Search Results (2,844)

Soil erosion is a critical global challenge and presents particularly alarming characteristics in the Amazon–Cerrado transition in Mato Grosso, a leading agricultural state in Brazil. Therefore, soil and water losses were evaluated under three soil cover conditions for corn and pasture cultivation (with vegetation cover, without vegetation cover, and without vegetation cover with soil scarified to a depth of 0.10 m) and four precipitation patterns (Advanced, Intermediate, Delayed, and Constant). The results showed that both soil cover and rainfall patterns directly influence the erosion processes. Soil loss increased up to sixfold under the Intermediate compared to the Constant rainfall, highlighting the strong influence of rainfall temporal distribution on erosion dynamics. The highest maximum runoff rates (MRR) and soil losses (SL) were recorded in tilled plots under maize cultivation, reaching 98.57 mm h⁻¹ and 5.90 g m⁻², respectively, under the intermediate pattern. In pasture areas, SL followed a similar pattern to the maize area, with maximum values of 6.96 g m⁻², but the MRR was recorded under the advanced pattern and in plots with cover (89.71 mm h⁻¹). This may be attributed to soil management conditions in pasture areas. Advanced and Intermediate patterns resulted in greater soil losses (3.78 and 2.04 g m², respectively), highlighting the impact of peak intensity timing on soil erosion. Greater soil losses were observed at the pasture experimental site than at the maize experimental site. Because the experiments were conducted at different locations with contrasting soil and management conditions, these differences should not be interpreted as being caused exclusively by crop type. The current study reinforces the need for erosion control and management strategies that account for natural variations in rainfall and soil cover to mitigate the negative impacts of land degradation on agricultural production and environmental sustainability. Full article

Background: Large language models (LLMs) are increasingly explored for automated educational assessment, while future educational platforms may combine grading, feedback, learner analytics, and personalization. The objective of this study was to determine whether student-history metadata can influence the numerical score assigned to the same answer. Methods: This study introduces GradeDrift-LLM, a controlled framework for measuring student-history-induced score drift in LLM-based automated grading. We evaluated 1000 Computer Science answers from 100 students across six student-history conditions and eight open-weight LLMs. For each grading instance, the submitted answer, question, reference answer, rubric-related information, scoring scale, and grading instruction were kept constant; only the student-history condition varied. Results: Across 39,997 valid paired comparisons, 83.92% showed no drift, 9.40% showed upward drift, and 6.68% showed downward drift. Mean absolute drift was 0.2137 points, and the 95th percentile absolute drift was 1 point. Positive-history frames tended to increase scores, whereas negative-history frames tended to decrease them. Drift was model-dependent, not uniformly explained by approximate scale, and present in both technical and argumentative answers; rare extreme deviations reached 10 points. Conclusions: Student-history metadata can influence LLM-generated grading scores despite explicit instructions to ignore it. Future LLM-based grading systems should separate answer-based scoring from learner-context-based personalization and validate score invariance under controlled learner-context variations. Full article

Large-scale and long-term hydrogen storage is a key requirement for a sustainable hydrogen-based energy system. Although underground hydrogen storage (UHS) in porous media has gained increasing attention, the behavior of hydrogen during cyclic injection and withdrawal remains poorly understood. This study experimentally investigates the effects of cyclic hydrogen injection cycles (3, 6, 9, and 12 cycles) on the physical and mechanical properties of both dry and brine-saturated Berea, which serves as a representative reservoir rock. Porosity, permeability, and mineral composition of the samples were measured before and after the injection cycles, while triaxial tests were conducted post-injection and compared with reference sister samples. Results show negligible and inconsistent mineralogical changes before and after hydrogen injection. Porosity remained nearly constant (<2 percentage variation), whereas permeability declined by more than 20% in samples, which can impact recovery efficiency. Mechanical properties remained largely unchanged, indicating stability. However, further experimental and modeling studies are required to better understand the observed permeability reduction and its implications for underground hydrogen storage (UHS). Full article

Engine-inlet fuel pressure in helicopters can fluctuate during flight manoeuvres because load-factor variation changes both fuel distribution in the tanks and the pressure balance along the fuel-supply pipeline. This simulation-only study investigates a passive pressure-regulation scheme based on a constant-pressure reducing valve (CPRV) for a representative five-tank helicopter fuel system. Mathematical models of the fuel tank, booster pump, jet pump, check valve, fuel-supply pipeline, and CPRV are integrated in AMESim and checked against reported tank-depletion data, fuel-centre-of-gravity data, and code-to-code benchmark results. A controlled same-model isolation check additionally compares a mobile CPRV spool with the same spool constrained at its fully open end stop while all pump, tank, line, demand, load, fluid, reference-pressure, and solver settings remain unchanged. In a longitudinal load-factor ramp to 1 g, the mobile-spool model maintains 1.8042–1.8047 barA, whereas the locked-open control gives 2.4287–2.5917 barA. Across the principal regulated simulations, the maximum absolute deviation from the 1.8 barA target is approximately 0.005 bar. These values are numerical results for the stated model and must not be interpreted as sensor-resolvable hardware accuracy or qualification evidence. Full article

