Search Results (120)

Electroencephalogram (EEG) signals serve as the core information carrier for brain–computer interfaces (BCIs); however, their highly non-stationary nature, extremely low signal-to-noise ratio, and significant inter-individual variability pose considerable challenges for signal decoding. Existing deep learning methods struggle to strike a balance between multi-scale, fine-grained feature extraction and efficient long-range temporal modeling. To overcome this limitation, this study proposes a novel deep cascaded architecture, MSResBiMamba, which deeply integrates multi-scale spatiotemporal feature learning with cutting-edge long-sequence modeling techniques. The model first utilizes an enhanced multi-scale spatiotemporal convolutional network (MS-CNN) combined with a SE-channel attention mechanism to adaptively extract local multi-band features and dynamically suppress redundant artefacts. Subsequently, it innovatively introduces an enhanced bidirectional Mamba (Bi-Mamba) module to efficiently capture non-causal long-range temporal dependencies with linear computational complexity, whilst cascading multi-head self-attention mechanisms to establish global higher-order feature interactions. Extensive experiments on the BCI Competition IV-2a dataset demonstrate that MSResBiMamba achieves outstanding classification performance in multi-class motor imagery tasks, significantly outperforming traditional methods and existing state-of-the-art neural networks. Ablation studies and t-SNE visualisations further confirm the model’s robustness in feature decoupling and cross-subject applications, providing a high-precision, high-efficiency decoding solution for BCI systems. Full article

Sharpness and cutting edge retention are critical performance metrics for kitchen knives. Their combined effectiveness is governed by the synergistic effects of edge geometry and material microstructure. The present study selected six representative knife steels, namely 3Cr13, 1.4116, 9Cr18MoV, T10, GCr15, and CPM 3V, to fabricate the experimental knives with edge inclusive angles of 18°, 24°, and 30°. Standardized CATRA cutting tests were conducted to evaluate the effects of material microstructure and edge geometry on initial cutting performance (ICP) and total card cut (TCC), serving as the direct metrics for sharpness and cutting edge retention, respectively. The underlying mechanisms responsible for the cutting behavior were elucidated through scanning electron microscopy, quantitative analysis of carbides, and measurements of edge wear volume. The roles of carbide number, size, and morphology in ICP and TCC were systematically analyzed. Furthermore, multivariate linear regression models were established to quantitatively correlate ICP and TCC with edge inclusive angle, material hardness, average carbide diameter, and edge width. The results indicate that the edge inclusive angle predominantly determines ICP, while TCC is primarily controlled by the synergistic interaction between carbide characteristics and matrix hardness. Although a smaller edge inclusive angle significantly enhances ICP, it also accelerates edge wear and reduces cutting efficiency. By comprehensively considering both ICP and TCC, an optimal edge inclusive angle range was identified for each material to achieve balanced cutting performance. This work provides experimental evidence and quantitative guidance for the material selection and edge geometry design of high-performance kitchen knives. Full article

The synthesis of continuous non-linear metal nanostructures at the micro and nanoscale remains a challenging frontier in nanotechnology due to inherent synthetic constraints. This study introduces an innovative chemical methodology for fabricating continuous rings and diverse geometries via the chemical fusion of gold nanorods (AuNRs) on a solid substrate. Initially, aqueous solutions of cetyltrimethylammonium bromide (CTAB)-coated AuNRs were deposited and dried on a solid substrate, resulting in the self-assembly of ring-like arrays. Subsequent chemical growth of the AuNRs in all dimensions was achieved using an aqueous solution of Au(I)/CTAB/Ascorbic Acid (AA), enabling their fusion into continuous structures. This approach permits the formation of arbitrary shapes by pre-arranging AuNRs, thereby opening new avenues for the exploration of non-linear nanostructures with potentially novel plasmonic and electronic properties. The capability to engineer such complex nanostructures is pivotal for advancing fields such as photonics, electronics, and sensing, where the unique optical and electronic properties of gold nanostructures can be exploited for cutting-edge applications. Furthermore, this technique shows a significant promise for the fabrication of various micro- and nanodevices and the seamless interconnection of components in integrated electronic circuits, potentially leading to more efficient and miniaturized electronic systems. The broader implications of this research are significant, offering a potential pathway to the development of nanomaterials and devices that could benefit various industries and technological processes. Full article

