Search Results (193)

This study presents an innovative bimodal approach for laboratory partial discharge (PD) analysis using a YOLOv8-based convolutional neural network (CNN). The main contribution consists, first, in the transformation of a conventional DDX-type electrical detector into a smart and autonomous data source. By training the CNN, a system capable of automatically reading and interpreting the data from the detector display—discharge magnitude and applied voltage—is developed, achieving an average training accuracy of 0.91 and converting a passive instrument into a digitalized and structured data source. Second, and simultaneously, an optical visualization system captures direct images of the PDs with a high-resolution camera, allowing for their morphological characterization and spatial distribution. For electrical voltages of 10, 13, and 16 kV, PDs were detected with a confidence level of up to 0.92. The fusion of quantitative information intelligently extracted from the electrical detector with qualitative characterization from optical analysis offers a more complete and robust automated diagnosis of the origin and severity of PDs. Full article

A sheet-type sinter-bonding material was developed to form thermally stable and highly heat-conductive joints suitable for wide-bandgap (WBG) semiconductor dies and high-heat-flux devices, and its bonding characteristics were investigated. To enhance the cost-competitiveness of the bonding material, Ag-coated Cu (Cu@Ag) particles were employed as fillers instead of conventional Ag particles. To facilitate accelerated sintering, a bimodal particle size distribution comprising several micron- and submicron-sized particles was adopted by synthesizing and mixing both size ranges. For sheet fabrication, a decomposable resin was used as the essential binder component, which could be removed during the bonding process via thermal decomposition. This approach enabled the formation of a sintered bond line composed entirely of Cu@Ag particles. Thermogravimetric and differential thermal analyses revealed that the decomposition of the resin in the sheet occurred within the temperature range of 290–340 °C. Consequently, sinter-bonding conducted at 350 °C and 370 °C exhibited significantly superior bondability compared to bonding at 330 °C. In particular, sinter-bonding at 350 °C for just 60 s resulted in a highly densified joint microstructure with a low porosity of 7.6% and high shear strength exceeding 25 MPa. The formation of the bond line was initiated by sintering between the outer Ag shells of the adjacent particles. However, with increasing bonding time or temperature, sintering driven by Cu diffusion from the particle cores to the outer Ag shells, particularly in the submicron-sized particles, was progressively enhanced. These results obtained from the fabricated sheet-type materials demonstrate that, even with the use of resin, rapid solid-state sintering between filler particles combined with the removal of resin through decomposition enables the formation of a metallic bond line with excellent thermal conductivity. Full article

The timely detection of falls among the elderly remains challenging. Single modality sensing approaches using inertial measurement units (IMUs) or vision-based monitoring systems frequently exhibit high false positives and compromised accuracy under suboptimal operating conditions. We propose a novel bimodal deep learning-based bimodal sensing framework to address the problem, by leveraging a memory-based autoencoder neural network for inertial abnormality detection and an attention-based neural network for visual pose assessment, with late fusion at the decision level. Our experimental evaluation with a custom dataset of simulated falls and routine activities, captured with waist-mounted IMUs and RGB cameras under dim lighting, shows significant performance improvement by the described bimodal late-fusion system, with an F1-score of 97.3% and, most notably, a false-positive rate of 3.6% significantly lower than the 11.3% and 8.9% with IMU-only and vision-only baselines, respectively. These results confirm the robustness of the described fall detection approach and validate its applicability to real-time fall detection under different light settings, including nighttime conditions. Full article

