Search Results (267)

Metallurgical lifting equipment operates under prolonged heavy-load, high-impact, and complex working conditions. The resulting failure reports contain rich field knowledge applicable to fault diagnosis and predictive maintenance. Nevertheless, reliably extracting traceable, structured knowledge from procedural and implicit maintenance records remains a significant challenge. To address this, the paper proposes a generative information extraction method for large language models (LLMs) that integrates ontology schema with inference chain constraints, targeting knowledge extraction and knowledge graph construction from failure reports of metallurgical lifting equipment, named generative constrained information extraction for operations and maintenance (GCIE-OM). A domain ontology schema is first constructed, defining seven entity types and nine relation types to establish explicit knowledge boundaries for structured LLM generation. An inference chain-assisted structured parsing method, termed IC-ASP, is then designed to guide the model through a sequential extraction pipeline comprising scene identification, scope of entity boundary, inference of relation type, evidence traceability with localization, and triple output. This stepwise process strengthens the model’s capacity to comprehend equipment hierarchies, fault evolution chains, and maintenance action logic. Building on this, ChatGLM or LLaMA serves as the backbone model and is adapted to the target domain via LoRA fine-tuning. Entity alignment and character-level source localization mechanisms are further introduced to establish precise mappings between generated outputs and their textual evidence in the source documents. The extracted results are ultimately converted into standardized knowledge triples and stored in a Neo4j graph database. Based on this, a prototype system for generative information extraction is designed and implemented to demonstrate the practical effectiveness and adaptability of the proposed method. Experimental results show that the proposed method outperforms baseline methods across entity recognition, relation extraction, and structured output quality, providing robust knowledge support for fault tracing and predictive maintenance of metallurgical lifting equipment. Full article

Triple-negative breast cancer (TNBC) lacks effective molecular targets, leading to poor prognosis. Previous computational methods to identify targets have suffered from low druggability, high complexity, and lack of robust validation. We propose a hybrid methodology combining Boolean network modeling with semidefinite programming (SDP) to analyze a TNBC cell line network. The resulting therapeutic pair underwent a multi-level validation framework, including Boolean simulations, statistical uncertainty quantification (bootstrap), sensitivity analysis, and orthogonal computational support from AlphaGenome, a deep learning model from Google DeepMind. Our analysis identified TK1 and VIM as a computationally robust therapeutic pair. Dual inhibition achieved 99.03% similarity to the apoptotic state with a 95% confidence interval of [98.79%, 99.26%], and was statistically superior to alternative pairs ($p<0.001$). The selection remained optimal across all tested model parameters, demonstrating high robustness. Importantly, the pair has full druggability because both targets have available specific inhibitors. Orthogonal computational evidence from AlphaGenome, stratified by mammary compartment, indicated that both targets exhibit moderate baseline expression in normal mammary epithelium (TK1 = 0.159, VIM = 0.143 in normalized RNA-seq units; n = 13 tracks per gene), with VIM showing a 2.2-fold higher expression in mammary stroma than in epithelium—a gradient consistent with its established role as a mesenchymal marker. Promoter-variant proxy analysis indicated near-zero transcriptomic perturbation upon simulated inhibition of either target in normal mammary epithelium (mean $|{log}_2\mathrm{FC}|<0.001$), supporting a favorable therapeutic window. Our methodology identified TK1–VIM as a computationally robust, druggable therapeutic candidate pair with biologically plausible mechanism of action. Gene-variability analysis identified TK1 and VIM as the highest-scoring candidates, with SDP optimization providing complementary, independent confirmation of this selection. This work provides a computationally grounded candidate strategy and a rigorous methodological benchmark for computational drug target identification; experimental validation remains an essential next step before clinical translation. Full article

