2026, 42(10):1-3.
Abstract:Taking the "information technology fundamentals" course as an example, this paper explores the application of digital media thinking in curriculum reform. Combined with WPS software teaching in computer practice, methods of using digital media thinking for teaching design are introduced, and its effectiveness is verified through teaching practice. Research shows that this method helps to enhance students' operational skills, innovation awareness, and autonomous learning ability, providing practical reference value for the curriculum reform of "information technology fundamentals".
2026, 42(10):4-6.
Abstract:With the rapid iteration of Artificial Intelligence (AI) technology and the continuous advancement of New Liberal Arts construction, higher education is facing new opportunities for interdisciplinary integration and teaching innovation. In the process of promoting curriculum reform, the social work major can explore the introduction of intelligent teaching methods to enhance students' practical ability, innovation ability, and comprehensive literacy. Based on the requirements of New Liberal Arts construction, this paper the main problems existing in the current social work curriculum in three aspects: design logic, teaching mode, and practical training, and proposes reform directions integrating AI technology, covering four dimensions: intelligent design of course content, adjustment of teaching methods, digital transformation of practical training, and optimization of evaluation mechanisms.
SU Yifei , LI Jing , XU Ruyu , ZHANG Lianchao
2026, 42(10):7-9.
Abstract:The rapid development of the big data industry has generated substantial talent demand. Currently, higher vocational colleges face practical challenges in cultivating talents for big data technology majors, including a disconnect between training objectives and industry demand, as well as curricula lagging behind technological advancements. To build a talent cultivation model oriented to industry demand, it is essential to optimize the curriculum system guided by job competency requirements, deepen school-enterprise cooperation through an industry-education integration collaborative mechanism, and establish a practical teaching system that integrates "post-course-competition-certification" pathways. The effective implementation of this model relies on systematic supports such as collaborative faculty development between schools and enterprises, the creation of teaching resources aligned with industry standards, the construction of practical teaching platforms based on real-world projects, and a diversified quality evaluation system.
2026, 42(10):10-12.
Abstract:To address the issues of frame rate drop and interaction latency caused by software-hardware adaptation limitations and network transmission loss in 3D simulation systems for anatomy, this study investigates the application of virtual simulation technology in the “anatomy” course. Organ models are reconstructed using image segmentation and the marching cubes algorithm, while mesh simplification and Physically Based Rendering (PBR) techniques are employed to optimize system performance. Gesture recognition and force feedback devices are integrated, and tissue manipulation and deformation haptic simulation are realized through collision detection and the mass-spring model. Computational fluid dynamics visualization methods are introduced to dynamically present blood flow velocity and pressure distribution. Experimental results show that the system achieves an average frame rate of 89.4 FPS and an interaction latency of 16.3 ms in static scenes. In complex interactive scenes, the frame rate remains above 68.5 FPS, with latency controlled within 28.2 ms, validating the effectiveness of the proposed method in improving system real-time performance and interaction authenticity.
2026, 42(10):13-15.
Abstract:The cultivation of cybersecurity awareness among students majoring in computer science in secondary vocational schools is of great significance. In response to the shortcomings of some students in terms of security knowledge and protective awareness, this paper analyzes the current situation and influencing factors based on literature review and questionnaire surveys. It further proposes strategies such as optimizing the curriculum system, strengthening practical teaching, organizing educational activities, and promoting collaboration among schools, families, and society. The study shows that systematic education and practical training can effectively enhance students' cybersecurity awareness and protection capabilities.
2026, 42(10):16-18.
Abstract:In the digital era, information technology literacy has become a crucial component of students' comprehensive professional competence. This paper presents a teaching reform practice centered on a project-based learning approach using intangible cultural heritage promotional video editing as the medium. It constructs a “conveying way through skill” teaching model that integrates the dissemination of ICH culture with the cultivation of video editing skills, achieving a synergistic improvement in both cultural literacy and technical competence. The findings indicate that this model not only effectively strengthens students' cultural identity and professional self-confidence but also significantly enhances their abilities in information acquisition, processing, and innovation, thereby providing a practical pathway and reference for the integration of technical and ethical education.
2026, 42(10):19-21.
Abstract:Starting from the basic principles of big data technology, this article explores its application path in the management of computer basic laboratories. By constructing a data collection, analysis, and decision support system, intelligent, refined, and visualized laboratory management can be achieved. The practical results have shown that big data technology has significant effects in laboratory resource scheduling, equipment maintenance, behavior analysis, and safety supervision, providing new ideas for the innovation of experimental teaching management models in universities.
2026, 42(10):22-24.
