Search Results (2,361)

In tool wear monitoring, the efficient fusion of multi-source sensor signals poses significant challenges due to their inherent heterogeneous characteristics. In this paper, we propose a Multi-Sensor Multi-Domain feature fusion Transformer (MSMDT) model that achieves precise tool wear prediction through innovative feature engineering and cross-modal self-attention mechanisms. Specifically, we first develop a physics-aware feature extraction framework, where time-domain statistical features, frequency-domain energy features, and wavelet packet time–frequency features are systematically extracted for each sensor type. This approach constructs a unified feature matrix that effectively integrates the complementary characteristics of heterogeneous signals while preserving discriminative tool wear signatures. Then, a position-embedding-free Transformer architecture is constructed, which enables adaptive cross-domain feature fusion through joint global context modeling and local feature interaction analysis to predict tool wear values. Experimental results on the PHM2010 demonstrate the superior performance of MSMDT, outperforming state-of-the-art methods in prediction accuracy. Full article

The growing demand for power and the rising awareness of the need to reduce carbon footprints have led to wider acceptance of photovoltaic (PV)-integrated microgrids. PV-based microgrids have numerous significant advantages over other distributed energy resources; however, creating a dependable protection scheme for the DC microgrid is difficult due to the closely resembling current and voltage profiles of PV array faults and line faults in the DC network. The conventional methods fail to clearly discriminate between them. In this regard, a fault-resilient scheme exploiting the inherent characteristics of Fourier–Bessel Series Expansion and Empirical Wavelet Transform (FBSE-EWT) has been utilized in the present work. In order to enhance the efficacy of the bagging tree-based ensemble classifier, Artificial Gorilla Troop Optimization (AGTO) has been used to tune the hyperparameters. The hybrid protection approach is proposed for accurate fault detection, discrimination between scenarios (source-side fault and line-side fault), and classification of various fault types (pole–pole and pole–ground). The discriminatory attributes derived from voltage and current signals recorded at the DC bus using the hybrid FBSE-EWT have been utilized as an input feature set for the AGTO tuned bagging tree-based ensemble classifier to perform the intended tasks of fault detection and discrimination between source faults (PV array faults) and line faults (DC network). The proposed approach has been found to outperform the decision tree and SVM techniques, demonstrating reliability in terms of discriminating between the PV array faults and the DC line faults and resilience against fluctuations in PV irradiance levels. Full article

Non-invasive ultrasound diagnosis, combined with deep learning, is frequently used for detecting thyroid diseases. However, real-time detection on portable devices faces limitations due to constrained computational resources, and existing models often lack sufficient capability for small object detection of thyroid nodules. To address this, this paper proposes an improved lightweight small object detection network framework called LISA-YOLO, which enhances the lightweight multi-scale collaborative fusion algorithm. The proposed framework exploits the inherent symmetrical characteristics of ultrasound images and the symmetrical architecture of the detection network to better capture and represent features of thyroid nodules. Specifically, an improved depthwise separable convolution algorithm replaces traditional convolution to construct a lightweight network (DG-FNet). Through symmetrical cross-scale fusion operations via FPN, detection accuracy is maintained while reducing computational overhead. Additionally, an improved bidirectional feature network (IMS F-NET) fully integrates the semantic and detailed information of high- and low-level features symmetrically, enhancing the representation capability for multi-scale features and improving the accuracy of small object detection. Finally, a collaborative attention mechanism (SAF-NET) uses a dual-channel and spatial attention mechanism to adaptively calibrate channel and spatial weights in a symmetric manner, effectively suppressing background noise and enabling the model to focus on small target areas in thyroid ultrasound images. Extensive experiments on two image datasets demonstrate that the proposed method achieves improvements of 2.3% in F1 score, 4.5% in mAP, and 9.0% in FPS, while maintaining only 2.6 M parameters and reducing GFLOPs from 6.1 to 5.8. The proposed framework provides significant advancements in lightweight real-time detection and demonstrates the important role of symmetry in enhancing the performance of ultrasound-based thyroid diagnosis. Full article

The suppression of Rayleigh backscattering noise in a resonant fiber optic gyro (RFOG) is accompanied by the emergence of residual amplitude modulation (RAM) effects, which impact the bias performance of the RFOG output. In this paper, we propose a double demodulation technique that integrates reciprocal modulation and RAM feedback. By utilizing reciprocal modulation–demodulation along with a RAM feedback control method, we effectively suppress both RAM and laser frequency noise. Furthermore, the inherent suppression characteristics of the double modulation–demodulation scheme facilitate effective backscatter noise reduction. As a result, the gyro angular random walk of the RFOG has improved to $3^{°}/\sqrt{\mathrm{h}}$, and the long-term bias instability has been enhanced to 0.1°/h over a test duration of 10 h. Full article