Decoupled maximum power point tracking control and output voltage control can be accomplished simultaneously using dual-duty cycle control. However, developed triple switch triple mode (TSTM) exhibits absence of the common ground between the solar panel and output load therefore causing the leakage current to flow which creates safety concern especially for household electrification. In addition to having a negative effect on the solar panel, leakage current increases power losses. Thus, this work proposes a unique TSTM dc-dc converter. The suggested converter has the following advantages: (1) The presence of a common ground between the output load and the solar panel eliminates the leakage current. (2) Reduced electromagnetic interference issues present due to leakage current. (3) Enhanced voltage gain over wider duty cycle. (4) Enables simultaneous decoupled control of MPPT and output voltage. (5) Absence of voltage oscillation across the switches. The proposed TSTM converter is an unique combination of switched inductor and switched capacitor. Both inductor and capacitors are connected in order to boost the level of voltage at the output terminal. The operating principle, design equations and device stress are analyzed in detail for the proposed TSTM. The comparison over existing converter in terms of voltage gain and switch stresses are highlighted in details. Lastly, a laboratory prototype (40/400 V) for 400 W is created and thoroughly tested in order to validate mathematical calculations. Full article

Postharvest seed cleaning is critical for ensuring high-quality rice seeds suitable for storage and planting. Traditional cleaning systems, which are often limited to one or two sieves, are insufficient for removing all impurities, resulting in reduced seed purity and potential germination issues. This study was designed to enhance the rice seed cleaning system by integrating a high-efficiency blower with a triple-sieving mechanism. The system utilized three sieves with progressively smaller mesh sizes to systematically separate contaminants such as dust, broken grains, husks, and other foreign particles. A controlled airflow from the blower distributes rice seeds uniformly across the sieves, optimizing separation while minimizing mechanical damage. Compared to existing conventional systems, the proposed design demonstrated significantly improved cleaning performance, resulting in higher seed purity levels and overall enhanced seed quality. The triple-sieve configuration, coupled with precise airflow control, led to more effective impurity removal and uniform seed handling. The improved seed-cleaning system offers several agronomic benefits, including reduced postharvest losses, increased seed germination rates, and improved crop establishment. By producing cleaner, higher-quality seeds, this system has the potential to support more efficient and productive rice cultivation. Full article

Target detection in synthetic aperture radar (SAR) images is constrained by three primary challenges. First, speckle noise overlaps heavily with the high-frequency features of target edges in the frequency domain, so standard convolutions cannot suppress noise without sacrificing edge texture. Second, the scattering signature of a SAR target varies markedly with viewing angle, and a fixed-parameter convolution kernel cannot accommodate this spatial non-stationarity. Third, deep and shallow levels of the feature pyramid differ in semantics and resolution, and a naive element-wise sum either introduces noise interference or loses small-target signals. We propose the Frequency–Spatial Detection Network (FSDNet), whose core FSDBlock cascades three operators to address these failure modes in turn. Wavelet Convolution (WTConv) projects features into Haar sub-bands and applies independent low- and high-frequency kernels prior to inverse-DWT reconstruction, suppressing noise while preserving edges. Receptive-Field Attention Convolution (RFAConv) generates location-conditional kernels and so adapts to non-stationary scattering. Spatial Context Self-Attention (SCSA) aggregates discrete scattering points into coherent target representations via long-range grouped attention. At the fusion stage, CGAFusion replaces FPN element-wise addition with a channel–spatial–pixel triple-attention soft switch that mitigates deep–shallow semantic mismatch. On HRSID, FSDNet attains mAP₅₀ $= 92.3$% and mAP_50:95 $= 68.6$%. On SSDD, it attains mAP₅₀ $= 98.7$% and mAP_50:95 $= 74.2$%. Both sets of results consistently surpass the baseline methods. Against the strongest YOLO baseline (YOLOv11n), FSDNet improves HRSID mAP₅₀ by $+1.7$ percentage points (pp) and mAP_50:95 by $+2.3$ pp, and SSDD mAP₅₀ by $+0.5$ pp and mAP_50:95 by $+2.7$ pp; against the capacity-fair YOLOv11s reference (∼51% more parameters), FSDNet still leads on mAP₅₀, mAP_50:95, recall, and F1. Ablation studies and power-spectral-density analyses corroborate the contribution of each module and confirm WTConv’s role in preserving high-frequency target features. Full article

