Search Results (1,710)

In this study, the continuous strengthening behavior of CoCrNi medium-entropy alloy at 1.2–4.2 nm twin spacings was revealed by molecular dynamics simulation. It was found that the yield strength increased linearly with the decrease in twin spacing, up to 12.526 GPa, and there was no softening inflection point. The strengthening mechanism is mainly due to the effective obstruction of coherent twin boundaries (TBs) to the dislocation slip, especially the stair-rod and Lomer–Cottrell lock structures generated by ISF and ESF stacking faults when crossing the interface. These structures significantly enhance the work-hardening capacity of the alloy by inducing dislocation stacking, although the very dense twin boundary will reduce the dislocation growth rate by limiting dislocation propagation. This precise interface control provides an important atomic-scale basis for the design of novel high-strength and high-work-hardening alloys. Full article

Although cloud storage is widely trusted by users and enterprises, externally stored encrypted and fragmented data remain vulnerable to reconstruction and inference attacks following partial exposure. Existing decoy-based defenses often rely on static configurations or randomly generated artifacts that can be filtered during adversarial analysis. This paper presents an Artificial Intelligence (AI)-controlled modular decoy generation method to enhance reconstruction resistance in distributed storage systems. The method operates as a system-agnostic post-fragmentation layer and does not require modification of encryption or storage architecture. Given encrypted fragments as input, decoys are generated using a supervised Extreme Gradient Boosting (XGBoost) regression model that adapts decoy quantity based on system telemetry and resource conditions. Decoys maintain statistical alignment with real encrypted fragments in size and Shannon entropy characteristics. To address scalability, the method is evaluated across small, medium, and large deployments comprising up to 413 externally exposed fragments and compared against fixed-ratio (10%, 20%) and randomized baselines. Experimental evaluation demonstrates increased adversarial uncertainty without altering legitimate reconstruction procedures or encryption mechanisms. Kolmogorov–Smirnov analysis indicates no statistically significant difference between AI-generated decoys and real fragments, whereas baseline decoys produce significant deviations in size and entropy distributions, supporting reconstruction resistance at scale in multi-cloud environments. Full article

The development of China’s marine food production system is receiving increasing attention, as its developmental level and obstacle factors will profoundly impact the nation’s future food security and nutritional supply. This study establishes a theoretical framework for evaluating the development level of marine food production systems based on three dimensions—resources, benefits, and governance—structured around the logical framework of “exogenous safeguard, endogenous drive, goal oriented”. First, a three-tier coding method based on grounded theory was employed to construct a Chinese marine food production system evaluation framework encompassing 28 specific indicators. Subsequently, a comprehensive weighting of these indicators was achieved by integrating fuzzy comprehensive evaluation with the entropy weighting method. Finally, based on the evaluation results and obstacle degree modeling, a comprehensive assessment study was conducted on 11 coastal provinces and cities, focusing on developmental level investigation and obstacle factor analysis. The results indicate that China’s marine food production system development level exhibits a trend of slow, fluctuating growth overall, maintaining an average annual growth rate of 3.23%. However, significant differentiation characteristics are emerging, with high regional heterogeneity and substantial variation in obstacle factors. Currently, the main constraints hindering the development of the marine food production system are insufficient human resource supply, uneven production resource distribution (higher in the north, lower in the south), and intensified fluctuations in comprehensive output. Finally, this study proposes three strategic recommendations: ecological restoration coupled with strict controls, comprehensive restructuring of the human resource support system, and establishing a multi-scale comprehensive evaluation mechanism. These strategies aim to disrupt the transmission mechanisms of different obstacle factors and accelerate the rapid development of the marine food production system. Full article

