Search Results (1,122)

Industrial maintenance and fault-diagnosis systems generate textual records, such as maintenance work orders, service requests, causal analyses, and troubleshooting solutions. These records contain domain-specific named entities that provide valuable knowledge for intelligent monitoring, fault diagnosis, maintenance decision support, and industrial knowledge graph construction. However, in practical industrial environments, maintenance records are strongly associated with specific equipment types, production processes, fault modes, and enterprise-specific terminology. As a result, entity schemas vary across systems, new entity types emerge with equipment updates, and high-quality annotation requires substantial domain expertise. These factors make it difficult to obtain sufficient labeled samples for each industrial entity type. Under such low-resource conditions, conventional supervised named entity recognition (NER) models tend to suffer from unstable entity boundary detection and biased entity representations. To address these challenges, this paper proposes a boundary-aware knowledge graph-guided adaptive prototype correction framework for few-shot NER in industrial maintenance information systems. The proposed framework first introduces a boundary-aware span detection mechanism to improve entity localization in noisy and irregular maintenance texts. A knowledge graph-guided adaptive prototype correction module is then designed to construct entity class prototypes from limited support examples, reducing prototype bias caused by sparse annotations. Experiments are conducted on two representative industrial datasets, MaintIE and CFDK, covering maintenance short texts and fault-diagnosis records. Experimental results show that the proposed framework achieves an average Micro-F1 improvement of 1.89 percentage points over the strongest compared baseline across 12 episodic settings on the two industrial datasets: three MaintIE coarse-grained settings, six MaintIE fine-grained settings, and three CFDK settings. The ablation and sensitivity analyses further indicate that boundary-aware span modeling and KG-guided prototype correction jointly contribute to low-resource entity classification. This study provides a data-efficient information extraction solution for AI-enabled industrial knowledge acquisition, fault diagnosis, and maintenance decision support. Full article

This paper presents the design, construction, and experimental evaluation of a low-cost, portable solar tracking system for monitoring ultraviolet solar radiation in real time. It integrates a Raspberry Pi Zero 2W as the embedded control unit, an AS7331 spectral sensor to measure UVA, UVB, and UVC irradiance, two 270° servomotors to position the system toward the sun, an NEO-6M GPS module to geolocate the system, and a DS3231 real-time clock to synchronize the time. To enable autonomous outdoor operation, a multistage power supply architecture based on a solar panel, a rechargeable battery, and LM2596 and MP1584EN DC-DC regulators was implemented. The tracking algorithm uses astronomical equations to estimate the solar azimuth and elevation and updates the sensor orientation during daylight hours. This allows the UV sensor to remain approximately normal to the incoming solar radiation. Experimental tests were conducted in Arequipa, Peru. The recorded data included UVA, UVB, and UVC irradiance; sensor temperature; geographic coordinates; time; and solar angles. The measured UV profiles exhibited the anticipated diurnal behavior: maximum values around solar noon, higher UVA levels than UVB levels, and minimal UVC levels due to atmospheric absorption. We compared the radiometric response with reference information from EarthKit, PVGIS 5.3, SAMPA, and a Davis Vantage Pro 2 weather station. We evaluated the solar positioning performance against Stellarium, NOAA, and the NREL Solar Position Algorithm. Across the complete five-day validation at three daily evaluation times, the maximum percentage errors were 0.0584% for azimuth and 0.5059% for elevation relative to the NREL SPA, NOAA, and Stellarium reference calculations. The results demonstrate that the proposed system constitutes an embedded, portable, autonomous, and low-cost platform for in situ monitoring of solar ultraviolet radiation. Due to its modular architecture, georeferencing capability, time synchronization, and independent power supply, the prototype can be used as a mobile measurement unit or as part of a distributed network of UV stations at various locations in Arequipa. In this regard, the system enables multipoint measurement campaigns, complements fixed weather stations, validates solar models, and generates local experimental data for the spatial and temporal assessment of the solar UV resource under real-world field conditions. Full article