Single image dehazing remains a fundamental challenge in computer vision due to the ill-posed nature of the inverse problem and the spatial heterogeneity of real atmospheric haze. Existing convolutional approaches suffer from two structural deficiencies: bounded receptive fields that fail to model large-scale haze gradients, and isotropic kernels insensitive to the directional patterns of atmospheric scattering. This paper proposes ACTA-AOD, a lightweight end-to-end dehazing network that addresses both limitations within a unified framework built upon the AOD-Net K-parameterization. The network integrates two complementary modules: (1) W-EMSAv2, a windowed efficient multi-scale attention module that reduces attention complexity from $\mathcal{O}(N^{2}C)$ to $\mathcal{O}(N{M}^{2}C/4)$ while preserving full-spectrum spatial information through pixel-shuffle reconstruction; and (2) the ACTA Fusion module, which combines structural-reparameterization-based asymmetric convolution with cross-dimensional Triple Attention for direction-sensitive local detail recovery at zero inference-time overhead. On the RESIDE benchmark, ACTA-AOD achieves peak signal-to-noise ratio (PSNR) of 26.02 dB and structural similarity index measure (SSIM) of 0.910 on indoor synthetic data, and 26.13 dB/0.910 on outdoor synthetic data, surpassing the AOD-Net baseline by $+3.41$ dB (indoor) and $+3.58$ dB (outdoor) in PSNR, and exceeding the strongest learning-based baseline (AECRNet, CVPR 2021) by $+1.17$ dB (indoor) and $+1.75$ dB (outdoor). The model processes images at 81 frames per second on a single GPU. Ablation studies and stratified robustness evaluation across five haze density levels confirm the complementary, synergistic contribution of each module. Full article

Silicon micromachining serves as the foundational enabling technology for high-precision MEMS inertial sensors. However, the relentless pursuit of enhanced sensitivity and multi-functionality in emerging applications encounters a fundamental bottleneck when confined to two-dimensional scaling. The evolution toward complex three-dimensional (3D) stacking architectures is an inevitable trajectory for devices including MEMS inertial sensors, yet performance is constrained by the limitations of conventional processes in fabricating and integrating intricate 3D hollow structures. Specifically, uniformity in large-area deep silicon etching, structural integrity of convex corners in wet etching, and residual stress induced by multi-layer wafer bonding have emerged as critical, shared challenges. To address these issues, this paper proposes a triple-layer wafer-level stacking scheme that synergistically combines wet/dry hybrid etching with low-temperature adhesive bonding. This stacking scheme incorporates an innovative linear compensation model for wet-etched convex corners, enabling high-precision fabrication of complex corner structures under deep etching conditions. Furthermore, a collaborative strategy involving temporary bonding and plasma flow-field optimization improves the uniformity and integrity of dry etching for large perforated structures. A low-temperature triple-layer wafer-level stacking process is developed, encompassing precise adhesive dispensing, optical alignment, and a stepped low-temperature curing profile, thereby achieving highly symmetric 3D integration with controlled adhesive distribution. The efficacy of this stacking scheme is validated through the fabrication of a symmetrically stacked triple-layer MOEMS accelerometer sensing element. Test results demonstrate a noise floor as low as 0.40 µg/√Hz and a bias instability of 1.81 µg over 10 min. Compared with a double-layer counterpart, improved performance is obtained. The wafer-level stacking scheme established in this work not only provides a viable pathway for pushing the manufacturing limits of high-precision inertial devices but also offers a generic methodology for tackling complex hollow structure formation and low-temperature integration, holding referential value for broader applications in high-precision 3D microsystems. Full article

Large language models (LLMs) frequently produce text that is fluent yet factually inconsistent with source documents. Detecting such inconsistency remains challenging, particularly when errors involve subtle entity substitutions, temporal distortions, or relational misattributions embedded within lengthy outputs. We propose Atomic Contrastive Verification (ACV), a training-free, graph-grounded fact-checking framework that decomposes both generated claims and source documents into atomic claims—minimal, self-contained factual units—and performs structured contrastive reasoning over each unit independently. For each atomic claim, ACV extracts a knowledge graph triple and generates contrastive claim variants through a multi-type perturbation taxonomy covering entity, relation, temporal, and quantitative dimensions. A novel Knowledge-Weighted Contrastive MMR mechanism, integrating graph-structural centrality and NLI-based logical diversity, selects the most discriminative subset of variants. Each selected variant is then pairwise compared against the claim; the resulting comparison responses are summarized to produce a per-claim verdict, and per-claim verdicts are aggregated into a document-level judgment. Experiments on the LLM-AggreFact benchmark (eleven subsets) demonstrate that ACV achieves competitive or superior performance compared to both specialized fine-tuned fact-checkers and large-scale LLMs. Beyond accuracy, ACV provides interpretable, claim-level error localization that existing methods cannot offer. Full article

