Optimal planning of power distribution networks with fault-tolerant configuration

To address the insufficient prediction accuracy of multi-state parameters in electro-hydraulic servo material fatigue testing machines under complex loading and nonlinear coupling conditions, this paper proposes a multivariate sequence-to-sequence prediction model integrating a Long Short-Term Memory (LSTM) encoder, a Gated Recurrent Unit (GRU) decoder, and a multi-head attention mechanism. This approach enhances prediction accuracy and robustness across different control modes and load spectra by leveraging multi-channel inputs and cross-variable feature interactions, thereby capturing both short-term high-frequency dynamics and long-term slow drift characteristics. Experiments using long-term data from real test benches demonstrate that the model achieves a stable MSE below 0.01 on the validation set, with MAE and RMSE of approximately 0.018 and 0.052, respectively, and a coefficient of determination reaching 0.98. This significantly outperforms traditional identification methods and single RNN models. Sensitivity analysis indicates that a prediction stride of 10 achieves an optimal balance between accuracy and computational overhead. Ablation experiments validated the contribution of multi-head attention and decoder architecture to enhancing cross-variable coupling modeling capabilities. This model can be applied to residual-driven early warning in health monitoring, and risk assessment with scheme optimization in test design. It enables near-real-time deployment feasibility, providing a practical data-driven technical pathway for reliability assurance in advanced equipment.

A Scalable V2X Architecture for Predictive Maintenance Utilizing Federated Learning in Smart Mobility

Integrating deep generative model with active learning for predicting immunotherapy responses in gastric cancer.

This study investigates an inpainting-based data generation approach for urban change detection and analyzes its impact on the model generalization performance. The proposed method constructs temporally consistent before and after image pairs by applying inpainting to existing images, thereby enabling seamless integration with existing change detection frameworks. Rather than replacing real-world datasets, the generated data were designed to increase the training diversity and mitigate dataset-specific biases. Comprehensive experiments were conducted using multiple representative change detection architectures across different datasets. The results show that incorporating the generated data alongside real training samples improves the cross dataset generalization performance, particularly in unseen target domains, whereas the degree of improvement varies depending on the model characteristics. By contrast, training with the generated data alone leads to limited generalization when evaluated on real test sets. These findings indicate that inpainting-based synthetic data can serve as a complementary training resource for enhancing the robustness of urban change detection. Future work will focus on reducing the domain gap between synthetic and real data, generating more diverse change scenarios, and further improving the reliability of data quality assessment.

This research focuses on analyzing the performance of the Control Pilot (CP) system on electric vehicle supply equipment (EVSE) through simulation and physical testing. The goal is to ensure the reliability of communication between EVSEs and electric vehicles (EVs). Using simulations on the Falstad Circuit Simulator, the study successfully showed that the CP system can recognize and respond to charging operational conditions with high accuracy, from condition A (not connected) to condition D (Requires ventilation) with a measurable error percentage in the range (0.8%-6.83%). However, the most crucial findings emerged from physical tests using dummy weights. Although the simulation was successful, real testing on Condition D (requiring ventilation) actually showed a communication failure. EVSE displays an "ERROR" message and stops charging. This failure highlights that under certain conditions, inaccurate communication can trigger the EVSE security system to cut off power. This indicates that there is a significant difference between the simulation results and operational reality, which is the main contribution of this research.

Open pit mining is a common technique for extracting surface mineral resources, but it poses safety risks due to terrain instability and slope movements. Accurate and efficient monitoring of surface changes is essential for operational safety and informed decision-making. Recently, UAV-based photogrammetric point clouds have emerged as a cost-effective and flexible alternative to traditional geodetic methods. Among the analysis tools, Multiscale Model to Model Cloud Comparison (M3C2) stands out for detecting 3D changes between epochs. However, standard M3C2 relies on internal accuracy, which may lead to overly optimistic assessments. This study introduces a visualization-based approach using realistic external-accuracy test statistics derived from GNSS control points. Within the approach, a new visualization metric based on the ratio of the norms of 2D and 3D change vectors is proposed. In cases where there is a predominant movement in the horizontal direction, the calculated ratio values are larger compared to their immediate surroundings. In this way, these locations become more perceptible and user attention is directed to the highlighted locations. Thus, it will be possible especially for non-specialized users to better examine these prominent locations in the images and in the field when necessary. In this way, decision makers will be provided with a practical tool to alert the field personnel working in the relevant locations and to ensure security quickly. The findings show that visualization supported by realistic accuracy values can significantly contribute to the identification of landslide areas that may be caused by excavation and are likely to progress over time in an open pit mine site.