Remote tower (rTWR) operations are reshaping air traffic control but introduce significant human-factor risks, notably cognitive fatigue induced by prolonged screen-based visual surveillance. To mitigate these risks in a safety-critical domain where missed detections can be catastrophic, we propose a non-intrusive, multimodal fatigue detection framework fusing ocular and cardiac signals. A high-fidelity simulation study with 36 controllers was conducted to collect eye-tracking and electrocardiogram (ECG) data, from which a 12-dimensional feature vector—integrating gaze entropy and heart rate variability (HRV)—was extracted. Addressing the severe class imbalance and scarcity of fatigue samples in physiological data, we developed a cost-sensitive XGBoost classifier combining SMOTE oversampling with a dynamically weighted loss function. Experimental results show that the proposed framework performed well under mixed-subject evaluation and improved sensitivity to fatigue events. Although a marked performance drop was observed under LOSO evaluation, personalized calibration partially alleviated this limitation, indicating the potential of the framework for real-time fatigue monitoring in remote tower operations. Full article

Magnetic and structural transitions can interact significantly, leading to an enhanced magnetocaloric effect (MCE), also known as the giant or colossal effect. In this study, we investigate how subtle microstructural changes impact the magnetocaloric behavior of a MnCoGeB_0.02 alloy fabricated via suction casting. We obtained conical samples and analyzed them to understand their structure and magnetic properties. X-ray diffraction patterns revealed a coexistence of a metastable high-temperature hexagonal phase and a stable low-temperature orthorhombic phase in different regions of each cone. The presence and proportion of these phases determine the degree of magneto-structural coupling, which in turn influences the MCE. The magnetic entropy change (|ΔS_Peak|) varied notably among the samples, ranging from 12.3 to 6 Jkg⁻¹K⁻¹ under a magnetic field change of Δµ₀H = 5.0 T. These findings demonstrate that even minor microstructural changes caused by differences in solidification during suction casting can lead to noticeable variations in magnetocaloric performance. Understanding and controlling these microstructural details is vital for optimizing the functional behavior of MnCoGe-based materials. Full article

This work aims to develop a controlled ball milling strategy for preparing FeCoCrNiCu high-entropy alloy (HEA) powders with improved compositional homogeneity while maintaining limited oxygen uptake. Specifically, a novel two-step combined ball milling strategy integrating gradient ball-size configurations with a sequential milling procedure is proposed and systematically evaluated. Compared with conventional single-step milling, the mixed-ball and two-step configurations enhance mechanical alloying (MA) efficiency and promote the formation of more stable FCC and BCC dual-phase structures, as confirmed by X-ray diffraction (XRD) analysis. Compositional standard deviation derived from energy-dispersive X-ray spectroscopy (EDS) measurements indicates improved macroscopic uniformity, while oxygen/nitrogen/hydrogen (ONH) analysis verifies that oxygen incorporation remains limited within the tested processing window. Systematic comparison of jar filling degrees and sampling interruptions further reveals the coupled influence of collision energy distribution and exposure frequency on oxidation behavior. The results demonstrate that controlled energy distribution and minimized atmospheric disturbance are critical for balancing alloying efficiency and oxygen control in FeCoCrNiCu powders. Full article

Laser powder bed fusion (LPBF) of AA7075 is severely constrained by a narrow process window and susceptibility to defect formation (hot cracking and porosity), which often dominates performance. In this study, 5 wt.% AlCoCrFeNi_2.1 high-entropy alloy (HEA) particles, volumetric energy density (VED = 74–222 J·mm⁻³), and subsequent T6 heat treatment were systematically investigated to reveal their combined effects on defect structure, crystallographic texture/substructure, and tensile behaviour. Quantitative EBSD shows a measurable grain refinement in the as-built state (average grain size 13.44 → 11.80 µm, ~12%) accompanied by a pronounced weakening of the <001> fibre texture (maximum MRD 4.94 → 2.38), indicating disrupted epitaxial growth and a more dispersed orientation distribution. After T6, the reinforced alloy retains a higher low-angle boundary fraction (31.62% vs. 24.17% in unreinforced AA7075) and a higher kernel average misorientation (0.80° vs. 0.60°), consistent with particle-stabilised substructure retention and retarded recovery. Across all VEDs, AA7075-HEA exhibits higher microhardness (compared with AA7075, the addition of HEA increases the hardness by roughly 20–50 HV) and tensile strength, with the intermediate VED (140.74 J·mm⁻³, T6 states) yielding the best performance. While macroscopic cracking is not fully eliminated, the results clarify that HEA-enabled texture/substructure modifications can contribute to enhanced defect tolerance and are more effectively translated into tensile performance when the as-built defect severity is controlled. These findings provide quantitative insights into defect–microstructure–property coupling in LPBF AA7075-HEA composites from as-built to T6 states. Full article

