Search Results (163)

Rapid urbanization and the increasing proportion of time spent indoors have intensified concerns regarding indoor air quality and public health. Fast and reliable prediction of indoor pollutant dispersion is essential for building design optimization and risk-informed emergency response. To enable efficient indoor pollutant dispersion prediction, this study proposes a graph neural network framework with dual-layer aggregation and graph augmentation to address several limitations of traditional GNNs, including restricted receptive fields, inefficient long-range information propagation, and limited ability to capture long-range dependencies. Graph data augmentation is employed to enrich the diversity of concentration field distributions, while a dual-layer aggregation mechanism is introduced to expand the receptive field and enhance message-passing efficiency. Specifically, node features are first aggregated from adjacent edges to capture local information and are then further aggregated across nodes to incorporate global contextual features, enabling the modeling of long-range physical dependencies. Validation across multiple building configurations demonstrates that the proposed model effectively captures physical characteristics of pollutant dispersion in distant rooms and accurately predicts toxic gas dispersion under different layouts, achieving a coefficient of determination $R^2$ of up to 0.97, while delivering computational speeds approximately one order of magnitude faster than conventional CFD solvers. Full article

Artificial intelligence has increasingly been applied to residential floor plan generation, producing outputs in diverse forms such as semantic raster images, graph-structured layouts, and vector polygons. However, these outputs still require substantial interpretation and reconstruction before they can enter conventional CAD and BIM workflows. This study develops a unified engineering–semantic framework for transforming heterogeneous AI-generated floor plans into editable CAD drawings and preliminary BIM information models. Three route-specific procedures—ColorPlan, TopoPlan, and PolyPlan—are established to parse and regularize the three representative input forms. The processed information is reorganized into a common engineering–semantic JSON core and subsequently mapped to native AutoCAD entities and Revit objects. The workflow was evaluated using 165 transformation-ready single-story residential samples. The effective CAD/BIM transformation rates reached 100.0%/97.7% for ColorPlan, 93.3%/85.3% for TopoPlan, and 100.0%/97.9% for PolyPlan. The results show that transformation stability is closely related to the completeness of semantic, geometric, topological, and opening information in the source representation. A controlled human–machine experiment further showed that the proposed workflow reduced CAD drafting time by approximately 85.2% and BIM modeling time by 90.0% compared with fully manual production. The study demonstrates a feasible approach for extending heterogeneous AI-generated layouts into editable and information-bearing engineering representations while substantially reducing repetitive drafting and preliminary modeling work. Full article

This study establishes a GCN–NSGA-III-based framework for determining gas-leak sensor locations. Candidate layouts are optimized simultaneously with respect to installation expenditure, spatial coverage, leak-identification performance, and time to alarm. In the proposed framework, the GCN extracts spatial correlations and leakage-risk features among candidate monitoring locations, whereas NSGA-III optimizes the network weights and thresholds to support the selection of improved sensor placement schemes. By combining the image-based spatial feature extraction capability of CNNs, the graph-structured feature learning capability of GCNs, and the multi-objective optimization strength of NSGA-III, the model achieves significant improvements in detection accuracy and risk assessment efficiency. The model was validated using a hybrid gas-leak dataset comprising 758 training samples, including 189 real-world monitoring samples and 569 simulated samples, and 229 test samples, including 73 real-world monitoring samples and 156 simulated samples. Its engineering applicability was further evaluated using a simulated chlorine leakage scenario at a chemical plant in Changzhou, China, with a leakage rate of 2 kg/s and an ambient easterly wind speed of 1 m/s. Experimental results confirm its reliability and practical applicability in real-world engineering contexts. Compared with the original pre-optimization sensor layout under the same leakage and environmental conditions, the proposed optimization strategy reduces deployment costs by 19%, increases monitoring coverage by 8.2%, improves detection accuracy by 14.1%, and shortens alarm response time by approximately 15%. Full article