Background: Cognitive behavioral therapy (CBT) has shown consistent efficacy in individuals with psychosis, as supported by many trials. One classical distinction is that between affective and non-affective psychosis. Few studies have specifically examined the possible moderating role of substantial affective elements. In this systematic review and meta-regression analysis, we assess how CBT response differs across the affective spectrum in psychosis. Methods: We included studies assessing various CBT modalities, including third-wave therapies, administered in people with psychosis. The study protocol is published in the Open Science Framework. Meta-regression was conducted to assess whether the proportion of participants with affective psychosis (AP), as proxied by a documented diagnosis of schizoaffective (SZA) disorder, moderated CBT efficacy across positive, negative, and depressive symptom domains. Results: The literature search identified 4457 records, of which 39 studies were included. The median proportion of SZA disorder participants was 17%, with a total of 422 AP participants represented. Meta-regression showed a trend toward lower CBT efficacy for positive symptoms with a higher SZA disorder proportion (β = +0.10 SMD per 10% increase in AP; p = 0.12), though it was not statistically significant. No significant associations were found for negative (β = +0.05; p = 0.73) or depressive symptoms (β = −0.02; p = 0.78). Heterogeneity was substantial across all models (I² ranging from 54% to 80%), and funnel plot asymmetry was observed in negative and depressive symptoms, indicating possible publication bias. Risk of bias assessment showed the anticipated inherent difficulty of psychotherapies in blinding and possibly dropout rates affecting some studies. Conclusions: Affective symptoms may reduce the effectiveness of CBT for positive symptoms in psychotic disorders, although the findings did not reach statistical significance. Other patient-level characteristics in psychosis could indicate which patients can benefit most from CBT modalities. Full article

This study explores the transformative potential of blockchain technology in overhauling conventional housing rental systems. It specifically addresses persistent issues, such as information asymmetry, fraudulent listings, weak Rental Agreements, and data breaches. A comprehensive review of ten academic publications highlights the architectural frameworks, underlying technologies, and myriad benefits of decentralized rental platforms. The intrinsic characteristics of blockchain—immutability, transparency, and decentralization—are pivotal in enhancing the credibility of rental information and proactively preventing fraudulent activities. Smart contracts emerge as a key innovation, enabling the automated execution of Rental Agreements, thereby significantly boosting efficiency and minimizing reliance on intermediaries. Furthermore, Decentralized Identity (DID) solutions offer a robust mechanism for securely managing identities, effectively mitigating risks associated with data leakage, and fostering a more trustworthy environment. The suitability of platforms such as Hyperledger Fabric for developing such sophisticated rental systems is also critically evaluated. Blockchain-based systems promise to dramatically increase market transparency, bolster transaction security, and enhance fraud prevention. They also offer streamlined processes for dispute resolution. Despite these significant advantages, the widespread adoption of blockchain in the rental sector faces several challenges. These include inherent technological complexity, adoption barriers, the need for extensive legal and regulatory adaptation, and critical privacy concerns (e.g., ensuring compliance with GDPR). Furthermore, blockchain scalability limitations and the intricate balance between data immutability and the necessity for occasional data corrections present considerable hurdles. Future research should focus on developing user-friendly DID solutions, enhancing blockchain performance and cost-efficiency, strengthening smart contract security, optimizing the overall user experience, and exploring seamless integration with emerging technologies. While current challenges are undeniable, blockchain technology offers a powerful suite of tools for fundamentally improving the rental market’s efficiency, transparency, and security, exhibiting significant potential to reshape the entire rental ecosystem. Full article

Terahertz (THz) has attracted substantial attention owing to its unique advantages in high-speed communications. However, conventional THz waveguide systems are inherently constrained by high transmission losses, stringent fabrication precision requirements, and extreme sensitivity to structural defects. Topological edge states with topological protection have driven significant advancements in THz wave manipulation. Nevertheless, the width of the topological waveguide based on edge states remains restricted. In this work, we put forward a type of spin photonic crystal with three-layer heterostructures, where large-area topological waveguide states are demonstrated. The results show that these topological waveguide states are localized within the region of Dirac photonic crystals. They also display spin-momentum-locking characteristics and maintain strong robustness against defects and sharp bends. Furthermore, a THz beam splitter and a topological beam modulator are implemented. The designed heterostructures expand the applications of multi-functional topological devices and provide a prospective pathway for overcoming the waveguide bottleneck in THz applications. Full article