To match the fluctuating renewable energy generation with varying industrial process steam demand, electric heat storage is a viable solution for energy transformation. However, the adoption of this technology by companies hinges on its economic feasibility. Proponents often argue that leveraging spot market electricity prices provides a competitive edge over conventional energy systems without storage. However, additional factors, such as grid fees, levies and taxes can substantially inflate storage charging costs. This study investigates the integration of an electric latent heat storage for process steam within a conventional natural gas boiler system in a paper manufacturing case study under German industrial electricity price regulation to evaluate the effects of electricity prices alongside various operational energy cost factors and rebates through a mixed-integer linear program. Three storage sizes (1 MWh, 100 MWh, 100 GWh) and four electricity procurement configurations (status quo, optimized fixed-price contract, day-ahead only, and a mixed fixed/day-ahead strategy) are analysed. For 100 MWh, the mixed strategy cuts energy costs by about 10% versus day-ahead pricing and around 13.5% versus the status quo contract, lowering the specific electricity price from 6.2 to 5.3 ct/kWh, while emissions range from the slight increases to 23% reductions when grid fees are removed. Full article

Advances in CMOS image sensors (CISs) have led to utilization in various industrial fields, including machine vision, medical, surveillance, the automotive industry, and the Internet of Things (IoT). One critical metric for CISs is the dynamic range (DR), which indicates the range of light intensity that can clearly capture images. As the technology evolves, wide dynamic range (WDR) becomes increasingly required for more diverse applications. To further advance these industries, this paper presents the active-pixel-type CIS design techniques and architectures developed to achieve WDR. These include the following: the basic concepts of the active pixel sensor, readout mechanism, and DR of the CIS; multiple exposure and dual conversion gain (DCG) schemes that are conventionally used to address a trade-off in the CIS; lateral overflow integration capacitor (LOFIC) and dual photodiode (PD) architectures that can improve the DR by utilizing trade-offs in the DR and exposure mechanism; CISs with logarithmic and linear–logarithmic (Lin-Log) responses to enable non-linear characteristics; and techniques that can be employed for higher sensitivity in dark conditions. This comprehensive study of various techniques and architectures can also be utilized for cutting-edge tech advances and future research, including neuromorphic array architecture. Full article

Background: The vehicle routing problem (VRP) is of great importance in the Industry 4.0 era because enabling technologies such as the internet of things (IoT), artificial intelligence (AI), big data, and geographic information systems (GISs) allows for real-time solutions to versions of the problem, adapting to changing conditions such as traffic or fluctuating demand. Methods: In this paper, we model and optimize a classic multi-link distribution network topology, including randomness in travel times, vehicle availability times, and product demands, using a hybrid approach of nested linear stochastic programming and Monte Carlo simulation under a time-window scheme. The proposed solution is compared with cutting-edge metaheuristics such as Ant Colony Optimization (ACO), Tabu Search (TS), and Simulated Annealing (SA). Results: The results suggest that the proposed method is computationally efficient and scalable to large models, although convergence and accuracy are strongly influenced by the probability distributions used. Conclusions: The developed proposal constitutes a viable alternative for solving real-world, large-scale modeling cases for transportation management in the supply chain. Full article