Advanced high-temperature fluid reactors (ARs), such as sodium fast reactors (SFRs) and molten salt cooled reactors (MSCRs) utilize high-temperature fluids at ambient pressure. To melt the fluid during reactor startup and prevent fluid freezing during cooldown, the thermal–hydraulic systems of such ARs include heater zones consisting of specific heaters with controllers, temperature sensors, and thermal insulation. The failure of heater zones due to insulation material degradation or improper installation, resulting in parasitic heat losses, can lead to fluid freezing. The detection of faults using a heat-transfer model is difficult because of a lack of knowledge of the experimental details. Data-driven machine learning of heater zone temperature time series offers a viable alternative. In this study, we benchmarked the performance of recurrent neural networks (RNNs) in an analysis of heat-up transient temperature time series of heater zones installed on a liquid sodium vessel. The RNN models include long short-term memory (LSTM) and gated recurrent unit (GRU) networks, as well as their bi-directional variants, BiLSTM and BiGRU. Anomalous temperature points were designated using a percentile-based threshold applied to residual fluctuations in the detrended temperature time series. Additionally, the impact of the exponentially weighted moving average (EWMA) method on detection accuracy was examined. The RNN models’ performance was assessed using precision, recall, and F₁ score metrics. Results demonstrated that RNN models effectively detect anomalies in temperature time series with the best models for each heater zone achieving F₁ scores of over 93%. To explain the variations in RNN model performance across different heater zones, we used Kullback–Leibler (KL) divergence to quantify the relative entropy between training and testing data, and the Detrended Fluctuation Analysis (DFA) to assess long-range temporal correlations. For datasets with strong long-range correlations and minimal relative entropy between training and testing data, GRU is the best-performing model. When the data exhibits weaker long-term correlations and a significant relative entropy between training and testing distributions, BiGRU shows the best performance. For the data sets with intermediate values of both KL divergence and DFA, the best performance is obtained with LSTM and BiLSTM, respectively. Full article

Accurate registration of multi-temporal LiDAR point clouds is essential for reliable monitoring of mining subsidence. Systematic errors in point clouds acquired at different times can arise from GNSS/INS positioning drift, sensor calibration bias, and differences in observation geometry. These errors typically manifest as global reference shifts or gradual distortions. When such errors are superimposed on real terrain changes, they can mask subsidence signals and introduce observational pseudo-differences, thereby increasing the difficulty of separating actual subsidence from artifacts. To address this issue, this study proposes Robust-Registration-Based Systematic Error Correction for Time-Series Point Clouds (RR-SEC), which establishes a consistent reference framework across epochs. The method does not assume that stable areas remain strictly unchanged. Instead, it identifies regions whose local change patterns are more temporally consistent using an information entropy analysis of multi-temporal differences. Under complex terrain, the method selects points with lower difference entropy as stable control points and uses them to constrain the registration process. It then performs Generalized Iterative Closest Point (GICP) rigid registration under these constraints to estimate the overall three-dimensional translation and rotation between point clouds from different periods. The estimated transformation is applied to the entire point cloud to correct inter-epoch reference mismatches and unify the coordinate reference across all epochs. Comprehensive validation using simulated complex terrain data containing rigid reference biases and non-rigid deformations, as well as UAV LiDAR data collected from the MuduChaideng Coal Mine, shows that, compared with the baseline GICP method, RR-SEC reduces alignment errors. It decreases the mean residual in stable areas by approximately 85%. The subsidence values computed from the corrected point clouds are more consistent with measured values, and the spatial deformation patterns are easier to interpret. RR-SEC demonstrates robust performance and can serve as a practical approach to improve the accuracy of deformation monitoring in mining areas and potentially other geoscientific applications. Full article

