Large-size concrete filled steel tubular (CFST) columns are increasing adopted in long-span and heavy-loaded structures. An extensive review of the existing test database has revealed that square CFST columns showed the size effect of the shear strength. Given the sudden and destructive nature of earthquakes, the study on the size effect of square CFST columns under cyclic-shear loading is vital. However, the previous relevant studies mainly focused on small-size square CFST members (i.e., B = 100–200 mm) under cyclic-shear loading or on size effect of CFST members under cyclic-bending. To fill this research gap, in this work, a total of 6 square CFST columns with varying sizes and shear span-to-depth ratios were tested under cyclic-shear loading. The failure modes and load-deformation curves were analyzed detailed. The results indicated that the cyclic-shear behavior, including the peak strength, ductility, strength degradation, energy dissipation capacity and stress of the steel tube, showed the size effect to varying degrees. Finally, the tested peak strengths were compared with current design provisions and the design recommendations were given.

Wood scrimber is an innovative wood product that offers a viable solution to the shortage of large-sized wood required for structural applications. This study provides comprehensive methodologies for the numerical modelling of structural-scale beams made of scrimber. Initially, a constitutive law based on continuum damage mechanics was developed for this product, integrated into ABAQUS/Explicit as a user-defined subroutine (VUMAT). It was found that the explicit subroutine could accurately capture failure modes, avoid convergence problems, and eliminate the need for specifying mesh-size dependent fracture energy. Subsequently, a calibration approach was proposed for the non-standard specimens to transform material parameters from experimental data to simulation data, addressing the errors between macroscopic specimens and microscopic elements caused by buckling and uneven stress distribution. Furthermore, to model the bending behaviour of a structural-scale scrimber beam, a strategy was proposed to account for the size effect. The results demonstrate that the size effect resulted in a 12.5

An innovative steel parking structure with castellated beams (CBs) is proposed. The high-cycle fatigue performance of the CB under vehicle storage loading is investigated. A finite element (FE) model of the CB was established, and the stress distribution characteristics under such loading conditions were analyzed. Fracture mechanics-based numerical simulations were performed for the critical regions of the tensile opening corners and tensile bottom flanges, capturing the entire process from initial crack propagation to final fatigue failure. For the web post fatigue details subjected to shear loads, high-cycle shear fatigue tests were conducted, and the corresponding shear fatigue S-N curves were fitted. On this basis, a comprehensive parametric study was further conducted to evaluate the effects of beam configuration, crack and loading parameters on fatigue performance. The results indicate that the tensile opening corner is the most critical region, where crack propagation is dominated by Mode I behavior, and crack propagation to the web-flange junction is adopted as the fatigue failure criterion for the CB. The web post region is identified as the secondary critical location, where the butt-welded joints under shear fatigue loading are prone to cracking at the weld-base metal interface. The expansion ratio is the governing geometric parameter of the CB, with a recommended range of 1.4 to 1.5. The stress ratio is the governing loading parameter. Furthermore, the initial crack aspect ratio exerts a more significant influence on fatigue life than the initial crack size.

Current design standards for wood-based beams face two major limitations: (1) inadequate material characterisation of the asymmetry between compression and tension, and (2) insufficient consideration for catenary action caused by horizontal constraints. The aim of this study was to propose a simplified analytical method used for the beam in which both the compression - tension material discrepancy and the catenary action were incorporated. Initially, the influence mechanism of horizontal constraints on the wood beam bending behaviour was numerically analysed. Then, the simplified analytical methods were derived based on the mechanism. Finally, these methods were used to quantitatively evaluate the influence of the horizontal constraints and compressive-to-tensile strength ratio (rho) on beam performance. The proposed methods accurately captured both asymmetric tension - compression behaviour and catenary action, achieving less than 10.6% deviation from numerical simulations. Based on the proposed analytical method, the performance of wood beams can be evaluated efficiently and comprehensively. The findings indicate that beams with a lower rho exhibit greater ductility and are more suitable for resisting extreme loads. The results also demonstrates that the capacity modification factor (gamma(RE) = 0.8) recommended in current seismic codes is non-conservative for wood with rho > 0.77. This study facilitates the application of wood beams.