The increasing environmental concerns surrounding plastic waste have intensified recycling efforts, particularly in the textile industry, where poly(ethylene terephthalate) (PET) is widely used for sustainable material production. The growing use of recycled PET (rPET) in textiles has prompted the need for reliable analytical methods to detect and quantify rPET content. This study differentiates between virgin and recycled PET by simulating mechanical recycling through five reprocessing cycles of three distinct PET grades, assessing changes in crystalline structure, intrinsic viscosity, molecular weight, and specific degradation markers. Differential Scanning Calorimetry revealed bimodal melting behaviour in reprocessed samples, while intrinsic viscosity and Gel Permeation Chromatography indicated molecular degradation. Notably, the release of dimethyl terephthalate (DMT) and dimethyl isophthalate (DMiP) was consistently observed as a function of degradation. These markers were identified and quantified using High-Performance Liquid Chromatography (HPLC) and Gas Chromatography (GC), with GC offering higher sensitivity and lower matrix interference. This study demonstrates that DMT and DMiP are robust chemical indicators of PET degradation and recycled content. This analytical approach, combining thermal, rheological, and chromatographic techniques, provides a scientifically sound and potentially cost-effective basis for traceability systems, certification protocols, and regulatory compliance in sustainable textile production. Full article

The rising demand for platinum-group metals, driven by their essential applications in catalysis, energy storage, and chemical conversion, underscores the need to identify new sources for their recovery. Waste solutions originating from industrial processes offer a promising alternative source of noble metals. However, due to their typically low concentrations, effective recovery requires a highly targeted approach. In this study, we present a synthetic waste solution containing trace amount of Rh(III) ions as both a medium for metal ion recovery and a direct precursor for catalyst synthesis. Using a bimodal water–ethanol solvent system, ultra-small rhodium nanoparticles were synthesized and subsequently immobilized onto activated carbon fibers (ACFs) within a microreactor system. The resulting Rh@ACF catalyst demonstrated high efficiency in the reduction of 4-nitrophenol (4-NP) to 4-aminophenol (4-AP), serving as a model catalytic reaction. The Rh@ACF catalyst, containing 4.24 µg Rh per milligram of sample, exhibited notable catalytic activity, achieving 75% conversion of 4-NP to 4-AP within 1 h. Full conversion to 4-AP was also reached within 5 min, but requires extra NaBH₄ addition to the catalytic mixture. Full article

This study presents the diagnostics and microclimate analysis of four case studies located in the Alps region in Valtellina and Valposchiavo. The primary focus is on evaluating and comparing microclimatic conditions, encompassing temperature (T°C), relative humidity (RH%), mixing ratio (MR), and dew point depression (DPD). The choice of the variables and statistic metrics depends substantially on the aim to identify the risk factor for the preservation of the historical materials of historical buildings, and the procedures for identifying the anomalies in the trends useful to study how to prevent these anomalies in the future. The paper has the target to support the activities of restorers and building managers for improving the restoration process. While various moisture detection methodologies have been studied, no single approach is preferred for analyzing moisture via microclimate monitoring in built heritage. Therefore, this research delves into the influence of various factors, including altitude, location, building type, structure, materials, orientation, and use, on the microclimatic parameters. Altitude and building use significantly influence indoor microclimates: unoccupied structures exhibit greater stability, whereas seasonal use increases condensation risks. Key risks included high RH% and critical T-RH zones (T > 25 °C + RH > 65%), exacerbating material stress. Probability density function (PDF) analysis reveals temperature and RH% distributions, highlighting bimodal T°C patterns and prolonged RH% in high-elevation exposed sites. The findings underscore the need for tailored conservation strategies and targeted interventions to mitigate microclimate-induced deterioration in Alpine heritage. Full article

Third-generation wide-bandgap (WBG) semiconductor power electronics exhibit excellent workability, but high-temperature packaging technology limits their applications. TLP, TLPS, and nanoparticle sintering have the potential to achieve a high-temperature-resistant joint at a lower bonding temperature. However, a long bonding time, voids in the joint, powder oxidation, and organic solvent residues impede their application. A novel interlayer and other approaches have been proposed, such as preformed Sn-coated Cu foam (CF@Sn), a Cu-Sn nanocomposite interlayer, self-reducible Cu nanoparticle paste, bimodal-sized Cu nanoparticle pastes, organic-free nanoparticle films, and high-thermal-conductivity and low-CTE composite paste. Their preparation, bonding processes, and joint properties are compared in this paper. Full article