Population ageing and the need to provide affordable, healthy, and energy-efficient housing represent important sustainability challenges in many European countries. Sustainable senior housing should not only reduce environmental impacts through low-carbon construction technologies but also ensure high indoor environmental quality and occupant well-being. This paper presents a preliminary assessment of predicted thermal comfort and local thermal discomfort in a prototype senior home constructed a prefabricated timber-based building system incorporating renewable and recycled materials and designed to support low operational energy demand. The research forms part of a broader development study of technology, in which indoor thermal conditions were monitored in a prototype building consisting of two 30 m² residential units intended for older adults. Predicted thermal comfort was evaluated using the PMV (Predicted Mean Vote) and PPD (Predicted Percentage of Dissatisfied) indices together with local thermal discomfort criteria. The analysis was based on short-term winter and summer measurement campaigns conducted in the prototype building. The results indicated category B thermal environment conditions in both winter and summer according to ISO 7730. In winter, local discomfort associated with a cool floor corresponded to category C, while summer conditions met category B requirements without significant local discomfort. The findings provide preliminary evidence that timber-based low-carbon construction technologies can support acceptable indoor thermal conditions while addressing environmental and social sustainability objectives related to an ageing population. The study also identifies the need for longer monitoring campaigns and future investigations involving older occupants to validate actual thermal sensation and further optimize sustainable senior housing solutions. Since the building was unoccupied during the measurements, the results should be interpreted as a prediction of predicted thermal comfort conditions rather than an assessment of thermal sensations experienced by older adults. Full article

With the rapid development of the Industrial Internet of Things, data-driven deep learning has achieved remarkable success in bearing fault diagnosis. However, traditional static models suffer from catastrophic forgetting when facing continuously emerging fault categories and limited edge storage. Existing class-incremental learning frameworks expose critical limitations when applied to 1D vibration signals on micro edge devices, including feature space oscillation, difficulty in anchoring lightweight classifiers, and prototype drift over long incremental cycles. To address these challenges, this paper proposes a novel end-to-end class-incremental fault diagnosis method based on lightweighting and feature calibration tailored for severe memory-constrained conditions. Specifically, a lightweight feature extraction mechanism based on an $L_2$ constraint is introduced to replace computationally expensive similarity distillation, effectively suppressing feature space oscillations and providing stable spatial coordinates for old knowledge. Moreover, a mandatory balanced center–margin hybrid replay (CAHM) strategy is designed to balance class representation while proportionally retaining class center prototypes and marginal hard examples, balancing the anchor accuracy of the Nearest Class Mean (NCM) classifier and the discriminability of the decision boundary. Furthermore, an ultra-low-cost linear prototype calibration module is constructed using a learnable affine transformation to actively redirect shifted old class prototypes with negligible inference latency. Extensive long-tail incremental experiments on the CWRU bearing dataset demonstrate that the proposed method forms a highly synergistic anti-forgetting closed loop. Under an extremely limited memory budget ($K=40$), the proposed framework achieves an outstanding final average accuracy of 98.92% after five incremental stages, significantly outperforming mainstream baselines such as iCaRL, PRIL, and SCKD and exhibiting exceptional robustness for continuous online monitoring on industrial edge devices. Full article