Abstract:This paper takes the “information technology” course in secondary vocational schools as the research object and constructs a systematic classroom evaluation system based on the textbook Information Technology published by Higher Education Press. Through in-depth analysis of the textbook content, course objectives, and the cognitive characteristics of secondary vocational students, a multi-dimensional evaluation indicator system is proposed, covering knowledge acquisition, practical skills, innovation ability, and comprehensive literacy. Corresponding evaluation methods and implementation strategies are also designed. Practical results show that this evaluation system can assess student' learning outcomes comprehensively and objectively, help stimulate their learning interest, and improve classroom teaching quality and talent cultivation effectiveness.
WANG Siqi , LI Jinghang , HE Xianghui , WANG Chunming , ZHANG Li
2026, 42(10):25-27.
Abstract:With the deep integration of digital intelligence technologies into the power industry, the knowledge system of power plant electrical equipment is undergoing a historic transformation. Firstly, analyzes the development characteristics of this field in the context of the digital age, including intelligent monitoring and early warning, equipment condition assessment, deep integration of primary and secondary equipment, and the digitalization of the entire lifecycle of information management. Secondly, it dissects the practical problems of traditional textbooks, such as outdated content, disconnect from actual enterprise needs, and weak practical components. Finally, it proposes a textbook compilation strategy centered on the production of textbook content, enterprise-oriented needs adaptation, digitalization of textbook format, and the integration of ideological and political education into the curriculum. Research shows that this strategy can effectively improve the practicality and applicability of textbooks, providing a useful theoretical reference for power engineering education to adapt to the digital transformation.
2026, 42(10):28-30.
Abstract:Against the backdrop of the ongoing advancement of Emerging Engineering Education and the rapid development of the information technology industry, “operating system” as a core course for computer science majors, has become a crucial tool for implementing the fundamental task of “foster virtue through education” through its curriculum-based ideological and political education. Addressing issues such as insufficient student-student adaptation, unnatural integration of ideological and political education with professional studies, and a relatively simplistic evaluation system in current curriculum-based ideological and political education practices, this paper systematically explores the inherent ideological and political elements within the course, constructing a teaching paradigm of “foundation building through ideological and political education, dual-drive empowerment, and diversified evaluation”. Furthermore, it proposes a systematic reform path from four dimensions: educational philosophy, integration of ideological and political education, teaching methods, and evaluation system. Practical results demonstrate that this paradigm effectively promotes the organic unity of knowledge transmission, ability cultivation, and value guidance, providing a referable and scalable implementation path for similar courses to conduct curriculum-based ideological and political education.
2026, 42(10):31-33.
Abstract:Based on the GeoGebra dynamic mathematics software, a dice-rolling probability experiment simulation program was designed and implemented to intuitively explore the relationship between the theoretical mean and the sample mean of a random variable. This program, through the construction of an interactive model, can simulate multiple dice-rolling experiments and record the changing trend of the sample mean with the increase of the number of experiments in real time, while also comparing and analyzing it with the theoretical mean.
MENG Qingrui , CHEN Lu , SUN Zhihong
2026, 42(10):34-36.
Abstract:“Foster virtue through education” is the fundamental task of education and a core embodiment of the educational policy in the new era. The implementation of ideological and political education in courses should not be limited to theoretical courses but should permeate all types of professional courses, organically integrating ideological and political elements into the entire teaching process to guide students in establishing correct values. Based on the concept of “Three-Dimensional Education”, this approach systematically promotes ideological and political education through full participation, comprehensive curriculum design, and full-process management throughout the university stage. It also proposes targeted improvement measures and evaluation methods to enhance the effectiveness of ideological and political education in information security professional courses.
MA Bing , GUO Yi , ZHOU Kehong , SUN Min
2026, 42(10):37-39.
Abstract:Addressing the problems of low resource utilization and delayed learning feedback in communication engineering classes, this paper proposes a teaching model based on the integration of big data and AI. This method constructs a closed-loop teaching system of "data collection—learning profiling—personalized recommendation—instant feedback", and a comparative experiment was conducted in the "principles of communication" course. Class A adopted the traditional teaching model, while Class B adopted a big data and AI-assisted teaching model. During the experiment, log data from the teaching platform and questionnaire data were collected, and statistical and comparative analyses were performed on four dimensions: on-time assignment rate, chapter test pass rate, learning activity, and course completion satisfaction. The results show that Class B improved by approximately 8% to 20% in all four indicators compared to Class A, and the learning activity showed a continuous upward trend.
ZHAI Yingying , XU Bin , HE Chang , YANG Xiaochun
2026, 42(10):40-43.