Pipe leakage in water distribution networks causes water loss, pressure decline, energy waste, and reduced service reliability. This study aims to support early localization of two simultaneous pipe leaks using the evaluated 11-sensor pressure-monitoring layout. A topology-aware deep learning framework was proposed on the EPANET 2.0 (Build 2.00.12) Net2 network using WNTR 1.3.2-based hydraulic simulation. Two-point leakage scenarios were generated by pipe-splitting strategy, and a 4 h early-stage pressure-residual window from 11 pressure sensors was used as input. The task was formulated as joint multi-label pipe identification and intra-pipe position regression. A bidirectional long short-term memory (BiLSTM) branch learned the temporal pressure response, while a graph attention network version 2 (GATv2) branch represented sensor–topology relationships. On the validation set, the model achieved a Top-2 F1 score (F1@2) of 61.18%. Both leaking pipes were identified in 31.65% of samples, and at least one true leaking pipe was included in the Top-2 candidates for 90.72% of samples. For correctly matched leak-point instances, the physical mean absolute error was 41.91 m. The results indicate that topology-aware temporal learning can provide pipe-level candidates and intra-pipe inspection distances for early two-point leak localization, although nearby and weak simultaneous leaks remain challenging. Full article

Diagnosing faults across different smart homes is hard: sensor names, layouts, and daily routines differ from home to home, so a model trained in one home rarely works in another. We study an ontology-guided framework for cross-home fault diagnosis, but our main contribution is a benchmark-validity audit—a systematic check of whether the datasets used to evaluate such systems actually measure fault detection. Using the public Center for Advanced Studies in Adaptive Systems (CASAS) smart-home datasets (homes hh101–hh110) and real household power data (HomeC, UMass Smart*), we show that much of the high cross-home accuracy reported on these benchmarks is an artifact of features that re-encode the labelling rules rather than evidence of transfer: when those features are removed, the macro-averaged F1 score (macro-F1) collapses toward the level obtained with randomly permuted labels. We therefore treat these datasets as semantic-transfer and benchmark-validity studies, not fault-detection results. The framework’s distinguishing component is a counterfactual calibration layer that returns a probability for its recommended intervention; on a controlled structural causal model with known interventions, it achieves a Brier skill score of $0.369$ for intervention-success probabilities. Separately, on the simulation-derived LBNL Fan Coil Unit benchmark, a conventional gradient-boosted multiclass fault classifier achieves accuracy comparable to a random forest but about six times lower expected calibration error ($0.026$ vs. $0.159$) under a scenario-matched split. This calibration advantage does not generalize to held-out simulation scenarios, where the calibration error rises to $0.372$; we report this negative result as a limitation. We are explicit about scope: the ontology reasoner and the real-stream causal graph are only partially implemented, and the counterfactual recommendations are validated only under controlled or simulated conditions, not in deployed homes. The results are intended for researchers who build or benchmark sensor-based fault-diagnosis models, for dataset curators, and for practitioners who need calibrated rather than merely accurate outputs. All code, the proxy-label rules, and the leakage audit are released. Full article

Cyclic cutwidth problem is a graph layout problem whose goal is to find an embedding $\varphi$ of the vertices of a graph G with n vertices onto a cycle $C_n$ so as to minimize the maximum cutwidth of graph G. According to the literature, this problem is NP-complete, and some exact results regarding cyclic cutwidth have been reported. For an integer $k>1$, a graph G with cyclic cutwidth k is k-cyclic cutwidth critical if every proper subgraph of G has cyclic cutwidth less than k and G is homeomorphically minimal. In this paper, from the point of view of graph decomposition, we first characterize decomposable structure of some k-cyclic cutwidth critical graphs. We ascertain that a k-cyclic cutwidth critical graph G with a centre $u_0$ has a subgraph decomposition with either two or three members, and we find that each member is a $\delta$-linear cutwidth critical subgraph of G with $\delta =k-2$ or $k-1$. The result reveals the structural relation between the cyclic cutwidth and the linear cutwidth of graph G. Full article