Beamforming is highly significant for the physical layer of wireless communication systems, for multi-antenna systems such as multiple input multiple output (MIMO) and massive MIMO, since it improves spectral efficiency and reduces interference. Traditional linear beamforming methods such as zero-forcing beamforming (ZFBF) and minimum mean square error (MMSE) beamforming provide closed-form solutions. Yet, their performance drops when they face non-ideal conditions such as imperfect channel state information (CSI), dynamic propagation environment, or high-dimensional system configurations, primarily due to static assumptions and computational limitations. These limitations have led to the rise of deep learning-based beamforming, where data-driven models derive beamforming solutions directly from CSI. By leveraging the representational capabilities of cutting-edge deep learning architectures, along with the increasing availability of data and computational resources, deep learning presents an adaptive and potentially scalable alternative to traditional methodologies. In this work, we unify and systematically compare our two unsupervised learning architectures for uplink receive beamforming: a simple neural network beamforming (NNBF) model, composed of convolutional and fully connected layers, and a transformer-based NNBF model that integrates grouped convolutions for feature extraction and transformer blocks to capture long-range channel dependencies. They are evaluated in a common multi-user single input multiple output (MU-SIMO) system model to maximize sum-rate across single-antenna user equipments (UEs) under 3GPP-compliant channel models, namely TDL-A and UMa. Furthermore, we present a FLOPs-based asymptotic computational complexity analysis for the NNBF architectures alongside baseline methods, namely ZFBF and MMSE beamforming, explicitly characterizing inference-time scaling behavior. Experiments for the simple NNBF are performed under simplified assumptions such as stationary UEs and perfect CSI across varying antenna configurations in the TDL-A channel. On the other hand, transformer-based NNBF is evaluated in more realistic conditions, including urban macro environments with imperfect CSI, diverse UE mobilities, coding rates, and modulation schemes. Results show that the transformer-based NNBF achieves superior performance under realistic conditions at the cost of increased computational complexity, while the simple NNBF presents comparable or better performance than baseline methods with significantly lower complexity under simplified assumptions. Full article

Islamic Microfinance Institutions (IMFIs) encounter distinct difficulties with credit scoring because they need to follow Shariah principles that combine riba bans with fair financial dealings regulations. Conventional credit scoring models exhibit two shortcomings: a poor capability to incorporate non-financial behavioral data and inadequate support for Islamic Microfinance Institutions’ requirements. Researchers use machine learning coupled with blockchain technology to create an adaptive Shariah-compliant credit scoring method that solves problems found in standard evaluation systems. Using a dataset of 1275 farmers with 52 weeks of transaction data, we implemented and compared three ML models: Linear Regression, Random Forest, and Gradient Boosting. Data preparation involved addressing 53% missing transaction data, followed by summing weekly financial activity to prepare it for predictive evaluations. Our analysis shows that the Random Forest model produced the best results with an R-squared value of 0.87 and a Mean Squared Error (MSE) of 12.4. In creditworthiness binary classification tasks, Gradient Boosting delivered an F1 score of 0.91 while maintaining precision at 0.89 and recall at 0.93. Blockchain integration exists to protect data through secure mechanisms that also conserve Islamic financial integrity and promote transparency. The research shows how ML and Blockchain technology enable fundamental changes in IMFIs by delivering elevated predictive accuracy, operational enhancements, and complete transparency. The conceptual framework guides ethical financial inclusion strategy by offering a solution for marginalized communities, but remains consistent with global sustainability objectives. The research established foundational elements for implementing cutting-edge technologies within IMFIs, which will promote new economic growth and build confidence in Shariah-compliant financial systems. Full article

Additive and coating technologies, such as high-velocity oxy-fuel (HVOF) thermal spraying and direct metal laser sintering (DMLS), often require extensive post-processing to meet dimensional and surface quality requirements, which remains challenging for nickel-based superalloys such as Inconel 718. This study presents the design and topology optimisation of a cutting tool with a linear cutting edge, capable of operating in turn-milling or turning modes, offering a viable alternative to conventional grinding. A non-optimised tool served as a baseline for comparison with a topology-optimised variant improving cutting-force distribution and stiffness-to-mass ratio. Finite element analyses and experimental turn-milling trials were performed on DMLS and HVOF Inconel 718 using carbide and CBN inserts. The optimised tool achieved significantly reduced roughness values: for DMLS, Ra decreased from 0.514 ± 0.069 µm to 0.351 ± 0.047 µm, and for HVOF from 0.606 ± 0.069 µm to 0.407 ± 0.069 µm. Rz was similarly improved, decreasing from 4.234 ± 0.343 µm to 3.340 ± 0.439 µm (DMLS) and from 5.349 ± 0.552 µm to 4.521 ± 0.650 µm (HVOF). The lowest measured Ra, 0.146 ± 0.030 µm, was obtained using CBN inserts at the highest tested cutting speed. All improvements were statistically significant (p < 0.005). No measurable tool wear was observed due to the small engagement and the use of a fresh cutting edge for each pass. The resulting surface quality was comparable to grinding and clearly superior to conventional turning. These findings demonstrate that combining topology optimisation with a linear-edge tool provides a practical and efficient finishing approach for additively manufactured and thermally sprayed Inconel 718 components. Full article

