Cortisol metabolites are a class of steroid substances that are secreted by the adrenal cortex in the face of various environmental stressors to regulate stress behavior and homeostasis in animals. In recent years, non-invasive methods of collecting fecal samples for analysis of cortisol metabolites have been widely used in non-human primate studies, since direct blood measurements of cortisol metabolites in wild animals are often not feasible. In this paper, the research progress of fecal cortisol metabolites in primates in recent years is reviewed, mainly, focusing on the relationship and role of cortisol in primate reproduction, intestinal microflora, social behavior and other aspects. At the same time, the detection methods of fecal cortisol metabolites are briefly summarized, and the potential application of detection of fecal cortisol metabolites in primates is pointed out, aiming to provide basic knowledge for further study of fecal cortisol metabolites in different primate populations, and theoretical reference for future research.

The popularization of pure electric vehicles still faces the challenges of uneven charging infrastructure layout and low service efficiency, and the load impact formed by large-scale disorderly charging will lead to voltage shift and increased network loss in the distribution network. As an important application type of pure electric vehicles, electric taxi charging demand is frequent and has the potential for regulation. In this paper, we study the scheduling process of fast charging, taking into account the scheduling feasibility of individuals through time and space aspects, introduces the fast charging virtual load to realize the dynamic change of charging reservation mechanism, and establishes a multi-objective optimization model by using the grid load profile situation and the fast charging monetary cost. At the same time, we also investigate the compensation mechanism based on the value of the power deviation, and the spatial load scheduling by taking into consideration of the balance of the utilization rate of the charging station and the fast charging time cost. In view of the defects of the classical Particle Swarm Optimization (PSO) algorithm, such as premature convergence of particles and that it is easy to fall into the local optimal solution, we propose the Genetic-Particle Swarm Optimization of Normal Distribution Decay Inertia Weight (NDGAPSO) by combining with the crossover mutation mechanism. The overall performance of the NDGAPSO algorithm is proved to be better than other improved PSO algorithms through simulation experiments in terms of solution quality, convergence performance, and running speed. Finally, the algorithm is used to solve the fast charging scheduling model, and the experiment proves that the scheduling optimization research in this paper can effectively take into account the interests of electric taxi charging users and power grid operators.

With the large-scale integration of distributed photovoltaic power into distribution networks, issues such as frequent changes in power flow direction and increasingly complex topological structures have emerged. Meanwhile, existing intelligent algorithms suffer from drawbacks such as being prone to getting stuck in local optima and having poor iterative performance. To address these challenges, an improved Zebra Optimization Algorithm (IZOA) is proposed, in which Latin hypercube sampling is employed for the initialization of the zebra population, a dynamic perturbation strategy is introduced during the foraging stage, and a golden sine search mechanism is incorporated in the defense stage, thereby enhancing the algorithm's ability to escape local optima and improve optimization accuracy. Subsequently, a fault current encoding method, switch function, and objective function suitable for multi-source bidirectional power flow are constructed, transforming the fault location problem into an optimization problem. Finally, an improved IEEE33-node distribution network model is established, and the effectiveness of the proposed algorithm is verified through simulation experiments and comparative analysis. The results demonstrate that the IZOA algorithm can rapidly and accurately locate single-point and multi-point fault sections. Compared with other algorithms, it achieves a 12.57% improvement in average accuracy and a 51.5% increase in iteration speed. The algorithm exhibits fault tolerance for minor distortions in fault information, showing significant performance advantages.