Titanium alloy TC4 countersunk head bolts (CHB) are widely used in spacecraft structures, but the research on CHB does not receive enough attention at present. There are still some more opportunities worthy of in-depth research, such as insufficient research on CHB of high-strength fasteners for aerospace applications, an insufficient combination of CHB simulation tests with real working conditions, and inspection and testing methods. In this study, through the combination of finite element simulation and experiments, the working conditions of the CHB connection structure bearing tensile load and CHB screwing were analyzed, and the requirements of the CHB connection structure and installation of CHB were optimized. Based on the single-bolt tensile simulation, the working conditions of multi-bolt connection structures under eccentric load and single-bolt composite laminate connection structures under tensile load were analyzed. Meanwhile, the structure of CHB was further optimized, and the simulation analysis model of the CHB tightening process was established. The research shows that the larger fixing bolt countersunk angle θ1 and the smaller countersunk fillet radius r, the better the ultimate bearing capacity of the connection structure will be. When the countersunk bevel angle of pressure plate θ2 was greater than or less than 100°, the clamping force–angle slope will decrease, while when θ2 was smaller, it will have a greater influence on the slope. The coaxiality Φ had little influence on the slope around the allowable tolerance range (0.3 mm), but the influence on the slope becomes greater when it exceeds the tolerance range. The research results provide a reference and basis for the layout of CHB and the use of composite materials in aerospace connection structures.

Certain fieldwork settings pose challenges that often test the conventional methodological and academic wisdom that we have gained as a part of our training. Our research acumen faces the real test when we particularly research vulnerable and marginalised communities. This article reflects on my experiences of conducting fieldwork using an autoethnographic approach among the marginalised ‘Watal’ community in Kashmir. As a female researcher, I encountered unique challenges related to gender and ethical issues while negotiating access to the community. In this article, I examine topics such as site selection, the complexities involved in navigating sociopolitical dynamics and the reflexive nature of conducting research in conflict-affected areas. This work underscores the interplay between personal identity and fieldwork experience, providing insights into the intricate challenges of ethnographic research within marginalised and excluded settings.

After the unmanned transformation of the transmission system of a heavy tracked vehicle, there are problems of low shift smoothness and long shift time during remote driving. Aiming to realize the smooth shift of tracked vehicles by remote driving, this study mines the manned driving data of the vehicle and develops an auxiliary decision-making model for the shift timing of remote driving. In addition, an integrated shift control strategy of tracked vehicle power transmission based on a hierarchical finite state machine is designed. The effectiveness of the proposed remote control shift strategy for tracked vehicles is verified by bench test and real vehicle test. During the entire shifting process, the average shift impact of the tracked vehicle is 9.02:text{m/}{text{s}}^{3}, the average shift time is 4.6:s, and the vehicle has good ride comfort。The research results can provide useful guidance and lay a foundation for the research of driving control and trajectory tracking control of this type of heavy tracked vehicle after unmanned transformation.