ObjectivePatent foramen ovale (PFO), a prevalent congenital cardiac defect, is linked to clinical conditions such as cryptogenic stroke and migraine. The genetic underpinnings of PFO remain poorly elucidated, particularly in Tibet. This study aimed to identify potential pathogenic mutations in Tibetan PFO patients via whole exome sequencing (WES) to clarify its genetic basis.MethodsEighteen Tibetan PFO patients diagnosed by echocardiography were enrolled. Peripheral blood samples underwent WES using Illumina HiSeq platform, followed by bioinformatics analysis to filter rare variants. Pathogenicity was assessed using predictive tools (SIFT, PolyPhen V2, and MutationTaster) and cardiac development-related gene databases (OMIM, HPO, HGMD, and MGI).ResultsIn this study, we identified four novel pathogenetic mutations in Tibetan PFO patients, including GABRP rs201584759 (c.421C>T: p. R141C), GJB4 rs200602523 (c.292C>T: p. R98C), RTTN rs199568901 (c.5410G>A: E1804K), and USH2A rs144768593 (c.5608C>T: p. R1870W). Further analysis indicated that GABRP, GJB4, and RTTN were significantly associated with the occurrence of congenital heart disease.ConclusionThis study first reveals genetic characteristics of Tibetan PFO patients, implicating GABRP, GJB4, RTTN, and USH2A mutations in disrupting cardiac developmental pathways, potentially contributing to the occurrence of PFO. Findings underscore genetic factors regarding PFO prevalence in populations living in high-altitude and provide insights for molecular research and precision medicine.

This study proposes an all-steel buckling-restrained brace employing low yield point steel as the core material (LYPSBRB), specifically designed for cable-stiffened latticed shell structures. Which incorporates a dual-layer restraint system with internal and external member, where rubber layers were placed between the core and restraining members to mitigate interfacial friction. Four specimen types were fabricated based on variations in external restraint member geometries and rubber layer configurations. Quasi-static tests were conducted to evaluate their load-bearing capacity and cyclic energy dissipation performance. Experimental results demonstrate that the LYPSBRB fabricated with LYP160 steel achieves a peek yield load of 173.84 kN, maintaining stable energy dissipation even under large deformations exceeding 2 % of the yield segment dimensions. Using a validated numerical simulation method, parametric analyses of LYPSBRB revealed that: cyclic energy dissipation capacity shows less than 3 % sensitivity to friction coefficient variations; friction coefficient directly influence uneven tension-compression behavior and buckling stability; A 60 % increase in core member length can double energy dissipation capacity before triggering global instability.

Wood scrimber is a novel material that addresses the shortage of large-diameter trees and the inconsistency in the mechanical properties of natural wood resources. However, the application of wood scrimber in construction is limited by the insufficient understanding of its structural performance. This study investigated the failure behaviour of wood scrimber beams from the perspectives of both material strength and structural stability, aiming to develop a design method for controlling failure modes. Initially, a finite element (FE) model was developed and validated against a previous bending test of scrimber beams. Based on the FE model, the strength failure mechanism of the scrimber beam was revealed. Subsequently, the instability failure of the beam was numerically and analytically investigated. The relationship between material parameters and geometric parameters was established, and then a design method was developed. It was found that the strength failure of wood-based materials can be classified into three distinct types in both quantitative and qualitative ways, according to the ratio of material tensile strength to compressive strength (rho). Type II failure (0.38 < rho < 0.67) was considered to be most favourable, as it represented an optimal balance between ductility and bending resistance. Notably, for wood scrimber beams, only Type I failure occurred, indicating that this material lacked ductility and might not be suitable for extreme load conditions. Moreover, wood scrimber beams exhibited superior stability against lateral torsional buckling compared to other wood-based materials, such as bamboo scrimber. It was also found that the guideline in the current national building code is inadequate for designing a wood scrimber beam with stability considerations. In contrast, the proposed design method offers superior accuracy and addresses instability problems.