Unplanned icing-induced outages of wind turbines during low temperature and cold wave weather lead to severe fluctuations in wind farm power generation, posing significant challenges to the prediction accuracy of traditional models under such extreme operating conditions. To address this issue, this paper proposes a short-term power forecasting method based on heterogeneous ensemble learning. Firstly, it constructs a wavelet transform-driven deep feature extraction network, in which wavelet transform decouples meteorological data into detail components and trend components. These components are processed separately, where Convolutional Neural Networks enhance the capture of spatial local features, while Long Short-Term Memory networks model temporal dependencies, followed by adaptive feature fusion via a cross-attention mechanism. Subsequently, a heterogeneous ensemble strategy builds a Stacking framework combined with Light Gradient Boosting Machine, Extreme Gradient Boosting, and Support Vector Regression serve as diverse Base Learners to fully exploit feature heterogeneity, while linear regression (LR) acts as the meta-learner to optimize prediction accuracy and robustness. Based on the real-world data from a wind farm in Jiangxi province during winter, the proposed method achieves a 4-hour-ahead prediction with mean absolute error, mean squared error, and coefficient of determination of 0.028, 0.119, and 0.618, respectively. The experimental results demonstrate that the proposed model significantly enhances forecasting accuracy under low temperature and cold wave weather conditions compared with both single models and conventional ensemble approaches.

To address the issues of weakened features for distant small-target defects and high model complexity in photovoltaic panel defect detection algorithms, this study proposes an improved lightweight algorithm named FEM-YOLO. Firstly, the C3k2 module is enhanced by integrating FasterBlock and EMA, constructing a C3k2-Faster-EMA structure to improve the network's ability to learn and capture features of defective targets. Subsequently, the Mona module is incorporated into the C2PSA block, optimizing the model's feature extraction and representation capabilities. Moreover, the MLCA mechanism is integrated into the backbone network to enhance the robustness of feature extraction for diverse targets. Finally, an additional P2 detection layer is added specifically for small targets, and an efficient detection head named EfficientHead is designed. This combination enhances the capability to capture micro-defects while simultaneously reducing model complexity. Experimental results demonstrate that, compared with the original YOLO11n model, the improved algorithm achieves increases of 1.9% in both mAP50 and mAP50-95 metrics. Furthermore, the model parameter count is reduced to 2.1×10⁶ and the model size is compressed to 4.4 MiB. Thus, the proposed FEM-YOLO algorithm significantly enhances detection accuracy while substantially reducing model complexity.

Steel surface defects exhibit diverse morphologies, complex structures, a high proportion of small targets, and susceptibility to interference from environmental factors, while existing defect detection models suffer from complex structures, large parameter counts, and poor detection accuracy and real-time performance. To address these issues, a lightweight and efficient steel defect detection algorithm (MHTD-YOLO11n) based on YOLO11n is proposed in this studyly. Firstly, a multi-scale grouped dilated convolution (MSGDC) module is introduced in this method, in which grouped convolutions with different dilation rates are integrated to achieve multi-scale feature fusion and enhance the detection capability for various types of defects. Subsequently, a Hierarchical Reciprocal Attention Mixer (H-RAMi) module is incorporated to compensate for pixel-level information loss caused by downsampled features. A C2PSA_TPA module is then designed, in which the KV cache size during inference is significantly compressed by leveraging Tensor Product Attention (TPA). Finally, the feature interaction module (C3K2_DFF) is reconfigured to enable the network to effectively combine multi-scale information under a larger receptive field, promoting improvements in both detection accuracy and speed.Experimental results show that compared with the YOLO11n algorithm, the mAP value and recall rate of the MHTD-YOLO11n algorithm are increased by 4.3 and 9.1 percentage points respectively, a detection speed of 258.3 frame/s is achieved, the parameter count and computational volume are reduced by 1.42×10⁶ and 3.4×10⁹ respectively, and the dual requirements of high accuracy and real-time performance in industrial quality inspection scenarios are met.

