furnace slag, superplasticizer, and curing conditions. Predictive performance was evaluated using Multiple Linear Regression (MLR), Elastic Net Regression (ENR), Support Vector Regression (SVR), and Gradient Boosting Machine (GBM), with GBM showing the highest accuracy. Further optimization was conducted using XGBoost, Light Gradient-Boosting Machine (LightGBM), CatBoost, and NGBoost with GridSearchCV and Optuna. LightGBM achieved the best predictive performance (mean absolute error (MAE) = 3.3258, root mean squared error (RMSE) = 4.6673, mean absolute percentage error (MAPE) = 11.19%, and R² = 0.8271). SHAP analysis identified the W/C and cement content as dominant predictors, with fresh water curing and blast furnace slag also exerting strong influence. These results support the potential of biochar as a partial cement replacement in low-carbon construction material. Full article

Recent applications of low-field NMR in alkali-activated materials (AAMs) often adopt interpretation models developed for Portland cement systems, overlooking the distinct influences of paramagnetic/ferrimagnetic components and free-water redistribution. This study investigates how paramagnetic or ferrimagnetic component and free water distribution influence low-field nuclear magnetic resonance (LF-NMR) and proton density magnetic resonance imaging (PD-MRI) characterization of alkali-activated materials (AAMs). Blast furnace slag, fly ash, and steel slag were activated with NaOH solution at liquid-to-solid ratios of 0.45 and 0.5, and analyzed across top, middle, and bottom layers. Slurries prepared with less mixing water and CaO-rich raw materials exhibited negligible settling and uniform relaxation behavior, whereas those with higher water content and CaO-deficient raw materials showed pronounced stratification, resulting in distinct gradients in signal intensity. The results indicate that the LF-NMR data interpretation of relatively dilute system may be unreliable as the relaxation time of protons will be extended after they transfer from bottom to the top of the slurry. A preliminary method for assessing slurry suitability for LF-NMR characterization is proposed for future validation. Full article

Metal Injection Molding (MIM) is an established, high-precision manufacturing route for small, geometrically complex metallic components, integrating polymer injection molding with powder metallurgy. State-of-the-art feedstock systems, such as Catamold (polyacetal-based), enable catalytic debinding performed in furnaces operating under ultra-high-purity nitric acid atmospheres (>99.999%). The subsequent thermal stages pre-sintering and sintering are carried out in continuous controlled-atmosphere furnaces or vacuum systems, typically employing inert (N₂) or reducing (H₂) atmospheres to meet the specific thermodynamic requirements of each alloy. However, incomplete decomposition or secondary volatilization of binder residues can lead to progressive hydrocarbon accumulation within the sinering chamber. These contaminants promote undesirable carburizing atmospheres, which, under austenitizing or intercritical conditions, increase carbon diffusion and generate uncontrolled surface carbon gradients. Such effects alter the microstructural evolution, hardness, wear behavior, and mechanical integrity of MIM steels. Conversely, inadequate dew point control may shift the atmosphere toward oxidizing regimes, resulting in surface decarburization and oxide formation effects that are particularly detrimental in stainless steels, tool steels, and martensitic alloys, where surface chemistry is critical for performance. This review synthesizes current knowledge on atmosphere-induced surface deviations in MIM steels, examining the underlying thermodynamic and kinetic mechanisms governing carbon transport, oxidation, and phase evolution. Strategies for atmosphere monitoring, contamination mitigation, and corrective thermal or thermochemical treatments are evaluated. Recommendations are provided to optimize surface substrate interactions and maximize the functional performance and reliability of MIM-processed steel components in demanding engineering applications. Full article

The increasing emphasis on sustainability in construction materials has led to a surge of research focused on recycled aggregate self-compacting concrete (RA-SCC). However, the critical gap in predicting the compressive strength of concrete remains challenging because of the nonlinear interactions among the mix’s constituents. The distinct contribution of this study is to develop an interpretable machine learning (ML) framework to accurately forecast the compressive strength of RA-SCC and identify the most influential mix parameters. A dataset comprising 400 experimental samples was compiled, incorporating eight input variables: age, cement strength, cement, fly ash, blast furnace slag, water, recycled aggregate, and superplasticizer, with compressive strength as the output variable. Four ML algorithms such as support vector regression (SVR), random forest (RF), Multilayer Perceptron (MLP), and extreme gradient boosting (XGBoost) were trained and optimized using Bayesian-based hyperparameter tuning combined with 10-fold cross-validation. Among the evaluated models, XGBoost demonstrated superior accuracy, with R² = 0.98 and RMSE = 2.95 MPa during training, and R² = 0.96 with RMSE = 3.25 MPa during testing, confirming its robustness and minimal overfitting. SHAP (SHapley Additive exPlanations) evaluation indicates that superplasticizer, cement, and cement strength were the most dominant factors influencing compressive strength, whereas higher water content showed a negative impact. The developed framework demonstrates that explainable ML can effectively capture the complex nonlinear behavior of RA-SCC, offering a reliable tool for mix design optimization and sustainable concrete production. These findings contribute to advancing data-driven decision making in eco-efficient materials engineering. Full article