The accurate dosing of polyaluminum chloride (PAC) is essential for achieving effective coagulation in drinking water treatment, yet conventional methods such as jar tests are limited in their responsiveness and operational efficiency. This study proposes a hybrid modeling framework that integrates artificial neural networks (ANN) with genetic algorithms (GA) to optimize PAC dosage under variable raw water conditions. Operational data from 400 jar test experiments, collected between 2022 and 2024 at the Yanahurco water treatment plant (Ecuador), were used to train an ANN model capable of predicting six post-treatment water quality indicators, including turbidity, color, and pH. The ANN achieved excellent predictive accuracy (R² > 0.95 for turbidity and color), supporting its use as a surrogate model within a GA-based optimization scheme. The genetic algorithm evaluated dosage strategies by minimizing treatment costs while enforcing compliance with national water quality standards. The results revealed a bimodal dosing pattern, favoring low PAC dosages (~4 ppm) during routine conditions and higher dosages (~12 ppm) when influent quality declined. Optimization yielded a 49% reduction in median chemical costs and improved color compliance from 52% to 63%, while maintaining pH compliance above 97%. Turbidity remained a challenge under some conditions, indicating the potential benefit of complementary coagulants. The proposed ANN–GA approach offers a scalable and adaptive solution for enhancing chemical dosing efficiency in water treatment operations. Full article

Repetitive DNA represents over 50% of the human genome and is an abundant component of circulating cell-free DNA (cfDNA). We previously showed that cfDNA levels and integrity can predict survival in elderly patients with cardiovascular disease. Here, we aimed to clarify whether a low-pass next-generation sequencing (NGS) approach can characterize the repeat content of cfDNA. Considering the bimodal distribution of cfDNA fragment lengths, we examined the occurrence of repetitive DNA subfamilies separately in dinucleosomal (>250 bp) and mononucleosomal (≤250 bp) cfDNA sequences from 24 patients admitted for heart failure. An increase in the relative abundance of Alu repetitive elements was observed in the longer fraction, while alpha satellites were enriched in the mononucleosomal fraction. The relative abundance of Alu, ALR, and L1HS DNA in the dinucleosomal fraction correlated with different prognostic biomarkers, and Alu DNA was negatively associated with the presence of chronic kidney disease comorbidity. These results, together with the observed inverse correlation between Alu DNA abundance and cfDNA integrity, suggest that the composition of plasma cfDNA could be determined by multiple mechanisms in different physio-pathological conditions. In conclusion, low-pass NGS is an inexpensive method to analyze the cfDNA repeat landscape and identify new cardiovascular disease biomarkers. Full article

The advent of alignment-free k-mer barcoding has revolutionized taxonomic analysis, enabling bacterial identification at phylogroup resolution within natural communities. We applied this approach to characterize Escherichia coli intraspecific diversity in human gut microbiomes using publicly available datasets representing diverse human physiological states. By estimating the relative abundance of eight E. coli phylogroups defined by their 18-mer markers in 558 fecal samples, we compared their distribution between gut microbiomes of healthy individuals, patients with chronic bowel diseases and volunteers subjected to various external interventions. Across all datasets, phylogroups exhibited bidirectional abundance shifts in response to host physiological changes, indicating an inherent bimodality in their adaptive strategies. Correlation analysis of phylogroup persistence revealed positive intraspecific connectivity networks and dependence of their patterns on both acute interventions like antibiotic or probiotic treatment and chronic bowel disorders. Along with predominantly negative correlations with Bacteroides, we observed a transition from positive to negative associations with Prevotella in Prevotella-rich microbiomes. Several interspecific correlations individually established by E. coli phylogroups with dominant taxa suggest their potential role in shaping intraspecific networks. Machine learning techniques statistically confirmed an ability of phylogroup patterns to discriminate the physiological state of the host and virtual diagnostic assays opened a way to optimize intraspecific phylotyping for medical applications. Full article