Automated and accurate segmentation of retinal vessels in fundus images provides pivotal evidence for ophthalmologists to effectively and non-invasively diagnose prevalent ocular and systemic diseases. However, existing methods often struggle to maintain the topological continuity of fine-diameter capillaries, leading to severe vascular discontinuity and fragmented segmentation results in challenging scenarios such as complex, irregular microvascular branches, pathological lesions, and high-noise conditions. To address these limitations, we developed a novel symmetric dual-branch network with frequency-aware decomposition and prototype refinement (FDPR-DBNet). Specifically, the network initially utilizes the discrete wavelet transform (DWT) to decompose input retinal images into high-frequency and low-frequency components, which are then processed by a structurally symmetric dual-branch encoder. In the high-frequency branch, the parallel atrous convolution activation (PACA) module is designed to explore fine-grained contour and edge patterns related to vessel terminals and microvessels. Concurrently, within the low-frequency branch, the spatial-frequency characteristic activation (SFCA) unit is constructed by introducing the selective state-space model ($S6$) and Fourier transform to extract salient structural backbones. Moreover, the spatial attention residual fusion (SARF) module and cross-frequency fusion (CFF) block are designed to establish a symmetric guidance mechanism, effectively reinforcing bidirectional feature interaction and alignment across different frequency spectra to eliminate vascular fragmentation. Furthermore, by embedding global and local window self-attention into the Transformer, we formulated the cross-scale enhancement (CSE) module, comprising global semantic enhancement (GSE) and local detail enhancement (LDE), to model multi-scale contextual semantic correlations and enhance the adaptive recognition of vessel structures. Ultimately, we embedded the multi-wise prototype characteristic refinement (MPCR) component into the decoder to correct cross-scale semantic features through a dynamic calibration mechanism, while introducing a new connectivity loss to strictly enforce topological continuity. Experimental results on four publicly available retinal image datasets (DRIVE, CHASE_DB1, STARE, and IOSTAR) demonstrate that the proposed model achieves competitive performance and effectively preserves vascular integrity even in the presence of fundus lesions and noise. Full article

Cross-condition fault diagnosis is important for ensuring the reliable operation of mechanical systems. However, most existing methods assume that the source and target domains share identical fault label spaces and data distributions, limiting adaptation to unknown faults and cross-condition shifts in practical industrial scenarios. To address this issue, a structured, reliable, domain-specific open-set domain adaptation method is proposed. The proposed method first constructs a domain-specific batch normalization-based feature extraction network, in which independent normalization branches model statistical discrepancies under different operating conditions; it then designs a cross-domain structured representation consolidation module to enhance feature discriminability through source-domain anchor compactness, target-domain multi-view contrastive, and prototype entropy regularization constraints; an open-set boundary learning mechanism is further introduced to establish a discriminative boundary between known and unknown classes; finally, a reliability-aware pseudo-label propagation strategy refines target-domain pseudo-labels and imposes separate prediction-consistency constraints on known and unknown classes. Experimental results on the CWRU bearing dataset and the self-built rolling bearing dataset show that the proposed method achieves average H-scores of 93.32% and 97.65%, respectively, on open-set transfer tasks. Compared with several baseline methods, the proposed method achieves a better balance between known-class recognition and unknown-class detection, thereby improving cross-condition open-set fault diagnosis performance. Full article

This work presents the design evolution, fabrication, and characterisation of an additively manufactured ring-type thermal airflow sensor for in-pipe flow monitoring, developed employing exclusively Fused Deposition Modelling (FDM) additive manufacturing technology and a commercially available Carbon Nanotube (CNT)-enriched Biopolymer Polylactic Acid (PLA) composite filament. The design evolution proceeds through three progressive stages. In the first stage, a flat heater element is characterised through Constant-Current (CC) Joule heating experiments in order to derive the corresponding Temperature Coefficient of Resistance (TCR) and Thermal Resistance from the obtained experimental data. Consequently, a Finite Element Method (FEM) model implemented in COMSOL Multiphysics^® and calibrated with the extracted material parameters validates the experimental temperature–power relationship and predicts the convective cooling behaviour at various airflow velocities. In the second stage, the geometry is optimised by introducing a conductive trace with a reduced-cross-section central region; as a result, an equivalent thermal localisation is achieved at approximately 26% lower supplied power with respect to the initial heating element, enabled by the design freedom inherent in the FDM process. We should note that the specific sensing geometry can also be directly embedded into any 3D-printed structural component (e.g., a bracket or housing), enabling simultaneous local thermal heating and/or thermal monitoring together with structural functionality within a single printed part. In the third and final stage—the target device—a fully monolithic ring-type airflow sensor is directly integrated into a 3D-printed pipe segment during the printing process. Under constant-current excitation at 40 mA, the device exhibits a monotonically decreasing resistance with increasing airflow (ΔR ≈ 117 Ω over 0–4 m/s) due to convective cooling, while in a single flow-interruption cycle, approximately 79% of the flow-induced resistance change was recovered upon flow removal, with a residual offset of approximately 3% of the heated baseline. A coupled electrothermal FEM model of the device further supports the experimental response by comparing the simulated temperature rise with the values inferred from resistance measurements, while also clarifying the role of the effective internal convective cooling conditions imposed by the pipe geometry. Key features of the proposed device are low raw-consumables cost, fast on-site manufacturing employing a commercially available desktop 3D printer, monolithic construction free of wire-bonded interconnections, and simplicity, indicating its potential for flow monitoring and condition-based maintenance systems aboard vessels as well as in a wide range of industrial sectors. We should note that the present characterisation was performed under laboratory conditions employing a single prototype per design stage; the effects of humidity, salt exposure, vibration, temperature cycling, and material-batch variability remain to be assessed prior to shipboard deployment. Full article