The emission of greenhouse gases from fossil fuels creates devastating effects on Earth’s atmosphere. Therefore, a clean energy source is required to fulfill the energy demand. Hydrogen is considered an energy vector, and the production of green hydrogen is a promising approach. Photoelectrochemical (PEC) water splitting is the best approach to produced green hydrogen, but the efficiency is low. To produce hydrogen by PEC splitting water, semiconductor photocatalysts have received an enormous amount of academic research in recent years. A new class of co-catalysts based on transition metals has emerged as a powerful tool for reducing charge transfer barriers and enhancing photoelectrochemical (PEC) efficiency. In this study, copper oxide (CuO) and tin oxide ($Sn{O}_2$) multilayer thin films were prepared by thermal evaporation to create an energy gradient between $Sn{O}_2$ and CuO semiconductors for better charge transfer. To improve the crystallinity and reduce the defects, the prepared films were annealed in a tube furnace at 400 °C, 500 °C, and 600 °C in an argon inert gas environment. XRD results showed that $Sn{O}_2$/CuO-600 °C exhibited strong peaks, indicating the transformation from amorphous to polycrystalline. SEM images showed the transformation of smooth dense film to a granular structure by annealing, which is better for charge transfer from electrode to electrolyte. Optical properties showed that the bandgap was decreased by annealing, which might be diffusion of Cu and Sn atoms at the interface. PEC results showed that the $Sn{O}_2$/CuO-600 °C heterostructure exhibits the solar light-to-hydrogen (STH%) conversion efficiency of 0.25%. Full article

The ordinary Portland cement (OPC) manufacturing process is highly resource-intensive and contributes to over 5% of global CO₂ emissions, thereby contributing to global warming. In this context, researchers are increasingly adopting geopolymers concrete due to their environmentally friendly production process. For decades, industrial byproducts such as fly ash, ground-granulated blast-furnace slag, and silica fume have been used as the primary binders for geopolymer concrete (GPC). However, due to uneven distribution and the decline of coal-fired power stations to meet carbon-neutrality targets, these binders may not be able to meet future demand. The UK intends to shut down coal power stations by 2025, while the EU projects an 83% drop in coal-generated electricity by 2030, resulting in a significant decrease in fly ash supply. Like fly ash, slag, and silica fume, natural clays are also abundant sources of silica, alumina, and other essential chemicals for geopolymer binders. Hence, natural clays possess good potential to replace these industrial byproducts. Recent research indicates that locally available clay has strong potential as a pozzolanic material when treated appropriately. This review article represents a comprehensive overview of the various treatment methods for different types of clays, their impacts on the fresh and hardened properties of geopolymer concrete by analysing the experimental datasets, including 1:1 clays, such as Kaolin and Halloysite, and 2:1 clays, such as Illite, Bentonite, Palygorskite, and Sepiolite. Furthermore, this review article summarises the most recent geopolymer-based prediction models for strength properties and their accuracy in overcoming the expense and time required for laboratory-based tests. This review article shows that the inclusion of clay reduces concrete workability because it increases water demand. However, workability can be maintained by incorporating a superplasticiser. Calcination and mechanical grinding of clay significantly enhance its pozzolanic reactivity, thereby improving its mechanical performance. Current research indicates that replacing 20% of calcined Kaolin with fly ash increases compressive strength by up to 18%. Additionally, up to 20% replacement of calcined or mechanically activated clay improved the durability and microstructural performance. The prediction-based models, such as Artificial Neural Network (ANN), Multi Expression Programming (MEP), Extreme Gradient Boosting (XGB), and Bagging Regressor (BR), showed good accuracy in predicting the compressive strength, tensile strength and elastic modulus. The incorporation of clay in geopolymer concrete reduces reliance on industrial byproducts and fosters more sustainable production practices, thereby contributing to the development of a more sustainable built environment. Full article

Phosphogypsum represents a gypsum-based solid waste originating from phosphoric acid production, which can be exploited for road filling after cement modification. This study delved into the composition design of high-volume phosphogypsum road base materials, aiming to ascertain their feasibility for subgrade filling, and refine the mix ratio. The main content of phosphogypsum was set at three high-proportion intervals of 86%, 88% and 90%, while the total content of inorganic curing agent was fixed at 0.5% of the total material. Within such a total amount, the proportion of bentonite was preserved at 20%, whereas the proportion of waterproofing agent was configured at three gradients of 20%, 25% and 30%, with the remaining part supplemented by powdered sodium silicate. Merged with trace amounts of inorganic curing agents, particularly the waterproofing agent component, the composite cementitious system comprising cement and ground granulated blast-furnace slag (GGBS) was leveraged to augment the key road performance and water stability of high-volume phosphogypsum-based materials. Material strengths were observed to be distinguishable under an array of phosphogypsum contents, which could be explained by the varying proportions of cement, GGBS and waterproofing agent. The test samples and microscopic products were dissected via XRD and SEM, demonstrating that the hydration products of the materials were predominantly C-S-H gel and ettringite crystals. Full article

Industrial processes such as cement, steel, and glass manufacturing rely heavily on fossil fuels for high-temperature heat, presenting a significant challenge for decarbonization. To enable continuous thermal output from intermittent renewable electricity, Electrified Thermal Solutions, Inc. is developing the Joule Hive™ Thermal Battery (JHTB), an electrically heated energy storage system capable of delivering process heat up to 1800 °C. The system employs electrically conductive firebricks (E-Bricks) as both heating elements and thermal storage media, arranged with insulating bricks (I-Bricks) to facilitate gas flow and heat exchange. The work combines experimental and numerical studies to evaluate the thermal, electrical, and structural performance of the JHTB. A small-scale charging experiment was conducted on a single E-Brick circuit in a 1500 °C furnace, showing good agreement with coupled thermal-electric finite element models that account for Joule heating, temperature-dependent properties, radiation, and natural convection. Structural modeling assessed stress induced by thermal gradients. In addition, a high-fidelity conjugate heat transfer model of the full JHTB core was developed to assess system-scale discharge performance, solving conservation equations with SST k-ω turbulence and radiation models. Simulations for two air channel geometries demonstrated the battery’s ability to deliver 5 MW of heat for at least five hours with air temperatures higher than 1000 °C, validating its potential for industrial decarbonization. Full article