Background: Attention-deficit/hyperactivity disorder (ADHD) is one of the most prevalent neurodevelopmental disorders, defined by persistent patterns of inattention, hyperactivity, and/or impulsivity that impair academic, social, and occupational functioning across the lifespan. Affecting approximately 5% of children and adolescents worldwide, ADHD frequently persists into adulthood, contributing to long-term psychosocial difficulties. Objective: To outline the limitations of current ADHD diagnostic approaches and highlight how immersive virtual reality (VR) may strengthen ecological relevance, subtype differentiation, and contextual sensitivity. Methods: We reviewed the shortcomings of existing diagnostic tools, including their restricted ecological validity and limited ability to capture real-world attentional demands. We then examined immersive VR as a platform capable of standardized, realistic testing environments and simultaneous acquisition of multimodal behavioral and physiological data. Results: Evidence suggests that VR-based assessments can enhance the precision of ADHD subtype profiling, support individualized behavioral characterization, and overcome several constraints of traditional diagnostic methods. Common VR paradigms and prior examples of VR-based diagnostic platforms are summarized. Conclusion: Immersive VR offers a promising approach for advancing ADHD assessment by improving ecological validity, enabling objective multimodal measurement, and supporting more personalized subtype differentiation. Further research may refine its integration into clinical workflows.

During the process of automatic parking, parking slot detection often becomes discontinuous due to visual blind areas and vehicle occlusion, causing the vehicle to lose relative positional information with respect to the parking slot, which may eventually lead to parking failure. To address this issue, this paper proposes a fusion method of AVM and vehicle chassis sensor information based on a dynamic Bayesian network. The proposed method enables simultaneous parking slot tracking and vehicle self-localization, thereby ensuring continuous and accurate detection of both the vehicle’s position and the parking slot’s position. Firstly, the coupling mechanism between parking slot tracking and vehicle localization is modeled based on dynamic Bayesian network. Additionally, A robust extended Kalman filter algorithm for accurate vehicle localization and parking slot tracking is used, which can fuse the information of vision and vehicle motion and reduce the impact of dynamic measurement error. Finally, the proposed algorithm is validated through real car tests with a light pickup truck. The test results show that the proposed algorithm can improve the accuracy of parking slot tracking and vehicle localization and can meet the functional requirements of automatic parking.

Discriminating between perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS), while suppressing interference from common surfactants, such as sodium dodecyl sulfate (SDS), remains a persistent challenge. In this work, we tackled this challenge through a rational side-chain engineering strategy, designing a series of cationic polythiophenes that exploit analyte-specific conformational transitions for unprecedented selectivity. Among them, PT32 was utilized as a trimodal (UV-vis absorption, colorimetric, and fluorometric) probe, enabling discriminative detection of PFOA/PFOS and distinction from other anionic surfactants. The system demonstrates clear visual and spectral discriminability, with PFOA inducing a distinct pink color, PFOS producing an orange-red response, and SDS showing negligible color change. This selectivity stems from variable conformational changes in the PT32 backbone (e.g., random aggregation, extended conformation, and supramolecular aggregation), which are driven by synergistic electrostatic, fluorophilic, and hydrophobic interactions. The PT32 probe exhibited remarkably LOD of 41.61nM for PFOA and 25.71nM for PFOS in fluorescence-based detection and achieved excellent recovery rates in real sample tests. Interestingly, the probe could be recycled post-detection and use multiple times. Hence, the side-chain engineering strategy not only addresses long-standing challenges in discrimination but also establishes a versatile design principle for developing advanced molecular probes.

Abstract Discriminating between perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS), while suppressing interference from common surfactants, such as sodium dodecyl sulfate (SDS), remains a persistent challenge. In this work, we tackled this challenge through a rational side‐chain engineering strategy, designing a series of cationic polythiophenes that exploit analyte‐specific conformational transitions for unprecedented selectivity. Among them, PT32 was utilized as a trimodal (UV–vis absorption, colorimetric, and fluorometric) probe, enabling discriminative detection of PFOA/PFOS and distinction from other anionic surfactants. The system demonstrates clear visual and spectral discriminability, with PFOA inducing a distinct pink color, PFOS producing an orange‐red response, and SDS showing negligible color change. This selectivity stems from variable conformational changes in the PT32 backbone (e.g., random aggregation, extended conformation, and supramolecular aggregation), which are driven by synergistic electrostatic, fluorophilic, and hydrophobic interactions. The PT32 probe exhibited remarkably LOD of 41.61 nM for PFOA and 25.71 nM for PFOS in fluorescence‐based detection and achieved excellent recovery rates in real sample tests. Interestingly, the probe could be recycled post‐detection and use multiple times. Hence, the side‐chain engineering strategy not only addresses long‐standing challenges in discrimination but also establishes a versatile design principle for developing advanced molecular probes.

