Search Results (22)

Flow past bluff bodies (e.g., circular cylinders) forms a canonical context for teaching external flow separation, vortex shedding, and the coupling between surface pressure and hydrodynamic forces in offshore engineering. Conventional laboratory implementations, however, often fragment local and global measurements, delay data feedback, and omit intelligent modeling components, thereby limiting the development of higher-order cognitive skills and data literacy. We present a low-cost, modular, data-enabled instructional hydrodynamics platform that integrates a transparent recirculating water channel, multi-point synchronous circumferential pressure measurements, global force acquisition, and an artificial neural network (ANN) surrogate. Using feature vectors composed of Reynolds number, angle of attack, and submergence depth, we train a lightweight AI model for rapid prediction of drag and lift coefficients, closing a loop of measurement, prediction, deviation diagnosis, and feature refinement. In the subcritical Reynolds regime, the measured circumferential pressure distribution for a circular cylinder and the drag and lift coefficients for a rectangular cylinder agree with empirical correlations and published benchmarks. The ANN surrogate attains a mean absolute percentage error of approximately 4% for both drag and lift coefficients, indicating stable, physically interpretable performance under limited feature inputs. This platform will facilitate students’ cross-domain transfer spanning flow physics mechanisms, signal processing, feature engineering, and model evaluation, thereby enhancing inquiry-driven and critical analytical competencies. Key contributions include the following: (i) a synchronized local pressure and global force dataset architecture; (ii) embedding a physics-interpretable lightweight ANN surrogate in a foundational hydrodynamics experiment; and (iii) an error-tracking, iteration-oriented instructional workflow. The platform provides a replicable pathway for transitioning offshore hydrodynamics laboratories toward an integrated intelligence-plus-data literacy paradigm and establishes a foundation for future extensions to higher Reynolds numbers, multiple body geometries, and physics-constrained neural networks. Full article

The pandemic forced educators to shift abruptly to distance learning, also referred to as e-learning education. Educational institutions integrated new educational tools and online platforms. Several schools, colleges, and universities began incorporating online laboratories in different fields of education, such as engineering, information technology, physics, and chemistry. Online laboratories may take the form of virtual laboratories, software-based simulations available via the Internet, or remote labs, which involve accessing physical equipment online. Adopting remote laboratories as a substitute for conventional hands-on labs has raised concerns regarding the safety and security of both the remote lab stations and the Online Laboratory Management Systems (OLMSs). Design patterns and architectures need to be developed to attain security by design in remote laboratories. Before these can be developed, software architects and developers must understand the domain and existing and proposed solutions. This paper presents an extensive literature review of safety and security concerns related to remote laboratories and an overview of the industry, national and multinational standards, and legal requirements and regulations that need to be considered in building secure and safe Online Laboratory Management Systems. This analysis provides a taxonomy and classification of published standards as well as security and safety problems and possible solutions that can facilitate the documentation of best practices, and implemented solutions to produce security by design for remote laboratories and OLMSs. Full article

Facial occlusion represents a significant challenge in the domain of facial expression recognition (FER). The absence of feature information due to occlusion has been demonstrated to result in a reduction in recognition accuracy and model robustness. To address this challenge, a multi-angle feature extraction (MAFE) method is proposed in this paper, aiming to enhance the recognition accuracy under occlusion conditions by employing multi-scale global features, local fine-grained features, and important regional features. The MAFE approach involves three core modules: multi-feature extraction, regional detail feature fusion, and consistent feature recognition. In the multi-feature extraction module, PTIR-50 and Swin Transformer are used to extract global features and fine-grained features, and at the same time, the five key points of the face are combined to crop out the important regions from the global features. The Regional Bias Loss (RB-Loss) is then utilized to guide the model to focus on the key information regions. The subsequent Regional Detail Feature Fusion module combines fine-grained features with those from the important regions. This process enhances the expressiveness of the features. The Consistent Feature Recognition module proposes consistent feature loss (con-feature Loss) to ensure that global features and fused features guide each other, forcing the model to focus on more discriminative expression features. The experimental results demonstrate that MAFE attains 89.42% and 86.94% accuracies on the Occlusion-RAFDB and Occlusion-FERPlus datasets, thereby surpassing the existing methods. Accuracies of 92.11% and 90.15% are also obtained on the original RAF-DB and FERPlus datasets. Full article