To address the current problems such as low efficiency of manual garbage collection, high labor costs, and reduced accuracy of garbage detection due to the complexity of the underwater environment, an underwater garbage detection algorithm based on improved RT-DETR network is proposed. Firstly, the lightweight network StarNet is adopted to replace the original backbone network to achieve the simplification of the model. Secondly, a new feature pyramid structure is designed, aiming to enhance the feature information of small targets, replacing the traditional method of adding P2 layers. It also integrates the CSPO (CSP-OmniKernel) module and the SPD convolution module to improve the model's extraction of global features and the fusion of multi-scale features. In addition, the WaveletUnPool module and the LDConv module are introduced to reduce the loss of feature information and optimize the upsampling and downsampling operations to further enhance the accuracy of small target detection. Finally, the Focaler-MPDIoU loss function is designed to replace the loss function of the original model, assigning different weights to samples of different difficulties, which optimizes the accuracy and speed of bounding box regression. The experimental results show that, compared with the original model, the SOP-DETR model has increased the precision rate, recall rate and mAP@0.5 by 7.7, 3.3 and 4.5 percentage points respectively, while reducing the computational load by 30.4%, effectively enhancing the garbage detection performance in the complex underwater environment.

Underwater images often suffer from color distortion and detail loss due to complex environments. Existing methods process color channels uniformly, ignoring their distinct characteristics, while Transformers underperform convolutional networks due to limited information utilization. To address these issues, a multi-scale attention and adaptive normalization-based underwater image enhancement network is proposed. The network consists of two stages: multi-scale feature extraction and feature enhancement with reconstruction. In the first stage, rich features are captured through multi-scale processing. In the second stage, an encoder-decoder structure is employed, incorporating a feature enhancement module and parallel skip connections to ensure feature integrity and structural consistency while maximizing information utilization. Experimental results demonstrate that the proposed network significantly improves color correction and detail preservation compared with existing methods, achieving superior qualitative and quantitative performance.

Recommendation systems are a crucial means to address the problem of information overload, and knowledge graphs serve as the most widely used auxiliary information source in recommendation systems. Existing knowledge graph-based approaches suffer from limitations including insufficient dynamic modeling of multi-semantic relationships, lack of modeling for neighbor sequence dependencies and long-range associations, and inadequate capability in extracting multi-level features from graph structures. To tackle these issues, this paper proposes a Multi-Semantic and Temporal Convolutional model for Knowledge Graph recommendation (MSTC-KG). The model dynamically fuses user-relation interactions through a multi-head attention mechanism and a gating mechanism to distinguish the semantic contributions of neighbor nodes, employs GRU to capture the temporal dependencies and long-range associations of neighbor sequences, and utilizes 1D convolutional neural networks to extract local-to-global multi-level features of graph structures. Compared with 14 baseline models on two public datasets, MSTC-KG achieves an AUC of 0.737 and 0.982, and an F₁-score of 0.680 and 0.938 on the Book-Crossing and MovieLens-20M datasets respectively, outperforming all other baseline models. Experiments demonstrate that through the architecture characterized by “dynamic multi-semantic differentiation + temporal dependency capture + hierarchical feature extraction”, the model effectively enhances recommendation accuracy and the capability to model complex graph structures.

This paper investigates the Toda system on connected finite graphs using degree theory. It is proven that all solutions of the Toda system with nonnegative prescribed functions are uniformly bounded, which implies that the Brouwer degree of such a Toda system is well-defined. By the homotopy invariance, the Brouwer degree of the Toda system with nonnegative prescribed functions is shown to be equal to 1. As a corollary, this suggests that the Toda system with nonnegative prescribed functions is solvable.

The communication and collaboration among multiple agents may be affected by random noise. Therefore, this paper investigates the problem of cluster mean square consensus for nonlinear multi-agent systems under fractional Brownian motion disturbances. Firstly, the stochastic noise in the study is modeled as fractional Brownian motion rather than standard Brownian motion. Secondly, the cluster mean square consensus of the nonlinear multi-agent system is examined, where the case of a single cluster reduces to mean square consensus. Based on distributed control theory, stochastic analysis theory, graph theory, and other frameworks, an infinitesimal operator is constructed, and a novel Lyapunov functional with a double-integral form is designed. A controller with time-varying control gains is developed, and sufficient conditions for achieving cluster mean square consensus of the system are derived. A numerical example is provided to validate the correctness of the conclusions and the effectiveness of the proposed method.