Background/Objectives: Heart disease (HD) is recently becoming the foremost cause of death worldwide, underlining the importance of early and correct diagnosis to improve patient outcomes. Although Internet of Things (IoT)-enabled machine learning approaches have demonstrated encouraging outcomes in screening, existing approaches often face challenges such as imbalanced dataset handling, influential feature selection identification, and the ability to adapt to evolving HD data forms. To tackle the aforementioned challenges, we present a sequential hybrid approach, RAVE-HD (ResNet And Vanilla RNN Ensemble for HD), that combines a number of cutting-edge techniques to enhance screening. Methods: Preprocessing phase includes duplicates removal and feature scaling for data consistency. Recursive Feature Elimination is employed to extract the most informative features, while a proximity-weighted random synthetic sampling technique addresses class imbalance to reduce class biases. The proposed RAVE model in RAVE-HD approach sequentially integrates a Residual Network (ResNet) for high-level feature extraction and Vanilla Recurrent Neural Network to capture the non-linearity of the feature relationships present in the HDHI medical dataset. Results: Compared to ResNet and Vanilla RNN baselines, the proposed RAVE model attained superior results: 92.06% accuracy and 97.12% ROC-AUC. Stratified 10-fold cross-validation validated the robustness of RAVE, while Sensitivity-to-Prevalence analysis demonstrated stable recall and predictable precision across varying disease prevalence levels. Additional evaluations, including bootstrap and DeLong analyses, showed statistical significance $(p<0.001)$ of the discriminative gains of RAVE. Minimum Clinically Important Difference (MCID) evaluation confirmed clinically meaningful improvements $(\ge$$3\%)$ over strong baselines. Cross-dataset validation using the CVD dataset verified robust generalization (92.4% accuracy). SHAP analysis provided interpretability to build clinical trust. Conclusions: RAVE-HD shows promise as a reliable, explainable, and scalable solution for large-scale HD screening, consistently performing well across diverse evaluations and datasets. Through statistical validation, the RAVE-HD approach emerges as a practical decision-support tool in HD predictive screening results. Full article

Femoroacetabular impingement (FAI) and hip dysplasia have been shown to increase the risk of hip osteoarthritis in affected individuals. MRI with T2 mapping provides an objective measure of femoral and acetabular articular cartilage tissue quality. This study aims to evaluate the relationship between hip morphology measurements collected from three-dimensional (3D) reconstructed computed tomography (CT) scans and the T2 mapping values of hip articular cartilage assessed by three independent, blinded reviewers on the optimal sagittal cut. Hip morphology measures including lateral center edge angle (LCEA), acetabular version, Tönnis angle, acetabular coverage, alpha angle, femoral torsion, neck-shaft angle (FNSA), and combined version were recorded from preoperative CT scans. The relationship between T2 values and hip morphology was assessed using univariate linear mixed models with random effects for individual patients. Significant associations were observed between femoral and acetabular articular cartilage T2 values and all hip morphology measures except femoral torsion. Hip morphology measurements consistent with dysplastic anatomy including decreased LCEA, increased Tönnis angle, and decreased acetabular coverage were associated with increased cartilage damage (p < 0.001 for all). Articular cartilage T2 values were strongly associated with the radiographic markers of hip dysplasia, suggesting hip microinstability significantly contributes to cartilage damage. The relationships between hip morphology measurements and T2 values were similar for the femoral and acetabular sides, indicating that damage to both surfaces is comparable rather than preferentially affecting one side. Full article

