Algal blooms in eutrophic lakes are shaped by non-linear responses to both current and antecedent environmental conditions, while their dominant drivers can vary across space and bloom stage. This study developed an explainable machine-learning framework for predicting algal density using high-frequency monitoring data collected from six stations in Lake Taihu, China, during 2021–2024. The lake was divided into eastern and western functional zones, and separate models were developed for stable, outbreak, and decline stages. Two predictor schemes were compared: W1 contained contemporaneous environmental variables, whereas W2 additionally incorporated 3–15 d lagged, moving-average, and cumulative predictors. Across four algorithms and seven data subsets, W2 outperformed W1 in all 28 matched comparisons: mean test-set R2 increased by 0.140, while MSE, RMSE, and MAE decreased by 41.2%, 24.0%, and 27.8%, respectively. XGBoost achieved the best overall performance, with a maximum test-set R2 of 0.901. An integrated ranking based on model-specific importance, permutation importance, and SHapley Additive exPlanations (SHAP) showed strong context dependence. Solar radiation, total phosphorus (TP), and the 15-d mean N:P ratio dominated the global model; TP and total nitrogen were prominent during eastern functional-zone outbreak and decline stages, whereas antecedent pH, permanganate index, precipitation, turbidity, conductivity, and N:P were more influential in western functional-zone stage-specific models. BIC-supported segmented SHAP analysis with bootstrap confidence intervals identified reproducible model-derived transition ranges, including stage-dependent TP transitions on the eastern functional zone, pH transitions near 7.8 on the western shore, and N:P transitions during bloom decline. These results demonstrate that antecedent environmental information improves prediction and that spatial-stage stratification reveals local response regimes that are obscured in a lake-wide model.
Abstract To address the issues of low strength, uneven particle distribution, and weak interfacial bonding in ultralight expanded polystyrene (EPS) concrete, this study proposes a powder precoating process. This process involves precoating the surface of EPS particles with a layer of powder material. The differences in the effects of this process and the ordinary process on the properties of ultralight EPS concrete were systematically compared. The results showed that the precoating process increases the compressive strength by 21.4%–25.6% for ultralight EPS concrete with a low density grade of 200 – 400 kg / m 3 , and the flexural strength is increased by up to 40.6% for the concrete with a density grade of 600 kg / m 3 . Additionally, it can effectively alleviate the phenomenon of reduced concrete strength caused by the incorporation of fly ash. The uniformity index of ultralight EPS concrete prepared by the precoating process reached 0.78, whereas that of the ordinary process was 0.59. Cross-section observations confirmed that the EPS particles are more evenly distributed, the number of pores is reduced, and the delamination phenomenon is basically eliminated. The slope of the thermal conductivity of the precoating process with increasing density decreases from 1.64 × 10 − 6 (of the ordinary process) to 1.57 × 10 − 6 , indicating that the precoating process provides better thermal insulation performance at the same density. Microscopic analysis revealed that the precoating process improves compactness by optimizing the structure of the interfacial transition zone, promoting the formation of calcium silicate hydrate (C─S─H), and reducing the residual calcium hydroxide ( Ca ( OH ) 2 ), without introducing new chemical bonds. This process provides technical support for the application of ultralight EPS concrete in energy-efficient buildings.
Abstract Alkali-activated lunar regolith is a promising extraterrestrial material for large-scale construction of lunar bases. However, the complex coupling effects of multiple influencing factors present significant challenges to strength prediction and mechanistic interpretation of alkali-activated lunar regolith simulant (AALRS) under extreme lunar environments. This study proposes an interpretable Stacking ensemble model integrated with hyperparameter optimization algorithms. The dataset includes features related to the lunar regolith simulant properties, mixture proportions, and environmental conditions that represent realistic lunar in situ construction scenarios. After tuning hyperparameters using six optimization algorithms, the Stacking model integrated with random search (RS) achieves the highest predictive accuracy and computational efficiency among the tested models [ R = 0.9482 , root-mean-square error ( RMSE ) = 5.58 MPa , mean absolute error ( MAE ) = 4.07 MPa , and computation time = 0.77 min ]. Feature contributions to strength and their interactions were analyzed using Shapley Additive Explanations (SHAP) and partial dependence plots (PDPs). Alkali content (AC), curing age (CA), and temperature were identified as the dominant features. Among all input features, the interaction between alkali content and initial temperature (IT) exhibits the highest interaction value. Parameters in the ranges of median particle diameter (MPD) < 100 μm , water-to-binder ratio (W/B) < 0.3 , CA > 7 days , AC > 0.06 , and IT > 60 ° C consistently exert positive contributions to predicted strength. Furthermore, a practical graphical user interface (GUI) based on the Stacking-RS model was developed to enhance user accessibility. This study establishes a data-driven paradigm that provides a generalizable framework for predicting and designing AALRS materials performance under extreme lunar environments.
Automated concrete wall construction requires coordinated wall-configuration design and loading schedules to control early-age settlement and lateral deformation. This paper combines early-age tests, matched-age finite-element simulations, and analytical calculations to evaluate the responses of 3D printed recycled sand concrete (3DPCRS) walls to different configurations and prescribed loading histories. Results showed that the four-rib configuration provided a favorable balance between lateral-deformation control and material use, whereas additional ribs did not further improve deformation control. The analytical assessment identified a net increase in the material-yield margin between successive loading stages, with the calculated gain in available material yield resistance exceeding the corresponding rise in equivalent construction demand. Compared with natural-sand 3D printed concrete, 3DPCRS specimens exhibited a 71% lower maximum short-term creep displacement but steeper post-peak softening, underscoring the distinct sustained-load and post-peak deformation responses. These findings inform pre-print wall-configuration comparison and preliminary loading-schedule screening in automated concrete construction.
The development of durable electrocatalysts for water and seawater electrolysis is hindered by the lack of a unified strategy for anodic/cathodic stability and chloride-induced corrosion. A hierarchical porous medium-entropy sulfide nanoflower, (NiFeCoMo)S-2, achieves ultra-stable water and seawater electrolysis through a multi-level design combining d-d coupling-enhanced M-S d-p hybridization with a hierarchical porous nanoflower architecture. Mechanistic studies in 1 M KOH reveal that the medium-entropy configuration restructures the electronic landscape, enhancing d-d orbital interactions and optimizing d-p hybridization. This drives a condition-adaptive surface evolution: OER reconstructs into a TMOOH active layer; HER maintains a stable low-valence sulfide backbone. The hierarchical porous nanoflower architecture provides a large surface area of 123.2 m(2)/g, facilitating mass transport. Leveraging this design, the catalyst achieves low overpotentials (237 mV for OER and 125 mV for HER at 100 mA cm(-2)) and exceptional durability exceeding 3000 h in 1 M KOH. When extended to alkaline natural seawater, it maintains excellent performance (267 mV for OER and 163 mV for HER at 100 mA cm(-2)) and > 2000 h of stability. These results highlight multimetal orbital-interaction-mediated modulation of M-S bonding and hybridization as a promising strategy for designing durable bifunctional sulfide electrocatalysts.