Robust object detection for autonomous driving requires perception models that remain reliable when visible imagery is degraded by darkness, glare, rain, fog, motion blur, or long-range small targets. Visible and thermal infrared cameras provide complementary evidence, yet many RGB–thermal detectors fuse modalities, mainly as aligned tensors, and may underuse relational structure in channel responses, spatial layouts, semantic scales, and modality-specific uncertainty. This paper presents TopoGraph-Fusion, a hierarchical graph-guided dual-modal object detector that formulates fusion as topology-aware reasoning rather than direct feature concatenation. The proposed framework builds a dual-stream backbone for RGB and thermal images, constructs channel-wise topology through a channel-topology graph aggregation module, derives relation-aware spatial and channel global attention from affinity graphs, and replaces fixed feature-pyramid communication with a Graph-Guided Feature-Pyramid Network. A topology-regularized detection objective further encourages stable cross-modal correspondence while suppressing noisy all-to-all connections. Experiments on M3FD, FLIR, RGBTDronePerson, and VEDAI512 cover road scenes, adverse illumination, drone–person perception, and aerial vehicle detection. Within this validation scope, the results and visual analyses indicate that topology-guided fusion improves small-object recall, cross-modal consistency, and robustness under modality imbalance. Full article

Reverse engineering of printed circuit boards is still largely manual when original computer-aided design documentation is unavailable. This paper presents a semi-automatic vision-based pipeline that prepares an editable KiCad schematic draft for use in an Electronic Design Automation (EDA) workflow from paired TOP and BOTTOM board images. The method combines color-profile estimation, pad and through-hole detection, trace segmentation, optical character recognition, component inference, an explicit evidence graph and schematic export with drawn wires. A separate readability step aligns symbols to a grid and reroutes the reconstructed nets with orthogonal wires; it does not change the reconstructed netlist. The primary quantitative evaluation used twelve synthetic KiCad fixtures and three solver configurations: the default sequential pipeline, an opt-in global component solver and an opt-in probabilistic contact solver. These fixtures provide controlled regression cases and are complemented by a small exploratory acquisition trial on real photographed boards. All configurations completed all runs and passed the export round-trip validation without falling back to label-only connectivity. This round-trip check confirms consistency between the internal reconstruction and the exported schematic, but it is reported separately from electrical correctness against the KiCad reference design. The stricter reconstruction-quality criterion still failed on four stress cases involving repeated component chains, long meandering variable-width traces, circular distractors near pads and two-sided transistor layouts. The probabilistic contact solver was therefore kept as an opt-in diagnostic mode rather than enabled by default; it reduced the global pin-to-pin netlist edit distance from 642 to 525 while preserving schematic export checks. The real-board trial indicates that pad and hole detection can transfer to simple photographs, with trace extraction remaining sensitive to uncontrolled illumination and weak copper contrast. The results support the use of the system as a human-in-the-loop reconstruction assistant and identify component grouping, trace-contact reasoning, real-photograph benchmarking and safe missing-edge activation as the main remaining research problems. Full article

Camera localization aims to estimate the six-degree-of-freedom camera pose from a single RGB image and is essential for robotics, autonomous navigation, and augmented reality. Despite recent progress in learning-based localization, robust pose estimation remains challenging under large viewpoint changes, occlusions, repetitive textures, and complex scene layouts. Existing methods mainly rely on local feature matching or independent feature regression, while the structural dependencies among multi-scale visual representations are not fully explored. To address this issue, we propose Uncertainty-aware Bidirectional Graph Learning (UBGL) for RGB-based camera localization. UBGL extracts hierarchical visual features and converts them into graph-structured representations, where nodes describe scene information at different semantic levels. A bidirectional graph learning module is introduced to exchange information between low-level geometric cues and high-level contextual features, enabling joint modeling of local correspondences and global scene structure. In addition, an uncertainty-aware relation modeling strategy estimates the reliability of graph connections and helps reduce the influence of unstable feature interactions during graph reasoning. The refined graph representations are then projected into dense scene-coordinate maps, and the final camera pose is recovered using a differentiable pose solver. Experiments on the 7Scenes and Cambridge Landmarks datasets show that UBGL achieves competitive localization accuracy, especially improving rotation estimation on the 7Scenes benchmark while maintaining comparable performance on Cambridge Landmarks. Ablation studies further demonstrate the effectiveness of bidirectional graph interaction and uncertainty-aware relation modeling. Full article