The advancement of smart grid technologies necessitates the integration of cutting-edge computational methods to enhance predictive energy optimization. This study proposes a multi-faceted approach by incorporating (1) Deep Reinforcement Learning (DRL) agents trained using data from digital twins (DTs) to optimize energy consumption in real time, (2) Physics-Informed Neural Networks (PINNs) to seamlessly embed physical laws within the optimization process, ensuring model accuracy and interpretability, and (3) blockchain (BC) technology to facilitate secure and transparent communication across the smart grid infrastructure. The model was trained and validated using comprehensive datasets, including smart meter energy consumption data, renewable energy outputs, dynamic pricing, and user preferences collected from IoT devices. The proposed framework achieved superior predictive performance with a Mean Absolute Error (MAE) of 0.237 kWh, Root Mean Square Error (RMSE) of 0.298 kWh, and an R-squared (R²) value of 0.978, indicating a 97.8% explanation of data variance. Classification metrics further demonstrated the model’s robustness, achieving 97.7% accuracy, 97.8% precision, 97.6% recall, and an F1 Score of 97.7%. Comparative analysis with traditional models like Linear Regression, Random Forest, SVM, LSTM, and XGBoost revealed the superior accuracy and real-time adaptability of the proposed method. In addition to enhancing energy efficiency, the model reduced energy costs by 35%, maintained a 96% user comfort index, and increased renewable energy utilization to 40%. This study demonstrates the transformative potential of integrating PINNs, DT, and blockchain technologies to optimize energy consumption in smart grids, paving the way for sustainable, secure, and efficient energy management systems. Full article

Tool wear detection is very important in CNC machine tool cutting. Once the tool is excessively worn, it is not only easy to cause the workpiece to be scrapped, but even to damage the machine. Therefore, common external turning tools of CNC machines are studied. The effect of tool nose wear on machining accuracy was analyzed by a building mathematical model. According to different wear conditions, a linear detection method based on edge images and input features was proposed to detect the main and secondary cutting edges, which helped determine the theoretical center of the tool nose and build a morphological visual model. For different error cases, the axial and radial error compensation strategies were proposed, respectively. By comparing the experimental data of four kinds of workpieces before and after compensation machining, the average errors of them were reduced separately, and the maximum value reached 79.2%, which verified the effectiveness of the compensation strategy. The intelligent compensation strategies will significantly improve the micro-machining accuracy and efficiency of the external turning tools in CNC machines. Full article

Intelligent ventilation systems can optimize airflow regulation to enhance mining safety and reduce energy consumption, driving green development in mineral resource extraction. This paper systematically elaborates on the overall architecture, cutting-edge advances, and core technologies of current intelligent mining ventilation. Building upon this foundation, a comprehensive intelligent mine ventilation solution encompassing the entire process of ventilation design, optimization, and operation is constructed based on a five-layer architecture, integrating key technologies such as intelligent sensing, real-time solving, airflow regulation, and remote control, providing an overarching framework for smart mine ventilation development. To address the computational efficiency bottleneck of traditional methods, an improved loop-solving method based on minimal independent closed loops is realized, achieving near real-time analysis of ventilation networks. Furthermore, a multi-level airflow regulation strategy is realized, including the methods of optimization control based on mixed integer linear programming and equipment-driven demand-based regulation, effectively resolving the challenges of calculating nonlinear programming models. Case studies indicate that the intelligent ventilation system significantly enhances mine safety and efficiency, leading to approximately 10–20% energy saving, a 40–60% quicker emergency response, and an average increase of about 20% in the utilization of fresh air at working faces through its remote and real-time control capabilities. Full article

In many practical scientific studies, two-way analysis of variance (ANOVA) and linear response surface methods are used for determining whether two or more similar compounds (substances, agents, or drugs) interact synergistically, antagonistically or independently. These models are often found lacking both in the means to assess interaction and with sufficient power. Using ten judiciously chosen illustrations (spanning fields as diverse as aquatic and environmental toxicology, botany, entomology, oncology, pharmacology, and virology), this paper introduces, explores, quantifies, illustrates, interprets, and extends several cutting-edge nonlinear assessment models and methods for measuring and describing interaction. Developed and used here are the Finney and Separate Ray models and extensions, a new cut line approach useful for so-called web designs, and extensions to more than two substances. As noted in the provided examples, the Finney model and extensions have the advantage of characterizing nonlinear interaction in a single measure, whereas the Separate Ray model extension is required when nonlinear interaction requires more than a single parameter for interaction assessment. The interaction results of the ten illustrations—of which antagonism is observed in four examples, synergy in five examples, and mixed results in one example—are summarized below. A discussion is also provided of efficient experimental design strategies as an aid to the practitioner and their future scientific studies. Full article