Browse Articles

Discover research articles across all indexed journals

Cultivating key disciplinary competencies among university students against the background of China’s new engineering education initiative

PLoS ONE Liyuan Zheng, Di Ge Feb 14, 2025 DOI: 10.1371/journal.pone.0318561

The development of students’ core disciplinary competencies is crucial to the quality of talent cultivation against the background of China’s New Engineering Education initiative. We condensed a literacy–ability–knowledge model of the core competence structure of university students based on the results of a survey and then applied this model with a special focus on competence structures. We coded the core competency indicators of the course area using a software engineering course as an example to construct a map of the core competencies of the course (area). New teaching and instruction methods to promote the development of students’ core competencies were then explored. Finally, to shape core subject competence, we suggest building a course recommendation system that aligns with an employment guidance service based on the certification of core subject competence to realize a new mode of cultivating competence.

Barriers to participation in Iran’s Integrated Geriatric Care Program: A qualitative study

PLoS ONE Neda Sadat Nazaripanah, Haidar Nadrian, Vali Bahrevar et al. Feb 14, 2025 DOI: 10.1371/journal.pone.0315034

Background A significant number of older adults still have unfulfilled needs related to healthcare and social services. Policymakers and health administrators can turn this challenge into a golden opportunity through proper planning and policymaking. This study aimed to identify barriers to participation in Iran’s “Integrated Geriatric Care Program (IGCP)”. Methods A qualitative study with conventional content analysis approach was conducted in Yazd City, Iran, from January to March 2023. A purposive sampling method with maximum variation was used. Nineteen healthcare providers and community-dwelling older adults participated in the study. Data were collected using semi-structured in-depth interviews until data saturation and were concurrently analyzed. Results Nine older persons (3 women and 6 men) with a Mean±SD age of 71.33±5.75 and ten healthcare providers (6 women and 4 men) with a Mean±SD age of 38.2±9.16 participated in the study. Two main categories, with ten sub-categories, emerged from the data analysis. The main categories were “personal challenges of older adults” and “structural defects of Integrated Geriatric Care Program”. Conclusion The participation of community-dwelling older adults in the Iranian IGCP is constrained by a combination of personal and structural factors. Our findings seem to be informative for health policymakers and health practitioners who aim to enhance the quality and quantity of healthcare services for older adults in developing communities.

Wave packet splitter and rectificator based on the Dirac delta potentials in a fractional medium

Scientific Reports M. Solaimani Feb 14, 2025 DOI: 10.1038/s41598-025-87001-y

Cloning and bioinformatics analysis of key gene ShOMT3 of podophyllotoxin biosynthesis pathway in Sinopodophyllum hexandrum

PLoS ONE Wei Liu, Haona Gao, Dan Zhao et al. Feb 14, 2025 DOI: 10.1371/journal.pone.0314919

Sinopodophyllum hexandrum (S. hexandrum) is an endangered traditional Chinese medicine as abundant podophyllotoxin with powerful anticancer activity. In this study, the rootstalks of S. hexandrum from different geographical locations in China [S1 (Gansu) and S2 (Shaanxi)] were used as research materials to clone the key gene pluviatolide O-methyltransferase 3 (ShOMT3) in the podophyllotoxin biosynthetic pathway. Subsequently, bioinformatics analysis of the ShOMT3 gene and its encoded protein was subjected to bioinformatics analysis using various analysis software including ProtParam, DeepTMHMM, SubLoc, Signal-P 5.0, and Swiss-model. The results of the analysis revealed that the CDS region of the ShOMT3 gene is 1119 bp long, encoding 372 amino acids. The theoretical molecular weight of the ShOMT3 protein is 41.32784 kD, and the theoretical isoelectric point (pI) is 5.27. The instability coefficient of the protein is 46.05, the aliphatic index is 93.58, and the grand average of hydropathicity (GRAVY) is 0.037, indicating that it is an unstable hydrophobic protein. The protein does not contain transmembrane domains or signal peptides, indicating that it is a non-secreted protein. Secondary structure prediction results suggests that the protein consists of alpha helices, random coils, extended strands, and beta-turns. Tertiary structure prediction results suggests that the protein functions as a monomer. In the phylogenetic tree, the ShOMT3 protein has the highest homology with Podophyllum peltatum (P. peltatum). The successful cloning and bioinformatics analysis of the ShOMT3 gene provide theoretical basis and excellent genetic resources for the molecular regulatory mechanism analysis of the podophyllotoxin biosynthetic pathway and molecular breeding in S. hexandrum.