The DC–DC boost converter is a challenging control target because of its nonlinear dynamics, wide operating range, and non-minimum-phase behavior under continuous conduction mode. These control challenges are particularly pronounced under large-signal transients, parameter variations, constant-power-load effects, and hardware constraints. This review examines deep reinforcement learning-based control of DC–DC boost converters from an engineering-oriented perspective. It covers learning-assisted classical control, direct duty-cycle control, and hybrid architectures, with attention to action design, reward formulation, observation timing, safety constraints, and validation fidelity. A structured search of Scopus, Web of Science Core Collection, and IEEE Xplore was used to identify boost-specific studies and transferable adjacent-converter evidence. Rather than ranking algorithms alone, the review organizes the literature around converter-aware and hardware-oriented learning control. The review argues that recent progress should not be interpreted as a simple replacement of classical control by deep reinforcement learning. Accordingly, algorithm choice, physical knowledge, action and reward design, observation timing, safety constraints, and validation fidelity are treated jointly. The available evidence suggests that progress toward credible practical deployment requires integrating converter physics, bounded or hybrid control authority, explicit safety constraints, and hardware-oriented validation. Full article

Predicting the fatigue lives of high-performance alloys, specifically aluminum 7075-T6 and Inconel 718, is essential for ensuring structural integrity in applications such as aerospace and energy. While Poisson’s ratio is typically treated as a constant within fracture mechanics and finite element analysis, it has been found to vary significantly with increased temperatures and substantial amounts of plastic deformation. Variations in Poisson’s ratio can, therefore, have a significant impact on local stress fields around cracks and the behavior at crack tips. This study introduces a novel approach by systematically isolating the effects of varying Poisson’s ratios on fatigue life cycles, stress distributions, and fatigue crack growth using finite element analysis with the robust ANSYS SMART crack growth feature. The results indicate a stark difference in the effects of Poisson’s ratio on the fatigue life of aluminum 7075-T6 compared to Inconel 718. A strong negative correlation exists between Poisson’s ratio and fatigue life cycle numbers for aluminum 7075-T6, whereas a more linear trend is observed for all fatigue life cycle numbers of Inconel 718. The underlying reasons for these trends lie in the differing sensitivities of elastic, shear, and bulk moduli between the two alloys. Overall, a higher Poisson’s ratio intensifies the maximum principal stress for both alloys. Additionally, an increase in Poisson’s ratio leads to a decrease in von Mises stress for both metals. Furthermore, these numerical results demonstrate that an increase in Poisson’s ratio corresponds to a decrease in the cyclic plastic zone size at the crack tip for both alloys, indicating enhanced hydrostatic constraint and reduced shear deformation. The findings presented herein underscore the necessity of eliminating the use of static values for Poisson’s ratio when evaluating the structural performance of high-performance alloys under extreme operational environments. Additionally, this research highlights several key areas where existing modeling approaches are lacking and establishes a framework for developing improved constitutive models for fatigue life prediction. Full article

Multi-sensor data fusion is essential for accurate target tracking and trajectory reconstruction. However, common forms of systematic error in multi-sensor observations, including constant biases, linear drifts, and saturating exponential drifts, can degrade measurement consistency and trajectory estimation accuracy. Within the B-spline-constrained Error Model Best Estimate of Trajectory (EMBET) framework, B-spline coefficients and systematic-error parameters may produce similar observation responses, allowing part of the systematic-error response to be absorbed into the spline-coefficient correction and thereby weakening the identifiability of the systematic-error parameters. To avoid the weak-identifiability mechanism associated with the joint parametric estimation of trajectory and systematic-error terms, a residual-driven ResCompFormer method is proposed for systematic-error compensation and target trajectory reconstruction. First, a B-spline-constrained EMBET model is established to analyze the coupling between B-spline coefficients and systematic-error parameters. Systematic-error estimation is then removed from the joint EMBET parameter-estimation problem and reformulated as observation-domain error-sequence prediction, and ResCompFormer is employed to capture temporal dependencies and cross-channel correlations in multi-sensor residuals. The predicted errors are fed back to correct the observations, followed by iterative trajectory re-estimation. Simulation results confirm the systematic-error absorption mechanism and show that the proposed method outperforms the considered model-driven and data-driven methods in both systematic-error compensation and trajectory reconstruction, including iterative and stepwise EMBET variants. Additional experiments demonstrate the robustness of the proposed method to variations in systematic-error characteristics and sensor availability. Full article