This mini-review emphasizes the potential of biomass-derived materials as sustainable components for next-generation electrochemical energy storage systems. Biomass obtained from abundant and renewable natural resources can be transformed into carbonaceous materials. These materials typically possess hierarchical porosities, adjustable surface functionalities, and inherent heteroatom doping. These physical and chemical characteristics provide the structural and chemical flexibility needed for various electrochemical applications. Additionally, biomass-derived materials offer a cost-effective and eco-friendly alternative to traditional components, promoting green chemistry and circular resource utilization. This review provides a systematic overview of synthesis methods, structural design strategies, and material engineering approaches for their use in lithium-ion batteries (LIBs), lithium–sulfur batteries (LSBs), and supercapacitors (SCs). It also highlights key challenges in these systems, such as the severe volume expansion of anode materials in LIBs and the shuttle effect in LSBs and discusses how biomass-derived carbon can help address these issues. Full article

The safe and reliable operation of autonomous vehicles is contingent on comprehensive testing. However, the operational scenarios are inexhaustible. Corner cases, which critically influence autonomous vehicle safety, occur at an extremely low probability and follow a long-tail distribution. Corner cases can be defined as combinations of driving task and scenario elements. These scenarios are characterized by low probability, high risk, and a tendency to reveal functional limitations inherent to autonomous driving systems, triggering anomalous behavior. This study constructs a novel corner case dataset using naturalistic driving data, specifically tailored for autonomous vehicle testing. A scenario marginality quantification method is designed to analyze multi-source naturalistic driving data, enabling efficient extraction of corner cases. Heterogeneous scenarios are systematically transformed, resulting in a dataset characterized by diverse interaction behaviors and standardized formatting. The results indicate that the scenario marginality of the dataset constructed in this study is 2.78 times that of mainstream naturalistic driving datasets, and the scenarios exhibit considerable diversity. The trajectory and velocity fluctuations, quantified at 0.013 m and 0.021 m/s, respectively, are consistent with the kinematic characteristics of real-world driving scenarios. These results collectively demonstrate the dataset’s high marginality, diversity, and applicability. Full article

Soft manipulators have garnered significant research attention in recent years due to their flexibility and adaptability. However, the inherent flexibility of these manipulators imposes limitations on their load-bearing capacity and stability. To address this, this study compares various variable stiffness technologies and proposes a novel design concept: leveraging the phase-change characteristics of low-melting-point alloys (LMPAs) with distinct melting points to fulfill the variable stiffness requirements of soft manipulators. The pneumatic structure of the manipulator is fabricated via 3D-printed molds and silicone casting. The manipulator integrates a pneumatic working chamber, variable stiffness chambers, heating devices, sensors, and a central channel, achieving multi-stage variable stiffness through controlled heating of the LMPAs. A steady-state temperature field distribution model is established based on the integral form of Fourier’s law, complemented by finite element analysis (FEA). Subsequently, the operational temperatures at which the variable stiffness mechanism activates, and the bending performance are experimentally validated. Finally, stiffness characterization and kinematic performance experiments are conducted to evaluate the manipulator’s variable stiffness capabilities and flexibility. This design enables the manipulator to switch among low, medium, and high stiffness levels, balancing flexibility and stability, and provides a new paradigm for the design of soft manipulators. Full article

The growing need for high-efficiency and reliable propulsion systems in naval applications, particularly within the evolving landscape of submarine warfare, has led to an increased interest in multiphase Permanent Magnet AC motors. This study presents a modelling and simulation approach for a 3MW, seventeen-phase Permanent Magnet AC motor designed for submarine propulsion, integrating an AI-based drive control system. Despite the advantages of multiphase motors, such as higher power density and enhanced fault tolerance, significant challenges remain in achieving precise torque and variable speed, especially for externally mounted motors operating under deep-sea conditions. Existing control strategies often struggle with the inherent nonlinearities, unmodelled dynamics, and extreme environmental variations (e.g., pressure, temperature affecting oil viscosity and motor parameters) characteristic of such demanding deep-sea applications, leading to suboptimal performance and compromised reliability. Addressing this gap, this research investigates advanced control methodologies to enhance the performance of such motors. A MATLAB/Simulink framework was developed to model the motor, whose drive system leverages an AI-optimised dual fuzzy-PID controller refined using the Harmony Search Algorithm. Additionally, a combination of Indirect Field-Oriented Control (IFOC) and Space Vector PWM strategies are implemented to optimise inverter switching sequences for precise output modulation. Simulation results demonstrate significant improvements in torque response and control accuracy, validating the efficacy of the proposed system. The results highlight the role of AI-based propulsion systems in revolutionising submarine manoeuvrability and energy efficiency. In particular, during a test case involving a speed transition from 75 RPM to 900 RPM, the proposed AI-based controller achieves a near-zero overshoot compared to an initial control scheme that exhibits 75.89% overshoot. Full article