Deciphering the code of resistance: a genomic and transcriptomic exploration of the Cystoisospora suis Holland-I strain

Scientific Reports Teresa Cruz-Bustos, Thomas Eder, Baerbel Ruttkowski et al. Feb 14, 2025 DOI: 10.1038/s41598-025-89372-8

Abstract Cystoisospora suis, a member of the apicomplexan order Coccidia and causative agent of neonatal porcine coccidiosis, poses a challenge to pig production due to the emergence of reduced efficacy of toltrazuril, the only EU-approved treatment. To address the critical gaps in understanding toltrazuril resistance and possibilities of early diagnostics, our study investigated the genetic basis of resistance through whole-genome DNA sequencing and transcriptome analysis of two C. suis strains, the toltrazuril-susceptible Wien-I and the resistant Holland-I. Additionally, we studied the mitochondrial genome and analysed mitochondrial gene expression in both strains. Our results show that genes encoding proteins involved in host-cell invasion displayed variable expression patterns and genetic mutations, suggesting adaptive changes in invasion mechanisms. Moreover, substantial fluctuations in the expression of genes linked to retrotransposons, accompanied by genetic alterations, were observed, highlighting their potential involvement in genomic rearrangements. Finally, our mitochondrial genome analyses revealed important insights into its genetic organization and conservation. Notably, the marked downregulation of CoI, CoIII and Cytb mRNA levels in the resistant strain Holland-I upon toltrazuril exposure highlights the dynamic response of mitochondrial genes to toltrazuril. These mitochondrial adaptations appear to be closely linked to the parasite drug resistance mechanism, potentially facilitating its survival under pharmacological stress. These findings enhance our knowledge of drug resistance mechanisms in Coccidia and highlight the need for novel management strategies, leading to the development of targeted treatments and controls.

Study on lean production management of new energy vehicle body painting based on the dual perspectives of digital transformation and VSM

PLoS ONE Qin Yang, Xizhe Wang, Hu Wu Feb 14, 2025 DOI: 10.1371/journal.pone.0318253

The automobile body painting production system exists in the backward management concept of focusing on production and ignoring environmental protection, focusing on traditional technology and unwilling to innovate, etc., high production cost and environmental protection standards, which have had a negative impact on the industry’s long-term growth. Based on the value stream mapping (VSM) analysis technology, the current situation of the value stream of the body spraying process of TJ Automobile Manufacturing Company is analyzed, the problem of waste in the production process of automobile body painting system affected by Eliminate Combine Rearrange Simplify (ECRS)a four-step problem-solving technique was studied and analyzed, a new method for the application of value stream graph analysis technology in the digital transformation environment is proposed, and a suitable production improvement scheme for the body painting system is formulated by using the new method, which increases the value-added operation time of the painting production process by 9.5% and reduces the non-value-added operation time by 12.5%. The number of operators was reduced by 18.8%, the value-added ratio was increased accordingly, the scrap rate was reduced by 8%, and the exhaust gas purification rate was increased by 66%, which brought good economic and social benefits to the enterprise and provided a successful case for the application of lean production VSM in the digital transformation environment. The paper extends knowledge in the area of digital transformation and lean production by helping us to establish and explain the relationship between lean production management methods and the promotion of new energy vehicle body coating performance from a dual perspective of digital transformation and VSM. None of the previous studies have considered the combined application of the studied digital transformation and VSM lean management approaches.

Individual differences in wellbeing are supported by separable sets of co-active self- and visual-attention-related brain networks

Scientific Reports Yumeng Ma, Jeremy I. Skipper Feb 14, 2025 DOI: 10.1038/s41598-025-86762-w

Abstract How does the brain support ‘wellbeing’? Because it is a multidimensional construct, it is likely the product of multiple co-active brain networks that vary across individuals. This is perhaps why prior neuroimaging studies have found inconsistent anatomical associations with wellbeing. Furthermore, these used ‘laboratory-style’ or ‘resting-state’ methods not amenable to finding manifold networks. To address these issues, we had participants watch a full-length romantic comedy-drama film during functional magnetic resonance imaging. We hypothesised that individual differences in wellbeing measured before scanning would be correlated with individual differences in brain networks associated with ‘embodied’ and ‘narrative’ self-related processing. Indeed, searchlight spatial inter-participant representational similarity and subsequent analyses revealed seven sets of co-activated networks associated with individual differences in wellbeing. Two were ‘embodied self’ related, including brain regions associated with autonomic and affective processing. Three sets were ‘narrative self’ related, involving speech, language, and autobiographical memory-related regions. Finally, two sets of visual-attention-related networks emerged. These results suggest that the neurobiology of wellbeing in the real world is supported by diverse but functionally definable and separable sets of networks. This has implications for psychotherapy where individualised interventions might target, e.g., neuroplasticity in language-related narrative over embodied self or visual-attentional related processes.