Bipartite-partial component consensus refers to a group dynamic behavior in a multi-agent system where some components of the state variables of some agents tend to be consistent over time, while the corresponding components of the state variables of the remaining agents tend to have opposite values over time. This paper for the first time explores the bipartite-partial component consensus problem for nonlinear leader-following multi-agent systems in directed signed networks via event-triggered control. By designing a novel event-triggered control protocol and an event-triggering function that dynamically adjusts according to the system state, the communication frequency among agents and the update frequency of controllers can be significantly reduced while guaranteeing desired system performance. Then, the sufficient conditions to achieve bipartite-partial component consensus are derived for the multi-agent systems, and it is proved that the multi-agent systems can effectively avoid the Zeno phenomenon under the designed event-triggered control protocol. Finally, a specific simulation example is used to verify the reliability of the article’s conclusion and the effectiveness of saving communication resources.

This study characterizes Bacillus velezensis YWC-01, a strain isolated from Daqu with biodegradative activity against deoxynivalenol (DON), and evaluates its probiotic potential through whole-genome sequencing and biological trait analysis. Primary and secondary screening via dilution-plate streaking isolated the strain from Daqu, followed by identification. Whole-genome sequencing employed Illumina NovaSeq and PacBio Sequel platforms. DON degradation efficiency was quantified using high-performance liquid chromatography (HPLC). Genomic functional annotation utilized multiple databases. Metabolic traits were assessed by predicting carbohydrate-active enzyme (CAZy) families and secondary metabolite gene clusters, alongside virulence factor screening and antibiotic resistance mechanism evaluation. Phylogenetic analysis based on gyrA gene elucidated evolutionary relationships. The results showed that, YWC-01 degraded DON at a maximum rate of 29.1% (initial concentration: 140 mg/L). Its genome spanned 4.04 Mb with 46.51% G+C content, encoding 3 928 genes (including 87 tRNAs and 131 CAZy families). AntiSMASH predicted 13 secondary metabolite gene clusters, including those for macrolactin H, bacillaene, and fengycin synthesis. YWC-01 exhibits limited pathogenicity and harbors genetic potential for diverse functional metabolites, supporting its utility as a probiotic agent.

Despite extensive use of insect mitochondrial genomes in phylogenetic and evolutionary studies, large-scale and systematic comparative analyses of protein-coding genes (PCGs) in Orthoptera remain limited. In this study, 13 mitochondrial PCGs from 404 Orthoptera species are analyzed, including representatives from both Caelifera and Ensifera, to investigate sequence length variation, nucleotide composition, codon usage patterns, and nucleotide substitution rates. The results revealed a clear AT bias across all PCGs, with ATP8 and ND6 exhibiting the highest levels of sequence variability, while COX1 was the most conserved. Start codons were predominantly ATN types, and TAA was the most frequently used stop codon, with incomplete stop codons (T or TA) widely observed. Codon usage bias analysis showed a strong preference for synonymous codons ending in A or U, with relative synonymous codon usage (RSCU) and effective number of codons (ENC) indicating substantial codon usage bias driven primarily by natural selection rather than mutation pressure. Furthermore, Ka/Ks analysis demonstrated that all PCGs were subject to strong purifying selection, with COX1 evolving at the slowest rate and ND4L at the fastest. These findings provide new insights into the molecular evolution of Orthoptera mitochondrial genomes and offer a valuable foundation for further research on insect phylogeny, molecular adaptation, and evolutionary genetics.