In 3D integrated circuits, a placement decision that looks acceptable from a 2D wirelength view can still create a local thermal or stress problem after stacking. This issue becomes more visible as the number of tiers and the density of vertical interconnects increase. We propose ThermIC, a placement framework that brings thermal and mechanical risk estimates into the placement loop rather than treating them only as post-layout checks. The novelty of ThermIC does not lie in treating graph neural networks, reinforcement learning, uncertainty-aware learning, or physics-informed regularization as individually new techniques. Instead, ThermIC contributes a placement-time coupling mechanism in which physically typed graph propagation, dense multi-constraint risk prediction, and action-level reinforcement learning feedback are jointly organized for stacked 3D-IC placement. ThermIC uses a heterogeneous graph encoder to carry thermal, stress, timing, and congestion information through the netlist; a constraint head to estimate local hotspot, stress-risk, timing-violation, and congestion probabilities; and a sequential placement policy trained with physics-informed penalties. We evaluate the method on ThermIC-Bench, a simulated corpus with more than 30,000 finite-element samples from 18 heterogeneous 3D-IC designs with 4–8 tiers. Because the present study does not include proprietary industrial circuits, silicon measurements, or a tape-out case, the experimental results are interpreted as simulation-based benchmark evidence rather than final industrial qualification. ThermIC connects the heat-kernel branch to the discretized heat-conduction equation and the stress-filter branch to linear thermo-elastic equilibrium, providing a mechanism-level basis for physical interpretability. The analysis distinguishes offline simulation/training cost from online deployment cost and reports complexity, runtime, and memory scaling for practical large-scale use. Under joint DRC, thermo-mechanical stress, and thermally coupled timing checks, ThermIC obtains an $82.1\%$ physical verification pass rate. The peak-temperature error is $3.1$ °C, the hotspot localization IoU is 0.89, and the number of placement-closure iterations is reduced by $3.7\times$ relative to the heuristic baseline. Together, these benchmark results indicate that early, differentiable multi-physics feedback can make 3D placement less dependent on late correction cycles. Full article

Images captured by vision sensors provide visual evidence for scene understanding, including object appearances, pixel-level regions, and spatial relations among entities. Panoptic Scene Graph Generation (PSG) constructs structured scene representations by grounding visual entities with panoptic masks and predicting relationships among objects and regions. In pair-then-relation PSG pipelines, subject–object pair recall is critical to final triplet recall. However, existing pair proposal approaches mainly score candidate subject–object pairs based on object–query feature matching, while mask-derived spatial cues such as object locations, relative geometry, and local layouts remain underexplored. Consequently, ground-truth subject–object pairs may be excluded from the Top-$K_r$ proposals before relation decoding. To address this problem, this paper proposes a Spatially Aware Pair Proposal Model (SAPPM), which incorporates mask-derived soft centroids, relative geometry, and local-neighborhood context into pair scoring. SAPPM uses Grouped Vector Attention (GVA) to model local spatial interactions and introduces a spatially adaptive gating module to calibrate spatial-branch contributions. Experiments on the PSG dataset under the Scene Graph Detection (SGDet) protocol show that SAPPM achieves competitive performance, reaching 32.53 R@20 and 27.36 mR@20. These results indicate that SAPPM improves PSG performance by enhancing ground-truth pair coverage in the candidate proposal set. Full article

Modular multi-gun DC fast chargers can improve converter-capacity utilization by allowing charging guns to share power modules, but additional internal reachability also increases switching devices, layout complexity, reconfiguration exposure, and fault-related burden. This paper investigates topology and gun-to-module-ratio co-design for modular DC fast chargers from a device-level architecture perspective. A unified screening framework is developed to compare fixed, ring, partitioned, and semi-flexible layouts under common demand patterns, coefficient settings, and probabilistic module outages. A normalized cost-aware planning score evaluates delivered charging service against architecture burden, while exact small-scale benchmarks, repeated-seed sweeps, hotspot cases, robustness analysis, and continuous-operation references are combined to separate robust conclusions from conditional ones. The results show that fixed topology is the most conservative and robust option under balanced demand and high switching burden, whereas partitioned topology gives the most statistically regular behavior across broad sweeps. Semi-flexible layouts are not globally superior; their advantage appears mainly under persistent hotspot demand and moderate switching burden. These findings position bounded module sharing as a conditional charger-design regime for hotspot-prone applications. The results apply under homogeneous-module, graph-level, structured-demand, and proxy-cost assumptions and provide planning-stage architecture-screening guidance instead of hardware-calibrated cost predictions. Full article