Multi-source domain adaptation (MSDA) in remote sensing (RS) scene classification has recently gained significant attention in the visual recognition community. It leverages multiple well-labeled source domains to train a model capable of achieving strong generalization on the target domain with little to no labeled data from the target domain. However, the distribution shifts among multiple source domains make it more challenging to align the distributions between the target domain and all source domains concurrently. Moreover, relying solely on global alignment risks losing fine-grained information for each class, especially in the task of RS scene classification. To alleviate these issues, we present a Multi-Source Subdomain Distribution Alignment Network (MSSDANet), which introduces novel network structures and loss functions for subdomain-oriented MSDA. By adopting a two-level feature extraction strategy, this model attains better global alignment between the target domain and multiple source domains, as well as alignment at the subdomain level. First, it includes a pre-trained convolutional neural network (CNN) as a common feature extractor to fully exploit the shared invariant features across one target and multiple source domains. Secondly, a dual-domain feature extractor is used after the common feature extractor, which maps the data from each pair of target and source domains to a specific dual-domain feature space and performs subdomain alignment. Finally, a dual-domain feature classifier is employed to make predictions by averaging the outputs from multiple classifiers. Accompanied by the above network, two novel loss functions are proposed to boost the classification performance. Discriminant Semantic Transfer (DST) loss is exploited to force the model to effectively extract semantic information among target and source domain samples, while Class Correlation (CC) loss is introduced to reduce the feature confusion from different classes within the target domain. It is noteworthy that our MSSDANet is developed in an unsupervised manner for domain adaptation, indicating that no label information from the target domain is required during training. Extensive experiments on four common RS image datasets demonstrate that the proposed method achieves state-of-the-art performance for cross-domain RS scene classification. Specifically, in the dual-source and three-source settings, MSSDANet outperforms the second-best algorithm in terms of overall accuracy (OA) by 2.2% and 1.6%, respectively. Full article

Optical filters are essential components for a variety of applicative fields, such as communications, chemical analysis and optical signal processing. This article describes the preparation and characterization of a new optical filter made of polyvinyl alcohol and incremental amounts of crystal violet. By using distinct solvents (H₂O, dimethyl sulfoxide (DMSO) and H₂O₂) to obtain the dyed polymer films, new insights were gained into the pathway that underlies the possibility of tailoring the material’s optical performance. The effect of the dye content on the sample’s main properties was inspected via UV–VIS spectroscopy analysis combined with colorimetry, refractometry and atomic force microscopy experiments. The results revealed that the colorimetric parameters are affected by the dye amount and are dramatically changed when the solvent used for film preparation is different. The rise in the refractive index upon polymer dyeing was due to the synergistic effect of the larger polarizability of the dye and the occurrence of hydrogen bonds among the system components. Spectral data evidenced that samples prepared in H₂O and DMSO preserve the absorption characteristics of the added dye, whereas H₂O₂ acts as an oxidizing agent and enhances transparency. Also, for the first two solvents, multiple absorption edges were noted as a result of dye incorporation, which was responsible for the occurrence of new exciton-like states, hence the band gap reduction. The films processed in H₂O were able to block radiations in the 506–633 nm range while allowing other wavelengths to pass with a transmittance above 90%. The samples attained in DMSO presented similar properties, with the difference that the domain of light attenuation was shifted towards higher wavelengths. Atomic force microscopy showed the dye’s effect on the level of surface roughness uniformity and morphology isotropy. The dyed polymer foils in non-oxidizing agents have suitable features for use as band-pass filters. Full article