Semi-supervised learning effectively mitigates the lack of labeled data by introducing extensive unlabeled data. Despite achieving success in respiratory sound classification, in practice, it usually takes years to acquire a sufficiently sizeable unlabeled set, which consequently results in an extension of the research timeline. Considering that there are also respiratory sounds available in other related tasks, like breath phase detection and COVID-19 detection, it might be an alternative manner to treat these external samples as unlabeled data for respiratory sound classification. However, since these external samples are collected in different scenarios via different devices, there inevitably exists a distribution mismatch between the labeled and external unlabeled data. For existing methods, they usually assume that the labeled and unlabeled data follow the same data distribution. Therefore, they cannot benefit from external samples. To utilize external unlabeled data, we propose a semi-supervised method based on Joint Energy-based Model (JEM) in this paper. During training, the method attempts to use only the essential semantic components within the samples to model the data distribution. When non-semantic components like recording environments and devices vary, as these non-semantic components have a small impact on the model training, a relatively accurate distribution estimation is obtained. Therefore, the method exhibits insensitivity to the distribution mismatch, enabling the model to leverage external unlabeled data to mitigate the lack of labeled data. Taking ICBHI 2017 as the labeled set, HF_Lung_V1 and COVID-19 Sounds as the external unlabeled sets, the proposed method exceeds the baseline by 12.86.

The combustion characteristic parameters of mining conveyor belts represent a crucial index for measuring the fire performance and hazard posed by combustible materials. An accurate prediction of its value provides important guidance on preventing conveyor belt fires. The critical parameters of a flame–retardant polyvinyl chloride gum elastic conveyor belt were measured under different radiative heat fluxes, including mass loss rate, heat release rate, effective heat of combustion and gas production rates for CO and CO2. The prediction method for the combustion characteristics of conveyor belts was proposed by combining a convolutional neural network with long short-term memory. Results indicated that the peak values of the mass loss, heat release, smoke production and gas production rates of CO and CO2 were positively correlated with radiative heat flux, whilst the time required to reach the peak value was negatively correlated with it. The peak time of the effective heat of combustion occurred earlier. Through deep learning modelling, mean absolute error, root mean square error and coefficient of determination were determined as 2.09, 3.45 and 9.93 × 10−1, respectively. Compared with convolutional neural network, long short-term memory and multilayer perceptron, mean absolute error decreased by 26.92%, 24.82% and 25.09%, root mean square error declined by 27.82%, 29.59% and 29.59% and coefficient of determination increased by 0.05 × 10−1, 0.06 × 10−1 and 0.06 × 10−1, respectively. The findings provide a quantitative reference benchmark for the development of conveyor belt fires and offer new technical support for the construction of early warning systems for conveyor belt fires in coal mines.

Objective:Atrial septal defect (ASD) is a common congenital heart defect with incompletely understood genetic underpinnings, particularly in specific ethnic groups. This study aimed to identify novel genetic variants related to ASD within the Tibetan population using whole exome sequencing (WES). Methods:Genomic DNA was extracted from blood samples of 17 Tibetan ASD patients. WES was performed using the Illumina HiSeq platform. After rigorous filtering, detection, and annotation of single nucleotide variations (SNVs) and insertion-deletions (InDels), potentially pathogenic variants were prioritized. Functional impact predictions were conducted using SIFT, PolyPhen V2, MutationTaster, and CADD databases to identify variants likely contributing to ASD etiology. Results:We identified nine high-confidence candidate variants in Tibetan ASD patients, including rs145116532 (ALKAL1, c.287G >A: p. R96Q), rs374798430 (AVL9, c.1267G >A: p.D423N), rs138933092 (C5, c.4432C >T: p.R1478W), rs141638421 (CRYAB, c.470G>A: p.R157H), rs147287319 (DOCK8, c.989G>A: c.1193G>A: p.R330Q, p.R398Q), rs141616597 (NTN3, c.1243C>T: p.R415C), rs117506395 (PIWIL1, c.2207C>T: p.T736M), rs142533677 (PLEKHG4, c.2246G>A: p.R749Q), and rs118203532 (TSC1, c.1460C>G: p.S487C). Function annotation further suggested potential associations of C5, CRYAB, PIWIL1, and TSC1 with congenital heart diseases. Conclusion:This first WES-based study of Tibetan ASD patients reveals population-specific genetic determinants. The nine novel candidate variants, particularly in C5, CRYAB, PIWIL1, and TSC1, provide preliminary insights into ASD etiology in high-altitude populations and highlight potential targets for future diagnostic biomarker development.