Evaluating elexacaftor/tezacaftor/ivacaftor (ETI; Trikafta™) for treatment of patients with non-cystic fibrosis bronchiectasis (NCFBE): A clinical study protocol

PLoS ONE Colin E. Swenson, William R. Hunt, Candela Manfredi et al. Feb 14, 2025 DOI: 10.1371/journal.pone.0316721

Background Non-cystic fibrosis bronchiectasis (NCFBE) is a disease that exhibits dilatation of airways, airflow obstruction, persistent cough, excessive sputum production, and refractory respiratory infections. NCFBE exhibits clinical and pathological manifestations similar to key features of cystic fibrosis (CF) lung disease. In CF, pathogenesis results from dysfunction of the cystic fibrosis transmembrane conductance regulator (CFTR), and diagnosis is made by demonstrating elevated sweat chloride concentrations (typically ≥60 mEq/L), two CFTR mutations known to be causal, multi-organ tissue injury, or combination(s) of these findings. Objective Based on a considerable body of evidence, we believe many patients with NCFBE have disease likely to benefit from drugs such as elexacaftor/tezacaftor/ivacaftor (ETI) that activate CFTR-dependent ion transport. ETI is currently prescribed solely for treatment of CF and has not been adequately tested or proposed for patients with NCFBE, many of whom exhibit decreased CFTR function. Accordingly, we are conducting a clinical trial of ETI in subjects carrying a diagnosis of NCFBE. Methods Participants will exhibit one disease-causing CFTR mutation and/or sweat chloride measurements of 30–59 mEq/L. Cutaneous punch biopsy or blood samples will be obtained for iPS cell differentiation into airway epithelial monolayers–which will then be tested for response to ETI. Each patient will be given CFTR modulator treatment for approximately four weeks, with monitoring of clinical endpoints that include FEV1 (forced expiratory volume in one second), sweat chloride, quality of life questionnaire, and weight. The study will evaluate response of patients with NCFBE to ETI, and test usefulness of iPSC-derived airway epithelial monolayers as a novel in vitro technology for predicting clinical benefit. Trial registration This trial is registered at clinicaltrials.gov (Identifier: NCT05743946. Date: 02/23/2023).

Association of 25-hydroxyvitamin D with Parkinson’s disease based on the results from the NHANES 2007 to 2018 and Mendelian randomization analysis

Scientific Reports Yan Xu, Jie Peng, Xiguo Zhou et al. Feb 14, 2025 DOI: 10.1038/s41598-025-87120-6

A predictive model to assess the risk of developing hyperlipidemia in patients with type 2 diabetes

PLoS ONE Rujian Ye, Xitong Huang, Hehui Yang et al. Feb 14, 2025 DOI: 10.1371/journal.pone.0315781

Background Type 2 diabetes (T2D) is increasingly recognized as a significant global health challenge, with a rising prevalence of hyperlipidemia among diabetic patients. Effectively predicting and reducing the risk of hyperlipidemia in T2D patients to mitigate their cardiovascular risk remains an urgent issue. Objectives The research sought to determine early clinical indicators that could predict the onset of hyperlipidemia in patients with T2D and to establish a predictive model that integrates clinical and laboratory indicators. Methods A cohort of T2D patients, excluding those with pre-existing hyperlipidemia or confounding factors, was analyzed. Clinical and laboratory data were used in a LASSO regression model to select key predictive variables. A nomogram was then constructed and evaluated using receiver operating characteristic (ROC) analysis and calibration. Results Among 269 participants, PCSK9 levels were significantly elevated in T2D patients with hyperlipidemia and exhibited a positive correlation with several lipid markers. LASSO regression identified six predictors: BMI, TG, TC, LDL-C, HbA1c, and PCSK9. The nomogram model exhibited robust predictive performance (AUC, 0.89 (95% CI: 0.802–0.977)) and showed good calibration. Conclusions This method effectively predicts the risk of hyperlipidemia in patients with T2D and provides a valuable tool for early intervention. PCSK9, as a key predictor, highlights its potential role in the pathogenesis of diabetes with hyperlipidemia and offers new avenues for targeted therapy.