This paper analyses the socio-economic disadvantages of women from different ethnic minority heritages in the UK. Using data from the Labour Force Survey (2014–2023), which contains detailed information on parental class and respondents’ socio-economic conditions, we examine four domains of life chances which are crucial for ethnic integration: educational attainment at the degree level, risks of unemployment, access to professional-managerial (salariat) position and earning power. We proceeded with the gross differences and then examined the differences by ethno-gender status and parental class combinations, controlling for many confounding factors. We also examined the net ethno-gender differences over the life course and the trends of social fluidity over the period covered and across the ethno-gender groups. We found that women from all ethnic origins were doing well in education but faced multiple disadvantages in the labour market, especially in access to the salariat and in earning power. Women of Pakistani/Bangladeshi heritages faced pronounced unemployment risks, especially at the earlier life stages. There is a significant increase in fluidity over the period covered, but this is marked by considerable ethnic and class differences, with Black Caribbean, Black African, Pakistani and Bangladeshi women from more advantaged class origins being unable to secure advantaged class positions and those from working-class families unable to make long-range upward mobility as effectively as White men. Overall, Bangladeshi, Pakistani, Black African and Black Caribbean women are found to be considerably disadvantaged, but there are also signs of social progress. Full article

Four-wheel independent drive and four-wheel independent steering (4WID-4WIS) vehicles provide increased redundancy in fault-tolerant control (FTC) schemes, enhancing heterogeneous fault-tolerant capabilities. This paper addresses the challenge of maintaining vehicle safety and maneuverability in the presence of actuator faults in autonomous vehicles, focusing on 4WID-4WIS systems. A novel unified hierarchical active FTC strategy is proposed to handle various actuator failures. The strategy includes an upper-layer motion controller that determines resultant force requirements based on trajectory tracking errors and a middle-layer allocation system that redistributes tire forces to fault-free actuators using fault information. This study, for the first time, considers multi-fault scenarios involving longitudinal and lateral coupling, calculating FTC boundaries for each fault type. Additionally, a fault tolerance index is introduced for 256 fault scenarios, using singular value decomposition to linearly represent the vehicle attainable force domain. Based on this, an adaptive velocity planning strategy is developed to balance safety and maneuverability under fault conditions. Matlab 2021a/Simulink and Carsim 2019 co-simulation results validate the proposed strategies, demonstrating significant improvements in fault-tolerant performance, particularly in complex and emergency scenarios. Full article

This paper is concerned with the mathematical modeling of complex living systems whose element microscopic state contains variables which can attain discrete values. Specifically, the main mathematical frameworks of the discrete thermostatted kinetic theory for active particles are reviewed and generalized. In the generalized thermostatted frameworks, which are based on nonlinear ordinary or partial differential equations, the elements of the system are viewed as active particles that are able to perform certain strategies modeled by introducing a functional-state variable called activity. Interactions, which are responsible of the evolution of the system, are modeled using the fundamentals of stochastic game theory and may be influenced by the action of an external force field coupled to a Gaussian-type thermostat. In particular, the interaction domain is modeled by introducing a weighted function and different non-homogeneous discrete frameworks are proposed and coupled with a specific thermostat. Two recent models derived within this approach are reviewed and refer to vehicular and pedestrian dynamics. Future research perspectives are discussed in the whole paper from theoretical and modeling viewpoints. Full article

The proliferation of classification-capable artificial intelligence (AI) across a wide range of domains (e.g., agriculture, construction, etc.) has been allowed to optimize and complement several tasks, typically operationalized by humans. The computational training that allows providing such support is frequently hindered by various challenges related to datasets, including the scarcity of examples and imbalanced class distributions, which have detrimental effects on the production of accurate models. For a proper approach to these challenges, strategies smarter than the traditional brute force-based K-fold cross-validation or the naivety of hold-out are required, with the following main goals in mind: (1) carrying out one-shot, close-to-optimal data arrangements, accelerating conventional training optimization; and (2) aiming at maximizing the capacity of inference models to its fullest extent while relieving computational burden. To that end, in this paper, two image-based feature-aware dataset splitting approaches are proposed, hypothesizing a contribution towards attaining classification models that are closer to their full inference potential. Both rely on strategic image harvesting: while one of them hinges on weighted random selection out of a feature-based clusters set, the other involves a balanced picking process from a sorted list that stores data features’ distances to the centroid of a whole feature space. Comparative tests on datasets related to grapevine leaves phenotyping and bridge defects showcase promising results, highlighting a viable alternative to K-fold cross-validation and hold-out methods. Full article