BackgroundPulmonary tuberculosis (PTB) remains a significant global health issue, with genetic factors playing a crucial role in susceptibility. Long noncoding RNA (lncRNA) C5orf64 has been implicated in immune responses and cancer, but its association with PTB risk has not been fully explored.Methods Genomic DNA was extracted from peripheral blood samples of 955 participants (474 PTB cases and 481 controls). Rs12518552 and rs2950218 in C5orf64 were genotyped using the Agena MassARRAY system. Logistic regression analysis was performed to assess the association between these polymorphisms and PTB risk. Stratified analysis was conducted to evaluate the influence of age, gender, and smoking status.ResultsRs12518552-G (OR = 0.82, p = 0.034) and rs2950218-T (OR = 0.77, p = 0.012) were associated with a reduced PTB risk. Stratified analysis revealed that rs12518552 was associated with a protective effect against PTB in individuals over 40 years old (OR = 0.73, p = 0.024), females (OR = 0.77, p = 0.034), and non-smokers (OR = 0.78, p = 0.040), and rs2950218 was also associated with a reduced PTB risk in individuals over 40 years old (OR = 0.73, p = 0.040), females (OR = 0.72, p = 0.046), and non-smokers (OR = 0.72, p = 0.011).ConclusionC5orf64 polymorphisms, particularly rs12518552 and rs2950218, are associated with a reduced risk of PTB. These findings suggest that C5orf64 polymorphisms contribute to genetic susceptibility to PTB, with implications for PTB targeted screening and personalized therapeutic strategies.

Instance segmentation plays an important role in the automatic diagnosis of cervical cancer. Although deep learning-based instance segmentation methods can achieve outstanding performance, they need large amounts of labeled data. This results in a huge consumption of manpower and material resources. To solve this problem, we propose an unsupervised cervical cell instance segmentation method based on human visual simulation, named HVS-Unsup. Our method simulates the process of human cell recognition and incorporates prior knowledge of cervical cells. Specifically, firstly, we utilize prior knowledge to generate three types of pseudo labels for cervical cells. In this way, the unsupervised instance segmentation is transformed to a supervised task. Secondly, we design a Nucleus Enhanced Module (NEM) and a Mask-Assisted Segmentation module (MAS) to address problems of cell overlapping, adhesion, and even scenarios involving visually indistinguishable cases. NEM can accurately locate the nuclei by the nuclei attention feature maps generated by point-level pseudo labels, and MAS can reduce the interference from impurities by updating the weight of the shallow network through the dice loss. Next, we propose a Category-Wise droploss (CW-droploss) to reduce cell omissions in lower-contrast images. Finally, we employ an iterative self-training strategy to rectify mislabeled instances. Experimental results on our dataset MS-cellSeg, the public datasets Cx22 and ISBI2015 demonstrate that HVS-Unsup outperforms existing mainstream unsupervised cervical cell segmentation methods.

The coexistence of gas and spontaneous coal combustion poses a serious threat to mine production safety. Therefore, studying the compound disaster of gas and spontaneous coal combustion and revealing the impact of gas concentration on the characteristic parameters of coal spontaneous combustion are of great significance for understanding the mechanism of compound disasters involving coal and gas and for comprehensive management. Based on theoretical analysis, this paper selects four typical types of cannel coal,non-caking coal,meager coal,anthracite coal as research subjects, and conducts programmed heating experiments on coal samples under different gas concentration conditions, aiming to provide a theoretical basis for the prevention and control of spontaneous coal combustion in a gas environment. Experimental results show that with the increase in temperature of the coal samples, the oxygen consumption rate and heating emission intensity exhibit a trend of slow increase followed by rapid increase. With the increase in gas concentration in the environment, the production of CO and CO2, oxygen consumption rate, and heat emission intensity of coal spontaneous combustion all show a decreasing trend. This is mainly because the high gas concentration occupies the adsorption sites on coal molecules, thereby hindering the reaction between coal and oxygen. Additionally, by calculating the thermodynamic parameters of the coal samples, it was found that the apparent activation energy increases with the increase in gas concentration, further confirming that the increase in gas concentration in the environment significantly inhibits the spontaneous combustion process of coal.