Soil microbial carbon use efficiency (CUE) is an essential metric for assessing soil carbon cycling efficiency. In contemporary biogeochemical models, microbial CUE is often treated as a constant, but it is intricately regulated by biotic and abiotic factors, especially in heterogeneous karst rocky desertification (KRD) ecosystems. In the Guizhou Huajiang KRD area, this study used ecological stoichiometry, high-throughput sequencing, analysis of variance, and partial least squares regression to evaluate CUE across three restoration types, Zanthoxylum bungeanum (ZB), Selenicereus undatus (SU), and Pennisetum × sinese (PS), with Zea mays (ZM) as control. Results showed that restoration types affected CUE. Soil bulk density (SBD), total potassium (TK), and available phosphorus (AP) were key predictors of CUE, significantly correlating with generalist communities such as Planctomycetota, Gemmatimonadota, unclassified_Gemmatimonadaceae, unclassified_Roseiflexaceae, and unclassified_Trichomeriaceae. Inconsistent variations in α- and β- diversity across types indicated CUE regulation was driven by integrated structural and functional responses. These findings highlight the importance of soil physical properties, potassium cycling, low-phosphorus adaptation, and generalist communities for species selection. Future studies must isolate vegetation and management effects, extending observations from winter to the full growing season to clarify seasonal plant-soil-microbial feedbacks. Full article

We prove the existence and uniqueness of weak solutions to an obstacle problem for a nonlinear elliptic operator in divergence form. The variational inequality under consideration involves an integral over the domain of the Frobenius inner product of the operator $\mathcal{S}(z,\nabla u-\mathcal{O}(u\left)\right)$ with the gradient difference $\nabla (v-u)$, plus the Euclidean inner product of u and $v-u$, which is required to be nonnegative for all admissible functions v. The admissible set consists of functions in the Sobolev space $W^{1,2}(\mathrm{\Omega };{\mathbb{R}}^m)$ with prescribed Dirichlet boundary trace and lying above a given obstacle $\psi$ almost everywhere. The obstacle condition $v\ge \psi$ a.e. models a lower bound constraint (e.g., a membrane or a displacement limit) that the admissible functions must respect, while the boundary value $\delta$ prescribes the Dirichlet data. The principal part contains a solution-dependent shift $\mathcal{O}\left(u\right)$, which is Lipschitz continuous, while $\mathcal{S}$ is assumed to be globally Lipschitz and strongly monotone with respect to equal shifts, with quadratic growth and coercivity. This structural framework can be interpreted in terms of symmetry: the strong monotonicity condition expresses a quantitative symmetry property of $\mathcal{S}$ with respect to equal shifts, and the shift $\mathcal{O}\left(u\right)$ introduces a symmetry-breaking coupling. The smallness condition ensures that this asymmetry remains under control. However, we do not pursue a full group-invariance or Lie-symmetry analysis; the symmetry perspective is used here as a heuristic and interpretative tool. The main difficulty lies in the mismatch of shifts when comparing two admissible functions. This is resolved by a three-term decomposition of the monotonicity estimate, combined with Young’s inequality and Poincaré’s inequality, under the smallness condition that the product of the Lipschitz constant of $\mathcal{S}$, the Lipschitz constant of $\mathcal{O}$, and the Poincaré constant is bounded above by one quarter of the strong monotonicity modulus. Existence follows from the Kinderlehrer–Stampacchia theorem; uniqueness is obtained from the same decomposition. The result unifies and extends previous contributions that treated either the lower-order term or the shift coupling separately, and it does so within a unified quadratic framework that avoids the technical overhead of variable exponents and Young measures. Full article