Antimicrobial resistance (AMR) has emerged as a critical global public health challenge. Vibrio cholerae, a major pathogenic bacterium, is exhibiting increasing resistance to antibiotics, complicating cholera control efforts worldwide. In this study, a comprehensive bioinformatic analysis of the distribution, abundance, diversity, and mobility of antimicrobial resistance genes (ARGs) in V. cholerae based on global genomic datasets is performed. Screening and comparative analysis of 81 132 V. cholerae genome sequences led to the identification of 16 570 ARG sequences, encompassing 83 ARG subtypes associated with 19 antibiotic classes. Notably, genes conferring resistance to multiple drugs, tetracyclines, macrolides, and sulfonamides were the most prevalent, accounting for 69.23% of total ARG abundance. The diversity and abundance of ARGs exhibited significant geographic heterogeneity, with markedly higher levels observed in middle-and low-income countries such as China, India, and Haiti, compared with those in high-income countries. Strikingly, 91.57% of ARG subtypes were associated with mobile genetic elements (MGEs), highlighting their central role in the acquisition and dissemination of resistance. Cluster analysis further identified 120 mobile ARG operational taxonomic units, 26.67% of which were detected in at least three countries. ARGs conferring resistance to aminoglycosides, chloramphenicol, and sulfonamides displayed widespread geographic distribution. Representative genes such as dfrA1, APH(6)-Id, and cat3 have undergone intercontinental dissemination, predominantly across Asia and Africa. This study provides critical insights into the AMR landscape and transmission mechanisms of V. cholerae, offering a scientific basis for the development of effective surveillance and control strategies to mitigate the threat of resistance in V. cholerae and other pathogens.

To improve the freeze-drying quality of fresh Dendrobium huoshanense, this study explored the effects of calcium chloride (CaCl₂) immersion pretreatment on the physical properties, active components, and underlying mechanisms of freeze-dried samples. The results showed that appropriate CaCl₂ treatment enhanced the moisture characteristics, texture, color, and polysaccharide content of freeze-dried samples. Among all treatments, the 3.0% CaCl₂ group exhibited the best performance, with a moisture content of 7.40%±0.19%, rehydration ratio of 3.02±0.03, hardness of 2.70±0.23 N, crispness of 1.34±0.11 N·S, and a minimal color difference (ΔE) of 2.02±0.07. Its polysaccharide content reached 32.47%±0.73%, an increase of 10.3% was made compared with the untreated control. Subsequently, the differences between extracted dendrobium pectin and calcium pectate prepared by reacting with CaCl₂ were analyzed. FT-IR analysis revealed a shift of the characteristic carboxylate peak from 1 591.2 cm^-1 to approximately 1 621.1 cm^-1, and NMR spectra showed significant changes in proton signals. The weight-average molecular weight of pectin (414 777 g/mol) increased gradually with CaCl₂ concentration (1.5%-6.0%) to 529 211 g/mol. SEM observations showed that the microstructure of calcium pectate changed from loose and collapsed to dense and compact. These findings indicate that Ca²⁺ forms a stable “egg-box” crosslinking structure with pectin in the cell wall, which is the core mechanism for quality improvement. This study provides theoretical and technical support for the high-quality and value-added processing of Dendrobium huoshanense.

Using a combination of multiple column chromatography and high-performance liquid chromatography separation techniques, the ethyl acetate fraction of Ferula conocaula Korovin was isolated and purified. Modern spectroscopic methods were employed to structurally characterize the obtained monomeric compounds. This study successfully isolated 19 monomeric compounds from the ethyl acetate fraction of Ferula conocaula Korovin. including 4 polyphenolic monomers, 14 caffeoylquinic acid monomers, and 1 flavonoid monomer. They are identified as caffeic acid (1), caffeic acid methyl ester (2), 3,4-dihydroxybenzaldehyde (3), 5-O-caffeoylquinic acid (4), 1-O-caffeoylquinic acid (5), chlorgenic acid (6), methyl chlorogenate (7), 3-O-trans-feruloylquinic acid (8), methyl 5-O-p-coumaroylquinic acid (9), methyl 5-O-feruloylquinate (10), acanthopacid G (11), 1-(3,4-dihydroxycinnamoyl) cyclopentene-2,3-diol (12), 4,5-di-O-caffeoylquinic acid 1-methyl ether (13), 4-O-caffeoyl-5-O-feruloylquinic acid methyl ester (14), methy 3,5-dicaffeoylquinate (15), 1,5-di-O-caffeoylquinic acid (16), 1,3-di-O-caffeoylquinic acid (17), methyl 4-O-E- feruloyl 5-O-E-caffeoyl-quinate (18), and luteolin 7-O-glucoside (19). It is worth nothing that compounds 1-6 and 8-19 were isolated from this plant for the first time. Glucose uptake activity assay experiments showed that compounds 7, 13, 15, and 16 had strong blood glucose lowering effects.