We propose Adaptive Inference, a portfolio management framework extending Active Inference to non-stationary financial environments. The framework integrates inference, control, and execution under endogenous uncertainty, modeling investment decisions as coupled dynamics of belief updating, preference encoding, and action selection rather than optimization over fixed objectives. In this approach, portfolio behavior is governed by the expected free energy (EFE) minimization, showing that classical valuation models emerge as limiting cases when epistemic components vanish. Using train–test evaluation on the ARKK Innovation ETF (2015–2025), we identify a Passivity Paradox: frozen belief transfer outperforms naive adaptive learning. A Professional Agent achieves a Sharpe ratio of 0.39 while its adaptive counterpart degrades to $-0.28$, reflecting belief contamination when learning from policy-dependent signals. Crucially, the architecture is not designed to generate alpha but to perform endogenous risk management that mitigates overtrading under regime ambiguity and distributional shift. Adaptive Inference Agents maintain long exposure most of the time while tactically reducing positions during high-entropy periods, implementing uncertainty-aware passive investing. All agents reduce realized volatility relative to ARKK Buy-and-Hold (43.0% annualized). Cross-asset validation on the S&P 500 ETF (SPY) shows that inference-guided risk shaping achieves a positive Entropic Sharpe Ratio (ESR), defined as excess return per unit of informational work, thereby quantifying the economic value of information under thermodynamic constraints on inference. Full article

Cemented carbides are essential in applications requiring exceptional hardness and wear resistance. However, the reliance on cobalt as a binder raises concerns related to cost, supply security, and health. High-entropy alloys (HEAs) are promising cobalt-free binders offering favorable mechanical properties and potential grain-growth control. This work presents a new approach for the development of Co-free WC-based cemented carbide employing an HEA binder designed through CALPHAD-guided simulations. An optimized composition corresponding to Al5Cr5Cu10Fe35Mn10Ni35 (at%) alloy is predicted to be FCC-dominant with minimal σ-phase formation and good compatibility with WC. A preliminary batch of powder of the proposed binder was produced by blending elemental powders, arc remelting, and ultrasonic atomization, yielding predominantly spherical particles with a dendritic microstructure. WC–HEA composites (WC–12 wt% HEA) were then prepared by ball milling, pressing, vacuum sintering, and sinter-HIP for a first evaluation of the microstructure and achievable hardness. The microstructure exhibited residual porosity without significant WC grain coarsening. XRD analyses showed the dominant presence of WC, along with FCC and M₃W₃C phases (M mainly Fe and Mn), indicating thermal interaction between the binder and WC. Despite these effects, the composite achieved a hardness of 1913 HV and retained a fine WC grain size (0.86 μm). The proposed design approach allowed the definition of a promising Co-free binder composition based on HEA with the expected microstructure, which will need further evaluation, especially aimed at investigating toughness properties as a function of the WC content. Full article

With the rapid development of the offshore wind power industry, the construction process of offshore wind power jackets faces numerous quality risks, particularly in welding, coating, and assembly operations. This paper aims to investigate the identification, assessment, and management of quality risks during the construction of offshore wind turbine foundation structures. By establishing a multidimensional quality risk assessment framework, key risk factors affecting quality were identified through expert interviews and brainstorming sessions. Comprehensive evaluations of these risk factors were conducted using the Entropy Weight Method (EWM), the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), and Grey Relational Analysis (GRA). The findings indicate that welding and coating processes pose the highest risks during construction. Based on these assessments, corresponding risk mitigation measures are proposed, including process optimization, automation enhancement, environmental control, and management system refinement. This study provides theoretical foundations and practical guidance for improving construction quality and reducing costs in offshore wind turbine foundation manufacturing. It advances quality risk management by introducing an integrated evaluation model that addresses the limitations of single-method approaches in complex construction scenarios. Full article