The routine use of prosthetic hands significantly enhances amputees’ daily lives, yet it often introduces cognitive load and reduces reaction speed. To address this issue, we introduce a wearable semi-autonomous hierarchical control framework tailored for amputees. Drawing inspiration from the visual processing stream in humans, a fully autonomous bionic controller is integrated into the prosthetic hand control system to offload cognitive burden, complemented by a Human-in-the-Loop (HIL) control method. In the ventral-stream phase, the controller integrates multi-modal information from the user’s hand–eye coordination and biological instincts to analyze the user’s movement intention and manipulate primitive switches in the variable domain of view. Transitioning to the dorsal-stream phase, precise force control is attained through the HIL control strategy, combining feedback from the prosthetic hand’s sensors and the user’s electromyographic (EMG) signals. The effectiveness of the proposed interface is demonstrated by the experimental results. Our approach presents a more effective method of interaction between a robotic control system and the human. Full article

Fused filament fabrication (FFF), colloquially known as 3D-printing, has gradually expanded from the laboratory to the industrial and household realms due to its suitability for producing highly customized products with complex geometries. However, it is difficult to evaluate the mechanical performance of samples produced by this method of additive manufacturing (AM) due to the high number of combinations of printing parameters, which have been shown to significantly impact the final structural integrity of the part. This implies that using experimental data attained through destructive testing is not always viable. In this study, predictive models based on the rapid prediction of the required extrusion force and mechanical properties of printed parts are proposed, selecting a subset of the most representative printing parameters during the printing process as the domain of interest. Data obtained from the in-line sensor-equipped 3D printers were used to train several different predictive models. By comparing the coefficient of determination (R²) of the response surface method (RSM) and five different machine learning models, it is found that the support vector regressor (SVR) has the best performance in this data volume case. Ultimately, the ML resources developed in this work can potentially support the application of AM technology in the assessment of part structural integrity through simulation and can also be integrated into a control loop that can pause or even correct a failing print if the expected filament force-speed pairing is trailing outside a tolerance zone stemming from ML predictions. Full article

Utilizing turbidimetry data, an examination is conducted on the behavior of solutions containing N-vinylpyrrolidone and vinyl propyl ether copolymer within a temperature range coinciding with the occurrence of a phase transition. The investigation reveals that within specific conditions prevailing in this domain, the emergence of entities denoted as hydrophobic–hydrophilic associates is conceivable. These entities are characterized by the presence of a relatively dense core, upheld by hydrophobic interplays, and they are proficient in effectively dispersing irradiation within the optical spectrum. Encircling this core is a hydrophilic periphery that impedes the formation of insoluble precipitates. The development of such associates transpires when hydrophobic interactions have attained a discernible prominence, although they remain inadequate to counteract the forces that drive the expansion of macromolecular coils. Under these circumstances, the energetically favored course of action entails the constitution of a core for the aforementioned associates, involving discrete segments from diverse macromolecules. Notably, the introduction of an additional constituent (ethanol) to the solution, which selectively mitigates hydrophobic interactions, serves to stabilize the hydrophobic–hydrophilic associations. Full article