Disposcope endoscope guidance versus fiberoptic bronchoscopy guidance for double-lumen tube intubation in patients undergoing thoracic surgery: a randomized controlled non-inferiority trial

Scientific Reports Wencai Jiang, Xinyu Zhou, Xu Zeng et al. Feb 14, 2025 DOI: 10.1038/s41598-025-89945-7

Risk and outcomes of healthcare-associated infections in three hospitals in Bobo Dioulasso, Burkina Faso, 2022: A longitudinal study

PLoS ONE Arsène Hema, Satouro Arsène Somé, Odilon Kaboré et al. Feb 14, 2025 DOI: 10.1371/journal.pone.0307346

Background Healthcare-associated infections (HAIs) are one of the most common adverse events in healthcare and represent a major public health problem. The present study was conducted to analyze the incidence, risk factors, and outcomes of HAIs through active surveillance in three hospitals in the city of Bobo Dioulasso, Burkina Faso. Methods A prospective, longitudinal, multicenter study was conducted from May 1 to November 30, 2022, in two district hospitals (Do and Dafra) and the Sourô Sanou Teaching Hospital (CHUSS), Burkina Faso. Consenting patients hospitalized for reasons other than infection, cancer, immunosuppression in the postoperative care ward of Do or of Dafra district hospitals, intensive care unit (ICU)/CHUSS, neonatal ward/CHUSS, and gynecology and obstetrics postoperative care ward/CHUSS during a 2-month inclusion period in district hospitals and 4 months for CHUSS wards. For this study, we used the operational definitions of the French Technical Committee for Nosocomial Infections and Healthcare-associated Infections, with slight modifications. Logistic regression was used to analyze risk factors of HAIs. Results Of the 664 patients enrolled, 166 experienced an HAI, with a cumulative incidence rate of 25% (95%CI: 21.7%–28.3%) or an incidence density rate of 36.7 per 1000 patient-days (95% CI: 31.7–42.9). Surgical site infections (SSI) (44%), followed by neonatal infections (42%) were the most common HAIs. Enterobacterales represented 60% of the bacteria identified in HAIs, and 38.9% of them were extended spectrum β-lactamase (ESBL) producers. Factors associated with HAIs were admission in the neonatal ward (aOR = 6.6; 95%CI:1.1–40.2), ICU (aOR = 3.3; 95%CI:1.3–8.5), previous hospital stay longer than two days (aOR = 2.0; 95%CI:1.2–3.3), or male sex (aOR = 1.8; 95%CI:1.1–3.0). In addition, HAIs were associated with longer follow-up, hospitalization, and mortality (18.1%; 95% 95%CI:12.1–24.4). Deaths were only recorded in the ICU and neonatal ward, with case fatality rates of 45.4% (95% 95%CI: 27.5–63.4) and 21.4% (95% 95%CI: 11.6–31.3), respectively, p = 0.019. Conclusions The incidence of HAIs was relatively high in the three hospitals in Bobo Dioulasso, Burkina Faso. It is essential to implement rigorous protocols for patient management, to reduce the incidence of HAIs and the spread of resistant pathogens in general and Enterobacterales in particular.

Experimental investigation on water retention property of rubble improved by modified hydrophilic polyurethane

Scientific Reports Zhaozhao Liu, Qun Chen, Lu Li et al. Feb 14, 2025 DOI: 10.1038/s41598-025-89239-y

Research on the aesthetic sensitivity evaluation of tourism mascots based on semantic differential method

PLoS ONE Jing Wang, Wenting Yao, Sheng Su et al. Feb 14, 2025 DOI: 10.1371/journal.pone.0318715