The key challenge in unaligned multimodal language sequences lies in effectively integrating information from various modalities to obtain a refined multimodal joint representation. Recently, the disentangle and fuse methods have achieved the promising performance by explicitly learning modality-agnostic and modality-specific representations and then fusing them into a multimodal joint representation. However, these methods often independently learn modality-agnostic representations for each modality and utilize orthogonal constraints to reduce linear correlations between modality-agnostic and modality-specific representations, neglecting to eliminate their nonlinear correlations. As a result, the obtained multimodal joint representation usually suffers from information redundancy, leading to overfitting and poor generalization of the models. In this paper, we propose a Mutual Information-based Representations Disentanglement (MIRD) method for unaligned multimodal language sequences, in which a novel disentanglement framework is designed to jointly learn a single modality-agnostic representation. In addition, the mutual information minimization constraint is employed to ensure superior disentanglement of representations, thereby eliminating information redundancy within the multimodal joint representation. Furthermore, the challenge of estimating mutual information caused by the limited labeled data is mitigated by introducing unlabeled data. Meanwhile, the unlabeled data also help to characterize the underlying structure of multimodal data, consequently further preventing overfitting and enhancing the performance of the models. Experimental results on several widely used benchmark datasets validate the effectiveness of our proposed approach.

The training of sound event detection (SED) models remains a challenge of insufficient supervision due to limited frame-wise labeled data. Mainstream research on this problem has adopted semi-supervised training strategies that generate pseudo-labels for unlabeled data and use these data for the training of a model. Recent works further introduce multi-task training strategies to impose additional supervision. However, the auxiliary tasks employed in these methods either lack frame-wise guidance or exhibit unsuitable task designs. Furthermore, they fail to exploit inter-task relationships effectively, which can serve as valuable supervision. In this paper, we introduce a novel task, sound occurrence and overlap detection (SOD), which detects predefined sound activity patterns, including non-overlapping and overlapping cases. On the basis of SOD, we propose a cross-task collaborative training framework that leverages the relationship between SED and SOD to improve the SED model. Firstly, by jointly optimizing the two tasks in a multi-task manner, the SED model is encouraged to learn features sensitive to sound activity. Subsequently, the cross-task consistency regularization is proposed to promote consistent predictions between SED and SOD. Finally, we propose a pseudo-label selection method that uses inconsistent predictions between the two tasks to identify potential wrong pseudo-labels and mitigate their confirmation bias. In the inference phase, only the trained SED model is used, thus no additional computation and storage costs are incurred. Extensive experiments on the DESED dataset demonstrate the effectiveness of our method.

To investigate the suppression effect of fluorine-containing explosion suppressants (FESs) on methane (CH 4 ) explosions and the corresponding suppression mechanism, a 20-L spherical explosion testing device was used to analyze and compare the effects of perfluorohexanone (C 6 F 12 O) and heptafluoropropane (C 3 F 7 H) on CH 4 explosion parameters. The explosion limit and pressure were obtained before and after adding these FESs. CH 4 explosion triangles under the action of the FESs were drawn, and the explosion hazard degree ( F ) was determined. The results reveal that the addition of C 6 F 12 O and C 3 F 7 H reduces the upper explosion limit (UEL) and lower explosion limit (LEL) of CH 4 , with increased concentrations of C 6 F 12 O and C 3 F 7 H causing faster decreases in the UEL than in the LEL. The explosion limit range of CH 4 with C 6 F 12 O is smaller than that with C 3 F 7 H. When the concentration of CH 4 is low, increases in the C 6 F 12 O and C 3 F 7 H concentrations first cause an increase and then a decrease in the explosion pressure, and the rise rate of explosion pressure increases. After adding the aforementioned FESs, F first increases and then decreases. When the concentration of CH 4 is high, the two FESs only inhibit the explosion pressure. Compared with C 3 F 7 H, C 6 F 12 O has a stronger inhibitory effect; C 6 F 12 O has a smaller CH 4 explosion triangle area, higher critical oxygen concentration (18.27% vs. 17.70%), and lower F value. Both FESs have promotion and inhibition effects on CH 4 explosion because of the comprehensive physical and chemical effects on CH 4 explosions; however, the performance of C 6 F 12 O is superior to C 3 F 7 H.