Understanding the spatial generation mechanisms of ride-hailing demand is crucial for sustainable urban mobility. However, existing literature largely assumes monocentric urban layouts and globally stationary spatial scales, often overlooking the severe topographical constraints inherent in “strip cities”. To bridge this gap, the present study proposes a novel dual-level analytical framework coupling the Spatially Embedded Laplacian Graph Partition (SE-LGP) algorithm with a Log-Gaussian Multiscale Geographically Weighted Regression (MGWR) model. Taking Jinan, China, as a quintessential strip city, we incorporate spatial penalties to decode its mobility dynamics. Macroscopically, we reveal that substantial topographic friction fragments the workday mobility network into a chain of 23 highly localized micro-circulations. This anisotropic friction results in a notable 41.70% intra-community retention rate, demonstrating that flexible mobility operates within confined functional basins rather than a unified citywide market. Microscopically, the MGWR uncovers significant multiscale spatial heterogeneity: the jobs–housing mismatch is strongly associated with demand at a global macro scale (bandwidth = 1335), whereas public transit integration operates predominantly at a localized micro scale (bandwidth = 44). Crucially, the interaction between topographical friction and infrastructure capacity unveils a highly localized pressure-valve effect (bandwidth = 46), indicating that physical road networks mitigate natural barriers strictly at a micro scale. Comparative analysis quantifies a “spatial fusion effect” during weekends; the relaxation of rigid tidal commuting reveals a structural invariance in built-environment scales (bandwidth = 1335), while the impact intensity of natural topographical friction undergoes a marked spatial inversion. This behavioral elasticity merges fragmented micro-circulations into larger regional communities ($k=20$). The findings indicate that flexible transit is strongly associated with scale-dependent and temporally elastic mechanisms. It provides insights for planners to transition from uniform city-wide fleet dispatching toward region-customized, temporally dynamic mobility management in topographically constrained metropolises. Full article

High spatial resolution wind speed data is very important for wind farm planning, design, operation and maintenance. But due to cost, site and other factors, it is impossible to build a large number of anemometer towers to obtain high spatial resolution measured data. Therefore, this paper proposes a method for generating wind speed data in renewable energy bases based on physics-informed neural networks, which incorporates fluid mechanics control equations such as the Navier−Stokes equation as physical constraints into the model training process. The model’s input includes the wind speed data and the wind direction data of the anemometer towers as input, as well as the geographical difference data between the input anemometer towers and the output point, enabling to learn the mapping relationship between geographical differences and wind speed differences at different locations, achieving the goal of generating high spatial resolution wind speed data. Using normalized root mean absolute error (NMAE) to measure the model error, the average wind speed error and the average wind direction error of the proposed wind speed data generation method on different test sets are 8.28% and 10.50%, which is lower than that of BP neural network and graph convolutional neural network, and can provide more refined data support for wind turbine layout planning and wind farm power prediction of renewable energy bases. Full article

In the fabric warehouse, order picking operations face high labor intensity and rising operational costs, requiring urgent optimization. This study investigates the order scheduling and task assignment problem within an elastic staffing framework, where temporary labor recruitment and real-time task allocation need to be adjusted dynamically in response to fluctuations in order volumes. Nevertheless, conventional approaches often suffer from severe computational bottlenecks under such highly dynamic conditions, and struggle to maintain optimal solutions when demand undergoes large and frequent fluctuations. To address these challenges, this study proposes a Graph Transformer Policy Network with Proximal Policy Optimization (GTP-PPO), which combines graph structure features with a global attention mechanism. First, the return picking strategy and the S-shaped picking strategy are compared and analyzed in the fabric warehouse scenario. The results reveal that the return strategy is more suitable for the studied warehouse layout. Subsequently, a mixed-integer programming (MIP) model and a GTP-PPO model are established for optimizing order dispatching and scheduling. Finally, an empirical analysis is carried out based on the peak order day of the year in the fabric warehouse. The results demonstrate that the proposed GTP-PPO model not only achieves near-global optimal solutions (gap < 4%) comparable to the MIP model, but also exhibits robust real-time decision-making capabilities under dynamically increasing order volumes and unexpected disruptions. Compared to the MIP model, the GTP-PPO approach reduces unskilled labor hours by 84.80% and decreases operational volatility by 27.60%, with only a 3.52% increase in operational costs. Full article