In order to better meet the emotional needs of consumers and enhance the design of mascots that integrate multicultural elements, this study uses sensory engineering theory to explore the correlation between emotional needs and aesthetic forms. Initially, tourism mascot forms were collected through market surveys, and then these forms were synthesized into test samples. At the same time, expert interviews were conducted to determine the adjective pairs that can accurately capture consumers’ emotional needs. Subsequently, tourists who came to Xi’an were invited to participate in a satisfaction survey. Using the semantic differential methods, representative adjective pairs were selected from five dimensions: market positioning, market trends, color decoration, stylistic features, and psychological perception as emotional evaluation criteria. Each tourism mascot was evaluated and rated according to these criteria. SPSS software was then used to conduct a one-way ANOVA on the evaluation scores to explore the relationship between the aesthetic evaluation of various mascots and consumers’ emotional needs. The experimental results show that the IP image of cultural and creative brands designed with Tang cultural elements is more contagious and avant-garde than the IP image of cultural and creative brands that incorporate Qin cultural elements. These designs can resonate more with young consumers. This study successfully establishes a mapping model that combines consumers’ aesthetic preferences with their emotional needs for tourism mascot design. This study provides valuable guidance for the design of Xi’an tourism mascots and effectively meets consumers’ emotional expectations.

Investigating the presence of different bottlenose dolphin ecotypes in the Mediterranean Sea

Scientific Reports Michela Bellingeri, Alice Nebuloni, Gabriella La Manna et al. Feb 14, 2025 DOI: 10.1038/s41598-025-88605-0

Academic writing challenges and supports for early-stage Chinese postgraduates: A mixed-methods study on teaching-research integration, supervisor engagement, and self-efficacy

PLoS ONE Huanhuan Zhang, XiaoShu Xu, Vivian Ngan-Lin Lei et al. Feb 14, 2025 DOI: 10.1371/journal.pone.0317470

Enhancing academic publication outcomes and writing skills in postgraduates is a key goal for higher education institutions, ideally through early intervention. This study examines the challenges and support needs of early-stage Chinese postgraduates in academic writing, as well as the factors influencing their publication outcomes and writing self-efficacy. Using snowball sampling, 73 postgraduates were surveyed through questionnaires and reflective journals. Thematic analysis and linear regression were applied to analyze the data. Findings reveal primary challenges in literature review, research methods, data analysis, and topic selection. Key support needs include supervisor guidance, knowledge resources, tools, and self-assessment. Results indicate that the year of study, meeting frequency, supervisor support, teaching-research nexus, and self-efficacy do not significantly impact publication outcomes. However, supervisor support and the quality of the teaching-research nexus positively influence self-efficacy. These findings highlight the importance of providing targeted support and fostering intrinsic motivation to enhance the writing skills and self-efficacy of early-stage postgraduate students.

Simultaneous distributed generation and electric vehicles hosting capacity enhancement through a synergetic hierarchical bi-level optimization approach based on demand response and Volt/VAR control

Scientific Reports Zenhom M. Zenhom, Shady H. E. Abdel Aleem, Essam Aboul Zahab et al. Feb 14, 2025 DOI: 10.1038/s41598-025-88635-8

Abstract In the context of sustainable development, electric vehicles (EVs) and renewable-based distributed generation (RDG) integration into the distribution networks (DNs) introduce various merits. They involve lowering harmful emissions, employing various types of energy sources, and encouraging the dependence of renewable energy. However, the two most challenging issues posing a grave danger to the technical operation of the DN would be the inappropriate integration of RDGs and EVs. Consequently, in order to guarantee safe operation, distribution system operator (DSO) is responsible for precisely identifying two crucial terms, distributed generation hosting capacity (DG-HC) and electric vehicles hosting capacity (EV-HC). Despite an increase in the amount of research on HC approaches, there is still a remarkable research gap in the discussion of models that effectively combine demand response (DR), smart inverters (SI) Volt/VAR control, DG-HC, and EV-HC targets simultaneously. This study offers a hierarchical bi-level optimization HC framework depends on both dynamic tariff-based DR and SI Volt/VAR control. In the lower layer, Participating customers’ load curves and EV aggregators’ charging demands are optimally adjusted, on a forecast basis, based on the proposed dynamic tariff. Nonetheless, the suggested multi-objective function—which includes DG-HC, EV-HC maximization, and loss minimization—is optimized by the DSO at the upper layer based on all of these optimal load curves. To further support the proposed objective function, DR is mixed with the optimal Volt/VAR controlling offered by SIs. In addition, the role of grid-connected EVs (GCEVs) on DG-HC increase is revealed with both uncoordinated and coordinated charging schemes. To check the proposed approach robustness, three types of loads are considered. Through comparison with three other optimization approaches, the effectiveness of the equilibrium optimizer (EO) is demonstrated when it is employed to solve the proposed optimization scheme. The suggested planning approach is applied on both the IEEE 33-bus test system, and a real DN with 59 buses in Cairo, Egypt. Several significant conclusions are validated by the obtained results. First, the DG-HC assessment differs significantly depending on whether the EV charging demand is considered or not. Considering the EV integration in IEEE 33-bus, the mean value of the optimal DG-HC increased by more than 133% during the day. Secondly, the implementation of the proposed dynamic-pricing DR program in the IEEE 33-bus DN significantly improved both the DG-HC and EV-HC, named as the combined DG-EV-HC, with improvements of around 34% and 27% for DG-HC and EV-HC respectively. Finally, in IEEE 33-bus, the combined DG-EV hosting capacity was improved by approximately 49.2% regarding DG-HC and 61.2% regarding EV-HC, using the proposed synergistic DR-Volt/VAR control enhancement technique.