Heavy metal pollution is widespread and increasingly severe, cadmium (Cd) in water is potentially toxic, and can be transferred and enriched into the human body through the food chain and other ways, endangering human health. Adsorption method has garnered significant attention in the treatment of heavy metal pollution, and different biomass materials can affect the adsorption effect of biochar on heavy metals in water to a certain extent. In this study, magnetic pomelo peel biochar (ZMF@BC) was prepared from agricultural waste pomelo peel as raw material via zinc-manganese-iron impregnation modification to investigate the removal effect and mechanism of biochar on Cd(Ⅱ) in water under different conditions. The adsorption performance of ZMF@BC on Cd(Ⅱ) was analyzed by coexisting ion competition adsorption, adsorption-desorption, etc. The adsorption mechanism was also explored using batch adsorption experiments, Fourier Transform Infrared Spectroscopy (FT-IR), and X-ray Diffraction Analysis (XRD). The results showed that at pH 6.0, the biochar adsorbed Cd(Ⅱ) with a maximum adsorption capacity of 26.3 mg/g due to the reduced ionic competitiveness, and the competitive adsorption of Cd(Ⅱ) by common cations coexisting in the water was in the following order: Na⁺> Mg²⁺> Ca²⁺> K⁺. The adsorption of Cd(Ⅱ) by ZMF@BC was in accordance with Langmuir and quasi-secondary kinetic models, suggesting monolayer chemical adsorption. The thermodynamic results showed that the adsorption process was spontaneous, and the adsorption-desorption could maintain 85.1% of the initial adsorption amount after five adsorption-desorption cycles. ZMF@BC can absorb Cd(Ⅱ) through ion exchange by using the Zn, Fe, and Mn in the metal oxides on its surface, and also be immobilized through the complexation mechanism specific to the iron oxides.

Nitrate nitrogen pollution in water bodies is becoming increasingly severe, posing a threat to human health. Sulphur-autotrophic denitrification (SADN) technology, with its high efficiency and low carbon footprint, has been widely applied for the removal of nitrate nitrogen from groundwater. Groundwater often contains high concentrations of divalent manganese (Mn²⁺), and the impact of Mn²⁺ on SADN remains unclear. This study conducted batch experiments using sulfur (S⁰) as an electron donor to analyse the effects of different Mn²⁺ concentrations on SADN performance and to investigate the response patterns of microbial community structure, nitrogen cycling, and sulfur oxidation functional genes to Mn²⁺ in the system. The results showed that low concentrations of Mn²⁺ (≤1 mmol/L) significantly promoted the SADN process, with the highest nitrate nitrogen removal rate reaching 0.735 mmol/(L·d). The proportion of Thiobacillus, the dominant microorganism in the SADN process, increased from 9.41% to 14.92%, and the relative expression levels of denitrification genes nirS and nosZ, as well as sulfur oxidation genes dsrA and soxB, also significantly increased. However, high concentrations of Mn²⁺ (>4 mmol/L) inhibited the SADN process, reducing the nitrate nitrogen removal rate to 0.674 mmol/(L·d). The proportion of SADN microorganisms Thiobacillus and Longilinea in the system decreased from 23.11% to 15.60%, and the expression of denitrification genes narG, nirS, nirK, norB, and sulfur oxidation genes dsrA were also inhibited. In summary, Mn²⁺ affects the denitrification performance of the SADN system by regulating functional genes and microbial community structure in the SADN system, in which nitrite nitrogen reduction as well as S⁰ and thiosulfate oxidation were important rate-limiting processes in SADN at different Mn²⁺ concentrations. In this paper, the response mechanism of Mn²⁺ to the SADN process was revealed at the level of community structure and functional genes, which can provide a theoretical basis for the application of SADN technology in the removal of nitrate nitrogen from Mn²⁺-containing groundwater.