As a complex structural form capable of simultaneously bearing upper-deck highway traffic, lower-deck highway traffic, and rail transit, the curved chord stiffened double-deck continuous steel truss bridge is distinct from traditional single-deck bridges. The spatial superposition of multiple traffic types within this structure may result in multiple components approaching their critical states concurrently. Despite prior research efforts on this structural type, the failure evolution process from local yielding to global collapse under mixed traffic loads remains ambiguous. This study addresses these questions through systematic numerical investigation of a nine-span bridge with a 300 m main span. A two-stage analytical approach is employed: a Midas/Civil analysis first identifies critically stressed regions, then ABAQUS multi-scale modeling enables refined analysis of critical components while maintaining computational efficiency. Twenty-nine combined traffic loading cases encompassing dual- and triple-category configurations are systematically analyzed. The results show that the ultimate load-carrying capacity coefficients range from approximately 7 to 18, with a minimum of 7.137, and the dual-level highway combinations exert greater influence than road–rail combinations. More importantly, three failure path convergence characteristics were discovered. First, the initial failure position under each working condition tends to be consistent, initiating at the lower chord near the top of the mid-span pier, which confirms that inherent structural defects exist at this location. Second, the gusset plate at the top of pier W6 appears as the second failure location in 48% of cases and ranks within the first four locations across all cases. Third, path similarity progressively increases with traffic diversity. Additionally, Q370qE steel exhibits 5–22% stress exceedance with variable critical locations depending on traffic conditions. Based on these convergence characteristics, a safety monitoring scheme is proposed: monitoring points need to be arranged symmetrically on both sides of the bridge on the top chords, bottom chords, web members, and wedge plates near the tops of the piers. Full article

The BeiDou-3 (BDS-3) Precise Point Positioning service (PPP-B2b) can realize decimeter-level positioning by broadcasting satellite orbit, clock offset, and code bias corrections via GEO satellites, enabling PPP without reliance on ground communication networks. However, the current PPP-B2b service only provides corrections for BDS-3 and GPS satellites, which limits the number of available satellites and may affect positioning performance in challenging environments. To further enhance the positioning performance, we propose to incorporate Galileo observation into the PPP-B2b positioning. A PPP model integrating PPP-B2b service and broadcast ephemeris was established. First, the accuracy of the Galileo broadcast ephemeris was evaluated using precise orbit and clock products as references. The results show that the mean signal-in-space range error (SISRE) standard deviation of Galileo broadcast ephemeris is 0.30, which is only a little worse than that of GPS from PPP-B2b service. Then, the positioning experiments were conducted under different elevation cutoff angles. The experiments were conducted using data from 94 reference stations in China over a 7-day period. The results demonstrate that the inclusion of Galileo satellites significantly increases the number of visible satellites and improves satellite geometry. Compared with the BDS-3/GPS dual-system PPP solution, the BDS-3/GPS/Galileo triple-system PPP solution reduces the horizontal convergence time by approximately 13.70–16.67% and the vertical convergence time by about 18.75–20.00% under cutoff angles from 7° to 30° based on the 68th percentile statistics. The 95th percentile results further confirm the advantage of the triple-system solution under a more stringent statistical criterion. Where convergence is achieved, the triple-system solution reduces the horizontal convergence time by approximately 6.0–7.3% and the vertical convergence time by about 15.3–26.0%. Moreover, the triple-system solution exhibits a smaller re-convergence jump under abnormal observation conditions. In addition, under high elevation cutoff conditions, the introduction of Galileo satellites effectively improves PPP availability, thereby enhancing the continuity and robustness of PPP. These results indicate that incorporating Galileo observation within the PPP-B2b framework can effectively improve PPP performance and provide a simple and practical approach for high-precision real-time positioning. Full article