ABSTRACT The powered roof supports in the longwall complex are placed and interconnected next to one another, typically about 125–250 units in a longwall face. It is an essential component in the longwall complex, which also includes a shearer (or plow) and an armored face conveyor. Powered roof support perform two main functions, the first and most important is to support and control the immediate roof in the work area, and the second is it moves the armored face conveyor and assembly forward as the shearer cuts the coal face. A robust, reliable and interconnected hydraulic system is used to perform these 2 vital functions. Research on developing the real time measurement system for monitoring the powered roof support operation parameters was based mainly on the data from longwall mining panels. For this purpose, a prototype measurement system for the powered roof support was designed and built. The structural elements measured are the roof, the stringer, the lemniscate system and the shield system. A mining wall was selected, in which the sections of the powered roof support were equipped with suitable sensors for measuring. Based on the adopted mathematical relationship, the longitudinal and transverse slope coefficient for the safe operation of the powered roof support was determined. The real conditions tests confirmed the correct operation of the measurement system.

Most of the modern companies use different agile methodologies for work processes. Such methodologies apply an iterative approach for more robust and efficient development and production. Scrum – is one of the most popular agile frameworks. Such kind of framework works with tasks that can be estimated using story points, that are represented by rounded Fibonacci sequence. Estimation is one of the most important parts of the working process as it helps to create a working plan and achieve goals. Also, the work process is split into sprints – fixed period of time (usually 1-4 weeks) during which team is supposed to develop a ready part of the product. Teams usually perform estimation manually based on the previous experience. Since estimation is not an accurate value, but rather a product of team’s skills, we can use fuzzy logic to create an automatic estimation system. We develop system that can estimate task with predefined parameters such as title, description, and type, based on previously completed tasks. In this case we use mock tasks with predefined estimation. The system performs calculation to the analyze input task text parameters, compare with the previous tasks, and outputs the estimated value. For testing purposes, we developed a program that uses Jira API, React, Forge, and .NET project. The input task is received from the Jira board. Calculations are performed on local server. The program outputs estimation to Jira task. Future work includes the further development of the algorithm, program and real-life tests.

Most available studies on the mechanical properties of apples lack the comprehensive results needed for the construction and validation of static FEM models. Researchers typically focus on either the flesh and epidermis or the whole fruit, often overlooking the maturity stage of the examined apples. They usually report only firmness or the starch index as indicators of maturity. Furthermore, many studies use store-bought apples, which is impractical for industrial applications since this fruit has already undergone various treatments before reaching the shelf. This article aims to determine the mechanical properties of apples necessary for constructing static FEM models that are both adequate and useful for the industry. The new Polish apple variety, Chopin, was selected as the research material. The study was conducted for three stages of apple maturity: development, ripening, and senescence. Mechanical properties of the flesh and skin were determined as material data for FEM models. Force-displacement curves and pressure-force functions were examined for future model validation. Using micro-computed tomography, the bruise volumes of fruit subjected to 20%, 50%, and 80% of the destructive force were determined. Significant differences were found between apples in the senescence stage and those in the development and ripening stages. Results of Micro-Ct and the results of modified and real compression tests of whole fruit have allowed us to formulate the research hypothesis regarding the influence of flesh cracking (characterized by local drops of force), influencing the bruise visibility and detection.