An innovative simplified method for incorporating the behaviour of structural joints in structural analyses of steel parking-structure is proposed in this study. The method is based on the conventional mechanical modelling of joints through special springs that accommodate the arching effect and moment-axial force (M-N) interaction. The characterization of these springs in terms of stiffness and resistance is performed using analytical formulae. The accuracy of the proposed method is validated through comparisons to results from advanced numerical methods conducted on a multi-scale approach. Its implementation in sub-structure models is then demonstrated, with the main advantage of extremely reducing the modelling and computational time. Validated sub-structure models are finally employed to investigate the influence of the joint properties on the response of parking structures under vehicle collision scenarios. The results show that considering the rigid and semi-rigid joints in the research are characterised by the welded and bolted connections in the research, for a structure built in a courtyard, bolted connections (semi-rigid joints) are recommended, while for the same structure built near an urban road, welded connections (rigid joints) should be considered to enhance the overall robustness.

To analyse the active groups in long -flame coal and its heat transport characteristics under oxygen -limited, the Fourier transform infrared spectrometer and laser flash apparatus were selected, and active groups and thermosphysical parameters of coal were received during 30-300 degrees C at 5, 7, 10, 13, 17, and 21 vol%, the correlation among them under oxygen -limited were obtained by grey correlation method. The results showed that the contents of -COOH, aromatics -CH, and -CH 2 structures were larger than other functional groups at 13 vol%. With rising temperature, an increase in oxygen concentration promoted the reduction of intermolecular hydrogen bonding, -CH 3 , and -CH 2 . Compared with content of intermolecular hydrogen bonding at 30 degrees C, its maximum reduction at 5, 7, 10, 13, 17, and 21 vol% reached 15.75%, 20.56%, 24.37%, 23.02%, 25.73%, and 30.39% at 300 degrees C. Meanwhile, with elevating oxygen concentration after 210 degrees C, thermal diffusivity and thermal conductivity gradually enlarged, which was mainly influenced by -C - - C- and -COOH. The specific heat capacity at 13 vol% was larger than that under other oxygen concentrations, which was mainly influenced by C-O, -COOH, and -C -- C-. Those findings can provide a theoretical basis for the control of underground coal spontaneous combustion.

Domain shift poses a significant challenge in speaker verification, especially in open-set scenarios where the speaker categories are disjoint between the source and target domains. To alleviate the domain shift, traditional domain adaptation methods typically align the source and target distributions in the speaker embedding space, but this may cause the overlap of embeddings from different speakers. To address this problem, this paper proposes to perform the domain alignment in a novel distance metric space, where the source and target domains exhibit the shared within-speaker and between-speaker categories. Thus, the discrepancy between the source and target domains arises only from the domain shift. We refer to the proposed method as Cross-Domain Distance Metric Adaptation (CDMA), in which the within- and between-speaker distance distributions in the target domain are aligned with the source distance distributions and further separated to minimize their overlap. This alignment and separation require estimating the within- and between-speaker distance distributions based on speaker labels, which are unavailable in the unlabeled target domain. Thus, we further propose a learnable speaker clustering method called Graph Convolutional Network with Graph Pruning (GCN-GP). This method generates high-quality pseudo-labels to estimate the two distance distributions in the target domain. Experimental results demonstrate that our method achieves state-of-the-art performance on the FFSVC2022 and VOiCES datasets.

Prickle is a sharp protrusion that covering plant and fruit calyxes of eggplant are considered undesirable agronomic traits, for they bring troubles and create additional costs for farmers. However, little is known about its regulatory genes and molecular mechanism of morphogenesis. In this study, two eggplant inbred lines, prickly '140' and prickleless '145' were applied to construct F1, BC1, and F2 offspring populations. Genetic analysis results showed that prickle's absence/presence on various organs in eggplant was controlled by only one dominant nuclear gene. The PRICKLE LOCUS (Pl) was fine mapped into a candidate interval with 28.3 Kb on chromosome 6 ultimately by adopting bulked segregant analysis combined with genome walking strategy. An auxin response factor SmARF18 (Smechr0602826.1) was deduced to be a candidate gene encoded by Pl, which possesses a non-synonymous single nucleotide polymorphism co-segregating with prickle phenotype in F2 population. The finding here could provide a basis to reveal the molecular regulatory mechanism of prickle morphogenesis in plants and breed prickle-free eggplant cultivars.