Scientific and accurate agronomic knowledge is key to ensuring efficient wheat production. China’s vast agricultural land spans a wide range of longitudes and latitudes, and agronomic practices are closely tied to temporal factors such as wheat growth stages. So agronomic knowledge exhibits significant spatiotemporal variability. Constructing a spatiotemporal knowledge graph of wheat production can offer multi-dimensional data support and enabling deeper knowledge services. Wheat agronomic knowledge is often fragmented and unstructured and efficiently extracting text segments of agronomic knowledge and agronomic knowledge triples are two key challenges. Because of the high proportion and significant production service value of attribute values in agronomic knowledge, an attribute-rich agronomic knowledge graph schema was created. According to the characteristics of agronomic texts, a keyword attention mechanism (KAM) was proposed and integrated with an improved BERT model for sentence-level feature extraction to create an extraction model AgronomicCorpusExtraction for agronomic knowledge text corpora. The agronomic knowledge of wheat production is characterized by non-standard syntax, complex multi-layer structures, diverse entity expression methods, and a wide span of scope, and existing extraction methods cannot achieve satisfactory results. To address the issue, a joint extraction model AgronomicTripleExtraction was proposed to extract entities, attributes, and relations in different phrases, firstly the BERT and BiGRU were used jointly to extract the long and short distance features, and the CRF was used by global normalization joint modeling to extract attributes, then intermediate features between the same type of attributes extracted by average pooling to segment different entities. At last, a relation-aware relation feature enhancement (RAFE) method was created and a MLP was used to extract relations based on the relation matrix constructed from the knowledge graph schema. Ablation experiments were conducted to evaluate the performance for AgronomicCorpusExtraction with and without KAM and that for AgronomicTripleExtraction under four conditions, the model with BiGRU, RAFE, and entity segment, without BiGRU, without RAFF, and without entity segment. The results indicate that the use of KAM improves F1-score by 0.128 and AgronomicTripleExtraction achieves F1 of 0.897, 0.875, 0.871 for attribute, entity and relation extraction when using the three modules simultaneously, and removing any single module leads to a certain degree of performance degradation. Comparative experiments were conducted between AgronomicTripleExtraction and some related state-of-the-art models published recently. Full article

In knowledge graph construction tasks, recent research often leverages Large Language Models (LLMs) to enhance the efficiency and accuracy of unstructured data processing. However, current LLMs rely on lexical co-occurrence statistical patterns, making it difficult to capture deep semantic relationships. Furthermore, existing research largely focuses on entity-relation extraction or semantic-level optimization, overlooking the inherent hierarchical logical structures within text paragraphs (e.g., chapter organization, paragraph coherence). This leads to insufficient semantic completeness and damaged structural consistency in the constructed knowledge graphs. To address this dual limitation, we propose LLM-S2KG, a semantic–structural information extraction method that integrates LLMs with semantic correlation analysis. This method achieves synergistic modeling of semantic depth and logical structure by simultaneously performing dual parsing of keywords and structure, discovering and completing semantic associations, and finally integrating these dual graphs for construction. Experiments show that in query tasks, LLM-S2KG improved the F1 score by 0.1183, 0.1412, and 0.0231 compared with KeyBERT, TF-IDF, and LLM-KG, respectively. In fill-in-the-blank QA tasks, it achieved an accuracy of 94.81%; and in open-ended QA tasks, an accuracy of 85.885%, moderately outperforming LLM Triple Extraction (73.308%), LLM Triple Extraction with Source Sentence Augmentation (80.085%), and Chroma Database Import (76.150%). In summary, LLM-S2KG provides a unified modeling paradigm for structured knowledge extraction using LLMs, featuring mutual empowerment and co-evolution of semantics and structure. Full article