Photovoltaic (PV) pumping systems offer a sustainable alternative to diesel solutions, yet their nonlinearity, intermittent irradiation, and complex motor-pump dynamics challenge energy extraction and reliability. Currently, these systems predominantly rely on classical Maximum Power Point Tracking (MPPT) algorithms such as Perturb and Observe (P&O) and Incremental Conductance (INC), which suffer from slow convergence, steady-state oscillations, and an inability to track Global MPP (GMPP) under uniform irradiance variation conditions. Furthermore, existing studies typically address MPPT optimization and motor control in isolation, without considering their coupled interaction, and rarely incorporate economic viability assessments. To address these limitations, this paper proposes an innovative control architecture integrating four advanced metaheuristic MPPT techniques, namely the Genetic Algorithm (GA), Gray Wolf Optimizer (GWO), Cuckoo Search (CS) algorithm, and Horse Herd Optimization Algorithm (HOA), with Model Predictive Control (MPC) for a Brushless DC (BLDC) motor-driven pumping system, supplemented by battery storage. Comprehensive simulations were conducted under both constant and variable irradiance profiles (1000 to 500 to 1000 W/m²) to evaluate dynamic performance, tracking accuracy, and system robustness. The results demonstrate that HOA and GWO significantly outperform GA and CS, achieving superior DC bus voltage stability with ripple values below 2.4 V, faster convergence times, reduced electromagnetic torque oscillations, and enhanced MPPT efficiency exceeding 99%. Under variable irradiance, HOA exhibits the fastest stabilization with minimal overshoot and superior disturbance rejection, while GA suffers from severe oscillations and CS displays sawtooth ripple patterns. A techno-economic analysis further confirms the economic viability of the proposed system, with HOA and GWO strategies yielding lower lifecycle costs, extended converter lifespans from 5 to over 12 years, and improved return on investment compared to conventional approaches. This integrated framework offers a robust, efficient, and economically sustainable solution for autonomous PV water pumping applications. Full article

ZIKV is an emerging mosquito-borne pathogen with the unique capability to cause severe congenital abnormalities. Mosquito-borne viruses are reliant on an optimal temperature within the mosquito host to spread to humans. Few studies explore how day/night temperature variations impact virus replication within the mosquito. Most laboratory studies are conducted at constant temperatures, or with very minimal temperature oscillations. In consequence, ZIKV transmission models and risk maps utilize biological data derived from constant temperature studies. We sought to determine if the use of viral replication data obtained in more biologically relevant, fluctuating temperatures alters the geographic range for ZIKV transmission risk. We used a growing degree day (GDD)-based model to determine ZIKV transmission potential across the US. Replication curves were conducted at eight different average temperatures between 18°C and 32°C with daily fluctuations of either +/−2.5°C or +/−5°C from the baseline. We found that at lower temperatures (18–22°C), fluctuation was beneficial for virus replication, where it had less of an impact at more optimal temperatures (24–30°C). Our transmission potential map exhibited the highest proportion of infected mosquitoes in southeastern United States, with discrete pockets of elevated potential across interior regions, including the southern Great Plains and lower Mississippi Valley. Full article

Luminescence thermometry has evolved through decades of research focused on optimising materials and on extracting temperature information from isolated spectral features such as luminescence intensity ratios, bandwidth, line shift and excited-state lifetime. Despite extensive material development, these conventional methods remain fundamentally limited by construction: only a small subset of pre-selected spectral features is exploited, while the bulk of the temperature-relevant information encoded in the full spectrum is systematically discarded. A paradigm shift is presented here: Luminescence Thermometry 2.0 (LT 2.0), implemented through the newly developed SpectraSensML platform, in which machine learning regression operates on the entire emission spectrum to deliver temperature readout. The approach is demonstrated on a Yb³⁺-doped phosphor emitting in the near-infrared biological transparency window across 100 to 700 K. Yb³⁺ is a particularly demanding case: only the single ²F_5/2 multiplet emits, and its weakly thermally coupled Stark sub-levels yield modest sensitivity under conventional intensity-ratio thermometry. A total of 27 regression algorithms drawn from four families, namely tree ensembles, physics-aware regression models, kernel and instance methods, and neural networks, are systematically benchmarked. A sensor-fusion estimator that combines the first three principal components reaches an average root-mean-square error of 0.36 K on an unseen-temperature test set, a seven-fold improvement over the best luminescence intensity ratio variant. Standard normal variate (SNV) normalisation is identified as the most effective preprocessing strategy because it isolates the band-shape deformations that encode temperature. Single-component approaches that rely on the first principal component alone are shown to be quantitatively sub-optimal: multi-component regressors that exploit the first three principal components reduce the temperature uncertainty by close to an order of magnitude. The structural reason behind the failure of decision-tree ensembles on unseen temperatures is explained: their piecewise-constant predictions cannot interpolate beyond training set-points. The open-source SpectraSensML application used to obtain the results is released alongside the manuscript to enable reproducible community benchmarks. Full article