Enhancing generalization capabilities and robustness in scenarios with limited sample sizes, while simultaneously decreasing reliance on extensive and high-quality datasets, represents a significant area of inquiry within the domain of radar target recognition. This study introduces a few-shot learning framework that leverages multimodal feature fusion. We develop a cross-modal representation optimization mechanism tailored for the target recognition task by incorporating natural resonance frequency features that elucidate the target’s scattering characteristics. Furthermore, we establish a multimodal fusion classification network that integrates bi-directional long short-term memory and residual neural network architectures, facilitating deep bimodal fusion through an encoding-decoding framework augmented by an energy embedding strategy. To optimize the model, we propose a cross-modal equilibrium loss function that amalgamates similarity metrics from diverse features with cross-entropy loss, thereby guiding the optimization process towards enhancing metric spatial discrimination and balancing classification performance. Empirical results derived from simulated datasets indicate that the proposed methodology achieves a recognition accuracy of 95.36% in the 5-way 1-shot task, surpassing traditional unimodal image and concatenation fusion feature approaches by 2.26% and 8.73%, respectively. Additionally, the inter-class feature separation is improved by 18.37%, thereby substantiating the efficacy of the proposed method. Full article

Imbalanced regression poses a significant challenge in real-world prediction tasks, where rare target values are prone to overfitting during model training. To address this, prior research has employed relevance functions to quantify the rarity of target instances. However, existing functions often struggle to capture the rarity across diverse target distributions. In this study, we introduce a novel Distance-based Relevance Function (DRF) that quantifies the rarity based on the distance between target values, enabling a more accurate and distribution-agnostic assessment of rare data. This general approach allows imbalanced regression techniques to be effectively applied to a broader range of distributions, including bimodal cases. We evaluate the proposed DRF using Mean Squared Error (MSE), relevance-weighted Mean Absolute Error ($\mathrm{MAE}\varphi$), and Symmetric Mean Absolute Percentage Error (SMAPE). Empirical studies on synthetic datasets and 18 real-world datasets demonstrate that DRF tends to improve the performance across various machine learning models, including support vector regression, neural networks, XGBoost, and random forests. These findings suggest that DRF offers a promising direction for rare target detection and broadens the applicability of imbalanced regression methods. Full article

Fire regime (FR) is a key element in the study of ecosystem dynamics, supporting natural resource management planning by identifying gaps in fire patterns in time and space and planning to assess ecological conditions. Due to the insufficient consideration of integrated characterization factors, especially the insufficient research on fire season types (FST), the current understanding of the spatial heterogeneity of fire patterns in China is still limited, and it is necessary to use FST as a key dimension to classify FR zones more accurately. This study extracted 13 fire characteristic variables based on Moderate Resolution Imaging Spectroradiometer (MODIS) burned area data (MCD64A1), active fire data (MODIS Collection 6), and land cover data (MCD12Q1) from 2001 to 2023. The study systematically analyzed the frequency, intensity, spatial distribution and seasonal characteristics of fires across China. By using data normalization and the k-means clustering algorithm, the study area was divided into five types of FR zones (FR 1–5) with significant differences. The burned areas of the five FR zones account for 67.76%, 13.88%, 4.87%, 12.94%, and 0.55% of the total burned area across the country over the 23-year study period, respectively. Among them, fires in the Northeast China Plain and North China Plain cropland areas (FR 1) exhibit a bimodal distribution, with the peak period concentrated in April and June, respectively; the southern forest and savanna region (FR 2) is dominated by high-frequency, small-scale, unimodal fires, peaking in February; the central grassland region (FR 3) experiences high-intensity, low-frequency fires, with a peak in April; the east central forest region (FR 4) is characterized by low-frequency, high-intensity fires; and the western grassland region (FR 5) experiences low-frequency fires with significant inter-annual fluctuations. Among the five zones, FST consistently ranks within the top five contributors, with contribution rates of 0.39, 0.31, 0.44, 0.27, and 0.55, respectively, confirming that the inclusion of FST is a reasonable and necessary choice when constructing FR zones. By integrating multi-source remote sensing data, this study has established a novel FR classification system that encompasses fire frequency, intensity, and particularly FST. This approach transcends the traditional single-factor classification, demonstrating that seasonal characteristics are indispensable for accurately delineating fire conditions. The resultant zoning system effectively overcomes the limitations of traditional methods, providing a scientific basis for localized fire risk warning and differentiated prevention and control strategies. Full article