The development of innovative materials with improved properties is required for the field of energy storage. This article proves that it is possible to utilize bio-derived fillers to tune the performance of biodegradable polymers. For this scope, eco-composites were attained by loading several amounts of walnut leaf powder (WLP) in hydroxyethylcellulose (HEC). Basic testing was conducted to emphasize the sample’s suitability for the pursued application. The rheological behavior was altered with the addition of WLP at low shear rates, which became more pseudoplastic, resulting in composite films with higher thickness uniformity. Wettability characteristics were used to analyze the macro-level adhesion of the platinum-containing samples, and the results showed that the presence of WLP led to the augmentation of interfacial compatibilization of the composite with the metal layer. The electron microscopy and atomic force microscopy scans showed the proper distribution of the WLP in the matrix. Local adhesion data derived from DFL-height curves further showed that the inclusion of WLP improves the adhesion capabilities at the nanoscale. The dielectric spectroscopy tests proved that the used biofiller leads to an enhancement in the permittivity of the composite with respect to the neat HEC. By accounting for all results, the generated eco-composites are suggested as alternative dielectrics for usage in the energy storage domain. Full article

This study aimed to apply an incremental tethered swimming test (ITT) with workloads (WL) based on individual rates of front crawl mean tethered force (Fmean) for the identification of the upper boundary of heavy exercise (by means of respiratory compensation point, RCP), and therefore to describe oxygen uptake kinetics (VO₂k) and time limit (t_Lim) responses to WL corresponding to peak oxygen uptake (WLVO_2peak). Sixteen swimmers of both sexes (17.6 ± 3.8 years old, 175.8 ± 9.2 cm, and 68.5 ± 10.6 kg) performed the ITT until exhaustion, attached to a weight-bearing pulley–rope system for the measurements of gas exchange threshold (GET), RCP, and VO_2peak. The WL was increased by 5% from 30 to 70% of Fmean at every minute, with Fmean being measured by a load cell attached to the swimmers during an all-out 30 s front crawl bout. The pulmonary gas exchange was sampled breath by breath, and the mathematical description of VO₂k used a first-order exponential with time delay (TD) on the average of two rest-to-work transitions at WLVO_2peak. The mean VO_2peak approached 50.2 ± 6.2 mL·kg⁻¹·min⁻¹ and GET and RCP attained (respectively) 67.4 ± 7.3% and 87.4 ± 3.4% VO_2peak. The average t_Lim was 329.5 ± 63.6 s for both sexes, and all swimmers attained VO_2peak (100.4 ± 3.8%) when considering the primary response of VO₂ (A_1′ = 91.8 ± 6.7%VO_2peak) associated with the VO₂ slow component (SC) of 10.7 ± 6.7% of end-exercise VO₂, with time constants of 24.4 ± 9.8 s for A_1′ and 149.3 ± 29.1 s for SC. Negative correlations were observed for t_Lim to VO_2peak, WLVO_2peak, GET, RCP, and EEVO₂ (r = −0.55, −0.59, −0.58, −0.53, and −0.50). Thus, the VO₂k during tethered swimming at WLVO_2peak reproduced the physiological responses corresponding to a severe domain. The findings also demonstrated that t_Lim was inversely related to aerobic conditioning indexes and to the ability to adjust oxidative metabolism to match target VO₂ demand during exercise. Full article

Molecular dynamics simulations of protein folding typically consider the polypeptide chain at equilibrium and in isolation from the cellular components. We argue that in order to understand protein folding as it occurs in vivo, it should be modeled as an active, energy-dependent process, in which the cellular protein-folding machine directly manipulates the polypeptide. We conducted all-atom molecular dynamics simulations of four protein domains, whose folding from the extended state was augmented by the application of rotational force to the C-terminal amino acid, while the movement of the N-terminal amino acid was restrained. We have shown earlier that such a simple manipulation of peptide backbone facilitated the formation of native structures in diverse α-helical peptides. In this study, the simulation protocol was modified, to apply the backbone rotation and movement restriction only for a short time at the start of simulation. This transient application of a mechanical force to the peptide is sufficient to accelerate, by at least an order of magnitude, the folding of four protein domains from different structural classes to their native or native-like conformations. Our in silico experiments show that a compact stable fold may be attained more readily when the motions of the polypeptide are biased by external forces and constraints. Full article