Purpose of the study: Students’ conceptual understanding of Solar System topics remains low due to the abstract nature of astronomical phenomena and the limited use of interactive visual learning media. This study aims to develop AR-based Solar System learning media that are valid, practical, and effective in improving students’ conceptual mastery. Methodology: The research employed a Research and Development approach using the 4D model. Participants comprised 46 eighth-grade students from Muhammadiyah 1 Junior High School in Palembang, selected through purposive sampling. Data collection involved teacher interviews, expert validation (3 experts), student practicality questionnaires (9 students) and pretest-posttest assessments (46 respondents),. The instruments included closed-ended Likert-scale items, open-ended feedback, and concept mastery tests. Data analysis was conducted using the Guttman scale for validity, the Likert scale for practicality, and the N-gain test for effectiveness. Main Findings: The developed AR learning media achieved a validity score of 100%, a practicality score of 96.82% (very practical), and high effectiveness, with an average N-gain of 0.71. Students’ posttest scores showed substantial improvement compared to pretest results, indicating enhanced understanding of planetary motion, rotation, and revolution. Novelty/Originality of this study: The findings confirm that AR-based learning media effectively facilitate conceptual change by transforming abstract Solar System concepts into concrete, interactive experiences. The novelty of this study lies in integrating realistic 3D visualization, curriculum-oriented design, and effectiveness testing. The results imply that AR media can serve as a viable instructional innovation in science education, particularly for abstract topics that require spatial reasoning.

The purpose of the study is to develop and verify a mathematical model of the cement-sand matrix (CSM) of concrete with an auxetic returnably-concave structure structure to enhance the energy-absorbing properties of construction materials used in special facilities (SF) of the Civil Defense of the Russian Federation. The work is based on the analysis of the physical and mechanical characteristics of composite materials modified with voids of various geometries, taking into account the requirements for protective structures operating under elevated dynamic loads of technogenic nature. Methods include mathematical modeling using the finite element method to create a verified model of the cement-sand matrix of concrete with compressive strength classes B30 and B60. Verification was performed by comparing the modeling results with data from real tests on a press, specifying boundary conditions, physical and mechanical properties of materials, and loading parameters. For the auxetic structure, returnably-concave structure cells with a concavity angle of 60° and an edge length of 15 mm were used, modeled for cubes of various sizes (scaling and overlapping of cells). The results demonstrate uniform stress distribution in the auxetic structure of the CSM without the "hoop" effect characteristic of standard concrete, with a reduction in compressive strength (down to 8.7 MPa for B30 and 17.4 MPa for B60) while increasing deformability up to 10 times. Stress-strain diagrams confirm increased compliance, and consequently, energy absorption. The practical significance lies in the application of the model for designing energy-absorbing protective structures of SF, enhancing their protection.

Yeast-derived biomolecules are redefining the boundaries of green nanotechnology. Biosurfactants, exopolysaccharides, enzymes, pigments, proteins, and organic acids—when sourced from carbohydrate-rich lignocellulosic hydrolysates—offer a molecular toolbox capable of directing, stabilizing, and functionalizing nanoparticles (NPs) with unprecedented precision. Beyond their structural diversity and intrinsic biocompatibility, these biomolecules anchor a paradigm shift: the convergence of biorefineries with nanotechnology to deliver multifunctional materials for the circular bioeconomy. This review explores: (i) the expanding portfolio of metallic and metal oxide NPs synthesized through yeast biomolecules; (ii) molecular-level mechanisms of reduction, capping, and surface tailoring that dictate NP morphology, stability, and reactivity; (iii) synergistic roles in intensifying lignocellulosic processes—from enhanced hydrolysis to catalytic upgrading; and (iv) frontier applications spanning antimicrobial coatings, regenerative packaging, precision agriculture, and environmental remediation. We highlight structure–function relationships, where amphiphilicity, charge distribution, and redox activity govern resilience under saline, acidic, and thermally harsh industrial matrices. Yet, critical bottlenecks remain: inconsistent yields, limited comparative studies, downstream recovery hurdles, and the absence of comprehensive life-cycle and toxicological evaluations. To bridge this gap, we propose a translational roadmap coupling standardized characterization with real hydrolysate testing, molecular libraries linking biomolecule chemistry to NP performance, and integrated techno-economic and environmental assessments. By aligning yeast biotechnology with nanoscience, we argue that yeast-biomolecule-driven nanoplatforms are not merely sustainable alternatives but transformative solutions for next-generation lignocellulosic biorefineries.