Anxiety among pregnant women in Addis Ababa, Ethiopia

PLoS ONE Hanna Melesse, Tigest Shifraw, Yemane Berhane Feb 14, 2025 DOI: 10.1371/journal.pone.0318718

Background Anxiety is one of the most prevalent mental health problems during pregnancy which result in various maternal and newborn complications affecting the health and well-being of the mother and the baby. In countries like Ethiopia, anxiety among pregnant women was not well explored, limiting the development of informed interventions. Thus, this research aims to assess the magnitude of anxiety and associated factors among pregnant women visiting ANC at public health centers in Addis Ababa, Ethiopia. Method We conducted an institutional-based cross-sectional study. Pregnant women who visited antenatal care (ANC) from April 1–14, 2021, were recruited from seven public health centers in Addis Ababa, Ethiopia. Data were collected using a structured questionnaire by trained data collectors. Poisson regression with a robust error variance estimate was used to calculate the prevalence ratio to identify the associated factors; the results were presented with an adjusted prevalence ratio (APR) and a 95% CI. Result A total of 323 pregnant women were recruited from the ANC clinics of public health centers in Addis Ababa. The overall prevalence of anxiety disorder was 12.70%, 95% CI (9.00%–16.10%). The prevalence risk of anxiety was higher among women who reported unplanned pregnancy compared to those who had planned pregnancy (APR 1.99, CI 1.04–3.81), and being single was associated with a higher prevalence risk than being married (APR 2.29, CI 1.14–4.59). Conclusion More than 1 in 10 women had anxiety during pregnancy. The prevalence of anxiety was about two-fold higher among single mothers and among mothers who had an unplanned pregnancy. The study suggests the importance of integrated mental health assessment and services in regular antenatal care, with an emphasis on identifying and caring for women at a disadvantage due to social and obstetric factors.

Incorporating bilingual translation templates into neural machine translation

Scientific Reports Fuxue Li, Beibei Liu, Hong Yan et al. Feb 14, 2025 DOI: 10.1038/s41598-025-86754-w

RETRACTED: Fast fault diagnosis of smart grid equipment based on deep neural network model based on knowledge graph

PLoS ONE Lin Jun, Zhou Chenliang Feb 14, 2025 DOI: 10.1371/journal.pone.0315143

The smart grid is on the basis of physical grid, introducing all kinds of advanced communications technology and form a new type of power grid. It can not only meet the demand of users and realize the optimal allocation of resources, but also improve the safety, economy and reliability of power supply, it has become a major trend in the future development of electric power industry. But on the other hand, the complex network architecture of smart grid and the application of various high-tech technologies have also greatly increased the probability of equipment failure and the difficulty of fault diagnosis, and timely discovery and diagnosis of problems in the operation of smart grid equipment has become a key measure to ensure the safety of power grid operation. From the current point of view, the existing smart grid equipment fault diagnosis technology has problems that the application program is more complex, and the fault diagnosis rate is generally not high, which greatly affects the efficiency of smart grid maintenance. Therefore, Based on this, this paper adopts the multimodal semantic model of deep learning and knowledge graph, and on the basis of the original target detection network YOLOv4 architecture, introduces knowledge graph to unify the characterization and storage of the input multimodal information, and innovatively combines the YOLOv4 target detection algorithm with the knowledge graph to establish a smart grid equipment fault diagnosis model. Experiments show that compared with the existing fault detection algorithms, the YOLOv4 algorithm constructed in this paper is more accurate, faster and easier to operate.