The weakly supervised aircraft fine-grained detection task remains challenging due to the lack of fine-grained category labels. In addition, the high inter-class similarity among aircraft categories and complex background interference further hinders discriminative feature learning. To address these challenges, we propose a weakly supervised fine-grained aircraft detection framework based on prior-knowledge prototype learning. Specifically, a Prior-knowledge-based Soft-label Generator (PSG) is designed to exploit geometric and semantic priors, enabling the propagation of fine-grained category annotations from a few annotated samples to large-scale unlabeled data through reliable soft-label generation. To alleviate category ambiguity caused by highly similar aircraft categories, a Text Prototype Alignment (TPA) strategy is introduced to align vision-language semantic knowledge with the remote sensing domain and construct discriminative semantic prototypes. Furthermore, a Feature Purification Module (FPM) is developed to suppress background noise through morphology-guided feature purification and enhance aircraft-specific representations. Experiments on the FAIR1M-Aircraft dataset demonstrate that the proposed method achieves 30.5% mAP with only 50 fine-grained annotated images, outperforming the baseline by 10.1 mAP points. Full article

This paper presents the design, digital implementation, and experimental validation of a new 50 W scaled prototype of a quadratic buck converter (48 V to 5 V, 10 A) controlled by a finite control set model predictive control (FCS-MPC) strategy. The converter utilizes its quadratic step-down topology to achieve high voltage conversion gain without extreme duty cycles, making it suitable for low-power applications requiring precise voltage regulation. The proposed methodology encompasses the theoretical design of the power stage, the development of the experimental prototype based on a C2000 Digital Signal Processor DSP, and a comparative performance assessment between the proposed FCS-MPC and a conventionally tuned PD controller. An iterative tuning and real-time validation process is employed to optimize both the converter parameters and the control law, ensuring closed-loop stability and enhanced dynamic response under line and load disturbances. The experimental results demonstrate that the FCS-MPC strategy significantly outperforms the PD controller in terms of output voltage regulation, settling time (4.2 s vs. 5 ms), and disturbance rejection (<2 ms recovery). The main contribution of this work is the construction of a new scaled prototype and the experimental validation of a predictive control strategy for a high-gain DC–DC converter, positioning the FCS-MPC-controlled quadratic buck converter as a viable solution for modern applications demanding high energy efficiency and robustness. Full article

This study presents a continuous geometric framework utilizing a novel quartic Overhauser (QOVR) spline with adjustable local parameters. The proposed spline is constructed as an alternative representation from an algebraic blending mechanism of three parabolic segments defined over a five-point control configuration. This structure enables flexible spline definitions through internal parametric knots without modifying the spatial locations of the control points. The resulting formulation maintains $C^1$ continuity, ensuring smooth transitions between adjacent spline segments. The QOVR spline offers structural advantages in interactive design and computational geometry environments. To evaluate the computational steps, a dedicated object-oriented prototyping framework was introduced to organize, visualize, and analyze the proposed spline. This arrangement facilitates continuous visualization, interactive adjustment of local parameters, and multi-segment implementations. Exported coordinates further allow independent numerical verification in external computational environments when required. The QOVR spline achieves a balance between geometric flexibility and boundary condition invariance. By allowing independent parametric knot adjustments without disrupting the global continuity constraints, the proposed design offers direct local morphological flexibility during geometric synthesis. This study introduces the QOVR spline integrated into an object-oriented prototyping framework, offering a distinct algebraic design for interactive geometric environments and computer-aided design (CAD) setups. Full article