Agricultural and livestock production plays an important role in Inner Mongolia, China, but its continued development poses challenges for greenhouse gas mitigation. Using multi-source data from 2005 to 2023, this study integrates the IPCC emission-factor approach, panel STIRPAT modeling, uncertainty and sensitivity analyses, and Ridge regression forecasting to quantify carbon-equivalent emissions and emission intensity from selected agricultural and livestock sources, examine emission changes and associated factors, and project future emissions under a trend-continuation scenario (TCS) and a policy-target scenario (PTS). Total emissions increased from 9.8472 Mt C-eq in 2005 to 14.5223 Mt C-eq in 2023, while carbon-equivalent emission intensity declined over the same period. Livestock-related emissions remained dominant, although the contribution of agricultural emissions increased over time, with cattle and sheep accounting for most livestock-related emissions. Panel estimates showed significant positive associations between emissions and year-end rural population, year-end large-livestock inventory, and real agricultural and animal husbandry output value per capita at constant 2005 prices. Monte Carlo analysis showed that emission-factor uncertainty had a notable influence on absolute emission levels, with a coefficient of variation of 9.04% in 2023; enteric-fermentation CH₄ emission factors for cattle and sheep were the dominant sources of uncertainty. Among the models compared, Ridge regression showed the strongest overall validation performance. Emissions are projected to continue increasing through 2035 under both the TCS and PTS, reaching 16.5974 and 16.5012 Mt C-eq, respectively, indicating that the currently quantifiable policy constraints alone are insufficient to reverse the projected upward trend. Future mitigation should therefore prioritize major livestock-related emission sources and regionally differentiated management, while further refinement of the emission inventory should focus on developing more regionally representative livestock emission factors. Full article

Residual stresses generated during laser powder bed fusion (LPBF) can cause substantial distortion in slender structures, compromising dimensional accuracy and structural reliability. This study tests the hypothesis that, under identical nominal processing conditions, geometry-dependent stiffness and constraint govern how the evolving thermally induced stress field is redistributed and manifested as warpage after support removal. Bridge-type TA15 titanium alloy cantilever specimens with different spans, thicknesses, and support densities were fabricated by LPBF, and their post-cut warpage was quantified by three-dimensional scanning. A coupled thermo-mechanical finite element model was validated against the measured deformation profiles and subsequently used to examine simulated stress evolution during deposition and redistribution after support removal. Cantilevers with different spans approached similarly high simulated surface tensile-stress plateaus in the constrained as-built state but exhibited markedly different measured warpage after cutting, showing that the as-built stress magnitude alone does not reliably rank post-release deformation. Increasing span reduced global flexural resistance and enlarged the effective bending arm, whereas increasing thickness enhanced flexural rigidity and suppressed curvature even when relatively high localized stress was retained. With the total support volume held constant, changing support density altered the system-level constraint through the combined effects of support-leg stiffness, support spacing, local thermal and mechanical response, and deformation compatibility. Together, these results provide an experimentally supported process-structure interpretation of LPBF cantilever distortion across controlled variations in span, thickness, and support distribution. Full article

The problem of load frequency control (LFC) becomes more severe with the extensive integration of photovoltaic (PV) generation owing to the intermittent nature of the source. In this study, a machine learning approach is developed to design the proportional–integral–derivative (PID) controller of a two-area PV–thermal LFC system, extending a prior proportional–integral (PI) benchmark to full PID action. A Random Forest model is trained to predict the relationship between the six PID gains and the closed-loop integral of time-multiplied absolute error (ITAE), yielding an accurate performance model (test R² = 0.933) that is subsequently searched by a metaheuristic optimizer to determine the controller gains; the resulting controller is termed ML-PID. The novelty of the approach lies in employing the learned model not as a controller or a physical-quantity predictor, as in existing ML-based LFC studies, but as a reusable performance model that maps the controller gains directly to the closed-loop index and guides the PID design. To isolate and quantify the contribution of the learned model, the same three optimizers, namely the Cheetah Optimizer (CO), the Grey Wolf Optimizer (GWO), and Particle Swarm Optimization (PSO), are also applied directly to the plant, yielding purely metaheuristic controllers (CO-PID, GWO-PID, and PSO-PID) that are compared against the machine learning-assisted designs under identical algorithms and computational budget, with CO selected on the basis of the Friedman and Wilcoxon tests. The proposed ML-PID-CO controller attains the minimum ITAE under a step-load disturbance, approximately 70% lower than that of the reference SCHO-PI controller and comparable to the directly optimized controllers, with reduced control effort. Under a simultaneous variation in the plant time constants, it is the most robust of all controllers, exhibiting the smallest Δf₁ undershoot and a performance that degrades about 4.2 times less than that of the reference. The results show that a learned performance model provides a good and reusable basis for PID design. It can be searched over repeatedly once built and reduces the per-design simulation burden relative to direct metaheuristic tuning, while the design is largely independent of the optimizer used. Full article