Distant supervision relation extraction (DS-RE) provides an efficient way to construct large-scale training data by automatically aligning knowledge base relations with unstructured texts. However, this process inevitably introduces erroneous labels because sentences containing the same entity pair do not always express the corresponding knowledge base relations. To address this problem, this paper proposes a noise-weighted unsupervised denoising framework that integrates sentence-level prior confidence estimation, multi-factor representation learning, fine-grained noise detection, and clustering-based label generation. The framework first estimates noise-aware prior weights by matching sentence instances with semantically similar relation triples. It then incorporates lexical, positional, and entity-type factors to enhance sentence representations. For detected noisy instances, an unsupervised clustering-based label generation module is used to regenerate relation labels rather than directly discarding them. Experimental results on the DSRED dataset show that the proposed method achieves 89.7% Precision, 90.6% Recall, 90.1% F1-score, and a PR-AUC of $0.942\pm 0.004$, outperforming the strongest baseline EFEAPN by 1.7 percentage points in F1-score. Statistical analysis further shows that the PR-AUC improvement remains significant after Bonferroni correction ($p_{\mathrm{adj}}=0.0094$). Module-level ablation experiments, sensitivity analysis, and clustering quality evaluation further verify the effectiveness of the noise weighting and clustering-based label generation modules. Supplementary experiments with Transformer-based encoders and cross-dataset evaluation further show that the main performance gain comes from the proposed denoising framework rather than from a specific sentence encoder. Full article

Assessing the quality of colorized images remains challenging as most colorization artifacts arise from high-level semantic errors in the form of implausible or unnatural color assignments, rather than conventional low-level distortions such as noise, blur, or compression artifacts. The evaluation process is further complicated by the inherently subjective nature of color perception and the difficulty of accurately modeling human responses to color. Motivated by the limitations of current image quality assessment metrics for this task, we propose TRIPSI (Triple-Source Realigned Integrated Perceptual Semantic Index), a hybrid full-reference framework which approaches colorization quality assessment as a pattern recognition problem in a learned semantic-aware feature space. TRIPSI fuses three complementary deep pre-trained models, TOPIQ, LIQE, and DreamSim, into a unified framework by applying rank normalization to individual model scores per dataset to ensure comparability across varying output scales before aggregating the scores with equal weights. LIQE captures explicit color distortions. TOPIQ focuses on semantically important regions and color saturation artifacts. DreamSim measures color-preserving pattern agreement between deep feature representations of a colorized image and its ground-truth reference color image in a learned semantic-aware embedding space. By explicitly incorporating color-aware semantic representations at multiple levels, results across multiple datasets show that TRIPSI closely reflects human perceptual judgments, highlighting the effectiveness of semantic pattern modeling for quality assessment in image colorization. Full article

The field of intelligent transportation on inland waterways is experiencing rapid growth, driven by the global pursuit of enhanced waterway safety, operational efficiency, and environmental sustainability. In real-world autonomous operation scenarios of unmanned surface vehicles (USVs), image-based 2D object detection methods are insufficient to meet the demands of 3D environmental modeling and accurate perception of dynamic objects. Existing 3D perception systems for USVs depend heavily on precise sensor calibration. However, projection offsets between point clouds and images—caused by water surface fluctuations and complex outdoor environments—hinder the practical deployment of these methods. To address these limitations, we propose a weak calibration multi-modal 3D object detection algorithm based on cross-view fusion, termed RCF-Free (Radar-Camera Fusion, Free from precise calibration). Inspired by autonomous driving solutions, we design a Triple-Path Cross-View Fusion module that achieves high-quality cross-view feature fusion without requiring accurate calibration parameters, while simultaneously detecting complete bird’s-eye view (BEV) bounding boxes. We further enhance the spatial layout comprehension of the visual branch through a Mobile Self-Attention Module (MAM) and effectively encode sparse point cloud features in BEV space using a dedicated BEV-Point feature encoder. Additionally, we reconstruct and introduce two water-related 3D object detection datasets, FloW-BEV and WaterScenes-BEV. Experimental results demonstrate that RCF-Free achieves $\mathrm{m}\mathrm{A}{\mathrm{P}}_{\mathrm{B}\mathrm{E}\mathrm{V}}50$ scores of 60.5% and 69.3% on the FloW-BEV and WaterScenes-BEV datasets, respectively, showing the effectiveness in water surface object detection. Moreover, on the DAIR-V2X-I dataset for autonomous driving scenarios, the model attains $\mathrm{m}\mathrm{A}{\mathrm{P}}_{3\mathrm{D}}50$ scores of 73.3%, 61.2%, and 61.2% across three task difficulty levels, illustrating strong cross-domain generalization capability. Full article