In the multivariate time-series fault diagnosis task for aviation electro-hydrostatic actuators (EHA), the overall signal morphologies of different fault categories are relatively similar, while the key discriminative differences are hidden in local segments and variations in variable coupling. Therefore, existing Transformer-based methods usually have difficulty characterizing local specificity. To address this issue, this paper proposes a Local Prototype-Global Generic Dual-branch Transformer (LPG-Former). First, to obtain local information capable of characterizing class differences, a class-specific discriminative prototype (CDP) is constructed. The CDP selects discriminative time points from the time-series samples of each class to capture key local morphological variations, and constructs local prototypes carrying class-related local differential features. To further improve the ability of the CDP to capture multivariate fault coupling relationships, a multivariate coupling-aware prototype matching strategy (MCPM) is designed. The MCPM extends univariate prototypes into multivariate local prototype blocks and jointly measures local dissimilarity, variable correlation, and trend consistency, thereby enabling prototype learning with awareness of multivariate coupling relationships. Finally, to fuse local discriminative information and global temporal information, a dual-branch Transformer is constructed. LPG-Former encodes the differential features between the CDP and the best-fit subsequence (BFS) of the input sample through a Local Prototype Transformer, and complements global generic information through a Global Generic Transformer, thereby achieving collaborative dual-branch representation. Experimental results on an eight-class EHA operating-state dataset show that LPG-Former achieves an accuracy of 98.74% and an F1-score of 98.78%, significantly outperforming classical methods such as InceptionTime and TapNet. Full article

To address the issues of trajectory deviation, attitude fluctuations, and reduced driving stability that electric tracked tractors often encounter on uneven terrain in hilly and mountainous fields, this paper proposes a driving stability control strategy based on torque distribution. First, based on the operational requirements of hilly and mountainous terrain, the overall structural design and key parameter calibration of a dual-side independently driven electric crawler tractor were completed. Second, a vehicle dynamics model and a C/D/E-level stochastic road profile model were constructed, and a four-stage progressive driving control strategy was designed, consisting of path analysis, torque distribution, heading-error compensation, and closed-loop feedback. Finally, a co-simulation platform was established using RecurDyn 2024–MATLAB/Simulink R2023b, and straight-line driving simulations were conducted under C-, D-, and E-grade stochastic road excitations at low speeds of 1–5 km/h. A physical prototype was also built and validated through field tests. The results indicate that, under the tested straight-line driving conditions, the proposed control strategy improves trajectory tracking accuracy, yaw stability, roll stability, and speed smoothness compared with the constant-torque distribution strategy. Field tests with the actual vehicle provided preliminary validation of the strategy’s effectiveness in improving driving stability and path-keeping capability under the measured uneven-terrain condition. Full article