Query–service matching in customer service systems faces a critical challenge of accurately aligning user queries expressed in colloquial language with formally defined services while balancing business objectives. Traditional keyword-based and embedding approaches fail to capture complex semantic nuances and cannot provide interpretable explanations. We address this problem by proposing a novel reasoning-enhanced framework that leverages large language models (LLMs) for structured multi-criteria evaluation. Our key innovation is a reasoning-first scoring architecture where the model generates detailed explanations before numerical scores, reducing score variance by 18% through conditional mutual information. We introduce a controlled stochastic perturbation mechanism with theoretically derived optimal parameters that balance diversity and relevance, alongside a knowledge distillation pipeline enabling 960× model compression (480B→0.5B parameters) while retaining 94% performance. Rigorous theoretical analysis establishes Pareto optimality guarantees for multi-criteria evaluation, information-theoretic entropy reduction bounds, and PAC learning guarantees for distillation. Experimental validation on real-world telecommunications data demonstrates 89% Precision@1 (15.3% improvement over baselines), 23% diversity enhancement, and 96× latency reduction, with deployment cost decreasing 1200× compared to direct LLM inference. This work bridges the gap between LLM capabilities and production deployment requirements through principled mathematical foundations and practical system design. Full article

Multi-label emotion detection (MLED) suffers from long-tailed label distributions and structured inter-label correlations, which jointly suppress rare label recall and yield incoherent predictions. We present Graph Neural Network-Enhanced Transformer with Hybrid Loss Weighting (GTHL-Emo), a unified framework that addresses both challenges without heavy additional machinery. First, an adaptive imbalance-aware training scheme combines binary cross-entropy, asymmetric focal, and pairwise ranking losses under a learned batch-wise controller, emphasizing rare labels while stabilizing thresholding. Second, a lightweight correlation alignment module learns transformer-based label embeddings and aligns their predicted affinities with empirical co-occurrence via Kullback–Leibler (KL) regularization, smoothing rare label predictions through correlated frequent labels. A transformer encoder with learnable attention pooling provides semantic representations, and a dynamic GraphSAGE layer captures inter-instance structural dependencies. Comprehensive evaluation across three Arabic benchmarks—SemEval-2018-Ec-Ar, ExaAEC, and SemEval-2025 (Track A, Arq)—demonstrates competitive or leading performance. On SemEval-2018-Ec-Ar, GTHL-Emo attained a Jaccard accuracy of 58.70%, micro-F1 score of 71.02%, and macro-F1 score of 60.48%. On ExaAEC, it achieved a Jaccard accuracy of 65.99%, micro-F1 score of 70.72%, and macro-F1 score of 68.71%. On SemEval-2025-Arq, it obtained a Jaccard accuracy of 41.47%, micro-F1 score of 56.78%, and macro-F1 score of 56.69%. Ablation studies revealed that the GraphSAGE structure and ranking loss contributed most significantly (1.45% and 1.46% Jaccard accuracy drops, respectively), while label correlation alignment provided consistent improvements across the scales. These findings demonstrate that jointly optimizing imbalance-aware objectives and label dependencies yields robust Arabic MLED with minimal overhead. Full article

Urban integration can lower average pollution, yet environmental benefits may be unevenly shared across cities within the same urban agglomeration. Such within-agglomeration disparities can weaken joint prevention and control, sustain unequal health risks, and hinder inclusive urban sustainability even when overall concentrations fall. Using a panel of Chinese metropolitan areas from 2005 to 2023, we examine whether metropolitan integration is associated with a more even distribution of pollution burdens among constituent cities. We measure within-agglomeration inequality using entropy-based indices for total emissions and emissions intensity, and capture integration intensity using cumulative policy attention and the years since integration began. We find that deeper integration is associated with lower pollution inequality, with larger reductions for inequality in total emissions than for inequality in emissions intensity. The decline emerges after integration begins and persists over time, and it remains robust to alternative measures and to an identification strategy that leverages predetermined historical connectivity. The equalizing association is most evident in metropolitan areas featuring high-primacy and high-ranking core cities, is reinforced by greater fiscal capacity and factor market integration, and is moderated by industrial lock-in. These results suggest that metropolitan integration, when supported by credible cross-city coordination and transition support in regions facing industrial lock-in, can promote cleaner and more equitable environmental outcomes within urban agglomerations. Full article