The complexity of the pore system hindered our understanding of the storage and transport properties of organic-rich shales, which in turn brought challenges to the efficient exploration and development of shale oil and gas. This study, based on elemental, mineralogical, petrographic, and petrophysical approaches, attempts to reveal the pore structure and fractal characteristics of a suite of Permian shales collected from the northeastern Sichuan Basin, China. The results showed that meso-pores make up the main proportion of the total pore volume in the Permian shale in this study; most of the pore size distribution patterns for micro pores and meso-macropores are bimodal. Pores related to clay minerals, organic matter pores, and intragranular dissolution pores are the main storage spaces in these shales. In these samples, ink-bottle pores dominate, with some slit and wedge-shaped ones developed. The morphology of the pores in the studied shales is mainly ink-bottle pores, with some slit-shaped and wedge-shaped pores. The fractal dimension D₂ is greater than D₁, indicating that the homogeneity of pore space is stronger than that of the specific surface area. Quartz in Permian shales inhibits the development of macro- and mesopore spaces and enhances pore heterogeneity, while clay minerals facilitate the development of macro- and mesopore spaces and attenuate pore heterogeneity. The organic matter content shows a negative impact on the macropore volume due to the stripped occurrence and matrix filling. This study has a vital significance for current exploration and development of shale gas in Permian strata in the Sichuan Basin and offers insights for Permian shales in other basins all over the world. Full article

Organic matter pores constitute a significant storage space in shale gas reservoirs, contributing to approximately 50% of the total porosity. This study employed a comprehensive approach, utilizing scanning electron microscopy, low-pressure N₂ adsorption, thermal analysis, image statistics, and fractal theory, to quantitatively characterize the structure and complexity of organic matter pores in the Wufeng–lower Longmaxi Formations (WLLFs). The WLLFs exhibit a high organic matter content, averaging 3.20%. Organic matter pores are typically well-developed, predominantly observed within organic matter clusters, organic matter–clay mineral complexes, and the internal organic matter of pyrite framboid. The morphology of these pores is generally elliptical and spindle-shaped, with the primary pore diameter displaying a bimodal distribution at 10~40 nm and 100~160 nm, potentially influenced by the observational limit of scanning electron microscopy. Shales from greater burial depths within the same gas well contain more organic matter pores; however, the development of organic matter pores in deep gas wells is roughly equivalent to that in medium and shallow gas wells. Fractal dimension values can be utilized to characterize the complexity of organic matter pores, with organic matter macropores (D_>50) being more complex than organic matter mesopores (D_2–50), which in turn are more complex than organic matter micropores (D_<2). The development of macropores and mesopores is a key factor in the heterogeneity of organic matter pores. The complexity of organic matter pores in the same well increases gradually with the burial depth of the shale, and the complexity of organic matter pores in deep gas wells is roughly equivalent to that in medium and shallow gas wells. The structure and fractal characteristics of organic matter pores in shale are primarily controlled by components, diagenesis, tectonism, etc. The lower Longmaxi shale exhibit a high biogenic quartz content and robust hydrocarbon generation from organic matter. This composition effectively shields organic matter pores from multi-directional extrusion, leading to the formation of macropores and mesopores without specific orientation. High-quality shale sections (one and two sublayers) have relatively high fractal dimension D_2–50 and D_>50 values of organic matter pores and gas content. Consequently, the quality parameters of shale and fractal dimension characteristics can be comprehensively evaluated to identify high-quality shale sections. Full article