High-speed electric machines enhance power density and eliminate the need for a gearbox in waste heat recovery microturbine systems. However, existing designs often suffer from high manufacturing costs and complex cooling requirements. This study presents the development, experimental validation, and comparative analysis of three high-speed machine designs. First, a lower-speed induction machine prototype, constructed using standardized components, was tested at an operating speed of 13,000 rpm. This prototype enabled experimental validation of the numerical model used for loss calculations. Experimental results showed total losses of 7.89 kW, closely matching the simulated value of 7.75 kW at an output power of 93.1 kW, i.e., an efficiency of 92.19%. Building on these findings, two smaller machine prototypes were developed: one featuring an induction squirrel-cage rotor and the other employing a surface-mounted permanent magnet rotor topology. Both machines were designed and evaluated using finite element analysis and conjugate heat transfer simulations. Their performance was analyzed under both sinusoidal and pulse-width-modulated voltage supply conditions. At an operating speed of 14,000 rpm, the permanent magnet machine outperformed the induction machine, achieving 63.2 kW of mechanical power and an efficiency of 96.21%, while operating at lower temperatures. In comparison, the induction machine delivered 52.4 kW of mechanical power with an efficiency of 94.64%. The primary novelty and contribution of this work lie in the implementation of a two-pole machine architecture capable of achieving an output power of 100 kW at operating speeds between 20,000 and 25,000 rpm. Compared with similar solutions reported in the literature, the proposed machines feature a simplified bearing arrangement and a more straightforward liquid-cooling system. These characteristics have the potential to reduce manufacturing costs and simplify maintenance during operation. Full article

This study proposes a novel approach for performance improvement of an acoustic energy harvester integrated with a flexible polyvinylidene fluoride-based piezoelectric nanogenerator by exploiting the thermoacoustic effect. A prototype of the acoustic energy harvester was first designed and constructed. The influence of the temperature difference across the thermoacoustic stack on the acoustic pressure and open-circuit voltage amplitudes was then experimentally examined. Subsequently, the effects of excitation frequency and driving voltage on the performance of the acoustic energy harvester were systematically analyzed. The results demonstrate that the thermoacoustic effect can be effectively employed to enhance acoustic oscillations and, consequently, improve the electrical output. As the excitation frequency changes, the acoustic oscillations inside the acoustic energy harvester and the open-circuit voltage of the piezoelectric nanogenerator can be either amplified or suppressed depending on the frequency range. In addition, optimal driving voltages exist at which the amplification of acoustic pressure and open-circuit voltage is maximized. Specifically, at an excitation frequency of 85 Hz, a driving voltage of 3.5 V, and a stack temperature difference of 101.5 °C, a maximum pressure amplification factor of 3.73 and a maximum voltage amplification factor of 1.15 are obtained. This study successfully demonstrates the feasibility of utilizing the thermoacoustic effect to amplify acoustic oscillations within a resonator, offering a new pathway for enhancing the performance of acoustic energy harvesters and expanding the application potential of thermoacoustic technology. Full article

In the leveling process of ultra-high-strength steel plates, sample scarcity—driven by high prototyping costs and small-batch production—coupled with a narrow and unevenly distributed feasible region due to high yield-to-tensile ratios and limited ductility, impedes the balanced optimization of plate shape quality and energy consumption. To address this issue, this paper develops an optimization framework for the balanced trade-off between these two objectives. First, a high-precision response surface model based on Box–Behnken experimental design and finite element simulation was constructed using initial wave height, entry roll reduction, exit roll reduction, and leveling speed as key process parameters; peak residual stress difference (characterizing potential sheet quality) and leveling energy consumption as co-optimization objectives; and post-leveling flatness as a constraint. Next, by introducing the NSGA-II multi-objective genetic algorithm, the Pareto optimal solution set for the quality and energy efficiency objectives was obtained, clearly revealing the trade-off relationship between the two; furthermore, the TOPSIS decision-making method was employed to select the comprehensive optimal process scheme that achieves a balance between quality and energy efficiency from the Pareto solution set. An adaptive recommendation curve for the leveling process parameters of MS1500 ultra-high-strength steel plates was established, covering an initial wave height range of 10.5–14.6 mm, thereby enabling intelligent parameter matching based on different incoming material conditions. Finally, industrial validation demonstrated that this optimized scheme significantly reduced leveling energy consumption while ensuring that post-leveling flatness meets the high-quality requirement of less than 3.5 mm·m⁻¹. This achieves a balanced optimization of quality and energy efficiency. This study provides a reliable theoretical basis and practical engineering solution for the efficient and environmentally friendly leveling production of ultra-high-strength steel. Full article