Forest trees have emerged as a promising passive solution for mitigating low-frequency ground vibrations generated by railway operations, offering ecological and cost-effective advantages. This study proposes a three-dimensional semi-analytical method developed for evaluating the dynamic responses of the coupled track–ground–tree system. The thin-layer method is employed to derive an explicit Green’s function corresponding to a har-monic point load acting on a layered half-space, which is subsequently applied to couple the foundation with the track system. The forest trees are modeled as surface oscillators coupled on the ground surface to evaluate the characteristics of multiple scattered wavefields. The vibration attenuation capacity of forest trees in mitigating railway-induced ground vibrations is systematically investigated using the proposed method. In the direction perpendicular to the track on the ground surface, a graded array of forest trees with varying heights is capable of forming a broad mitigation frequency band below 80 Hz. Due to the interaction of wave fields excited by harmonic point loads at multiple locations, the attenuation performance of the tree system varies significantly across different positions on the surface. The influence of variability in tree height, radius, and density on system performance is subsequently examined using a Monte Carlo simulation. Despite the inherent randomness in tree characteristics, the forest still demonstrates notable attenuation effectiveness at frequencies below 80 Hz. Among the considered parameters, variations in tree height exert the most pronounced effect on the uncertainty of attenuation performance, followed sequentially by variations in density and radius. Full article

Various factors influence the formation and adjustment of network freight prices, including transportation costs, cargo characteristics, and policies and regulations. The interaction of these factors increases the difficulty of accurately predicting network freight prices through regressions or other machine learning models, especially when the amount and quality of training data are limited. This paper introduces large language models (LLMs) to predict network freight prices using their inherent prior knowledge. Different data sorting methods and serialization strategies are employed to construct the corpora of LLMs, which are then tested on multiple base models. A few-shot sample dataset is constructed to test the performance of models under insufficient information. The Chain of Thought (CoT) is employed to construct a corpus that demonstrates the reasoning process in freight price prediction. Cross entropy loss with LoRA fine-tuning and cosine annealing learning rate adjustment, and Mean Absolute Error (MAE) loss with full fine-tuning and OneCycle learning rate adjustment to train the models, respectively, are used. The experimental results demonstrate that LLMs are better than or competitive with the best comparison model. Tests on a few-shot dataset demonstrate that LLMs outperform most comparison models in performance. This method provides a new reference for predicting network freight prices. Full article

Metal–organic frameworks (MOFs) have characteristics such as a large specific surface area, distinct functional sites, and an adjustable pore size. However, the inherent low conductivity of MOFs significantly affects the charge transfer efficiency when they are used for electrocatalytic sensing. Combining MOFs with conductive materials can compensate for these deficiencies. For MOF/metal nanoparticle composites (e.g., composites with gold, silver, platinum, and bimetallic nanoparticles), the high electrical conductivity and catalytic activity of metal nanoparticles are utilized, and MOFs can inhibit the agglomeration of nanoparticles. MOF/carbon-based material composites integrate the high electrical conductivity and large specific surface area of carbon-based materials. MOF/conductive polymer composites offer good flexibility and tunability. MOF/multiple conductive material composites exhibit synergistic effects. Although MOF composites provide an ideal platform for electrocatalytic reactions, current research still suffers from several issues, including a lack of comparative studies, insufficient research on structure–property correlations, limited practical applications, and high synthesis costs. In the future, it is necessary to explore new synthetic pathways and seek; inexpensive alternative raw materials. Full article

Accurate weld defect detection is critical for ensuring structural safety and evaluating welding quality in industrial applications. Manual inspection methods have inherent limitations, including inefficiency and inadequate sensitivity to subtle defects. Existing detection models, primarily designed for natural images, struggle to adapt to the characteristic challenges of weld X-ray images, such as high noise, low contrast, and inter-defect similarity, particularly leading to missed detections and false positives for small defects. To address these challenges, a multi-dimensional feature fusion model (MADet), which is a multi-branch deep fusion network for weld defect detection, was proposed. The framework incorporates two key innovations: (1) A multi-scale feature fusion network integrated with lightweight attention residual modules to enhance the perception of fine-grained defect features by leveraging low-level texture information. (2) An anchor-based feature-selective detection head was used to improve the discrimination and localization accuracy for five typical defect categories. Extensive experiments on both public and proprietary weld defect datasets demonstrated that MADet achieved significant improvements over the state-of-the-art YOLO variants. Specifically, it surpassed the suboptimal model by 7.41% in mAP@0.5, indicating strong industrial applicability. Full article