Abstract:The rapid development of Artificial Intelligence (AI) technology has placed new demands on higher education. Integrating value guidance into AI curriculum systems and constructing a curriculum-based ideological and political education model that meets the needs of the times has become an important direction. The "value-and technology-driven" AI curriculum-based ideological and political education model combines project-based teaching with blended learning to organically integrate professional knowledge with ideological and political education. This model centers on project practice, incorporating elements of technological ethics and social responsibility into key aspects such as algorithm design and system development. This enables students to master core AI technologies while establishing correct values and enhancing their sense of social responsibility. Simultaneously, it optimizes the teaching structure through blended learning, supports self-directed learning with online resources, and deepens understanding through offline discussions, constructing a closed-loop teaching system that promotes the coordinated development of "knowledge—ability—value". Practice has shown that this model can effectively improve students' technological application abilities and ideological and political literacy, enhancing the precision and sustainability of teaching.
HU Fuzhi , QUAN Yue , WANG Shi
2026, 42(10):44-47.
Abstract:With the widespread application of information technology in education, this paper explores a solution for reforming the “microcomputer principle and interface technology” course by introducing knowledge graphs to improve teaching quality and student learning outcomes, addressing the issues of abstract content and complex knowledge relationships. Based on the course characteristics, a systematic analysis was conducted on key aspects of knowledge graph construction, including data collection, entity recognition, relation extraction, and visualization storage. Specific application solutions were designed from three dimensions: personalized learning path planning, intelligent recommendation of teaching resources, and teaching evaluation. Furthermore, corresponding solutions were proposed to address the challenges faced in knowledge graph construction and application, providing feasible optimization directions for course reform.
2026, 42(10):48-52.
Abstract:In the context of digital education, and facing the rapid development of database technology and the higher requirements of national smart education for applied undergraduate teaching, it is necessary to systematically explore the construction path of database course cluster teaching and research offices. First, the construction goals of database course cluster teaching and research offices should be clearly defined based on the development needs of digital and smart education. Second, the digital teaching reform of database course cluster teaching and research offices should be comprehensively promoted from aspects such as the construction of a digital course cluster system, optimization of the teaching staff, construction of teaching resource databases, improvement of institutional guarantees, and development of distinctive directions, so as to continuously improve the teaching quality and management level of the course cluster.
2026, 42(10):53-57.
Abstract:Digital transformation of higher vocational college professional clusters has become an inevitable path to achieve high-quality development. However, at present, both theoretically and practically, there is still a lack of evaluation tools and analytical frameworks that can systematically diagnose the transformation status and reveal the evolutionary laws. Based on this, this article constructs a STRPG maturity model containing five core dimensions: "Strategy & Organization (S), Technology & Platform (T), Resource & Curriculum (R), Pedagogy & Mode (P), and Governance & Evaluation (G)", and divides it into five development stages from "Initial" to "Leading", providing a basis for the systematic evaluation of the current status of digital transformation. On this basis, the evolution path of digital transformation of higher vocational college professional clusters is further analyzed, revealing the basic laws of its transformation from external driving to endogenous development and from technology application to paradigm innovation, and clarifying the key tasks and driving mechanisms of each stage of evolution. The theoretical framework constructed in this article integrates diagnostic, evaluation, and guidance functions, providing systematic theoretical support and practical reference for vocational colleges to clarify their transformation direction and optimize their development path.
2026, 42(10):58-60.
Abstract:Taking the "information technology" course in primary school as the research object, this paper discusses the application practice of GenAI in Project-Based Learning (PBL). The project-based learning environment of primary school information technology based on GenAI was constructed, easy to operate cases were designed, classroom teaching experiments were carried out, students' learning data and feedback were collected, and the influence of GenAI on students' creativity, information technology application ability and learning interest was analyzed. The research shows that GenAI can improve students' informationization practice ability and learning initiative, and provide a new paradigm and practical direction for primary school information technology teaching.
2026, 42(10):61-63.
Abstract:Taking the construction of an electrical automation training laboratory in a vocational college as the research object, this paper explores methods for optimizing the training environment, intelligent teaching design, and teaching effect improvement empowered by artificial intelligence (AI). By introducing intelligent experimental equipment, data acquisition and analysis systems, and AI-assisted teaching platforms, an intelligent training environment is constructed, achieving real-time monitoring, automatic evaluation, and personalized guidance of the experimental process. Experimental results show that the AI-empowered training model significantly improves students' operational skills, problem-solving abilities, and learning motivation, providing practical experience for electrical automation education.
2026, 42(10):64-67.
Abstract:To explore the impact of integrating Generative Artificial Intelligence (GenAI) with project-based teaching on the higher-order thinking skills of vocational education students, this study employed a randomized controlled experimental design with 100 higher vocational students as participants. Combining scale assessments and semi-structured interviews, the study conducted repeated-measures ANOVA and NVivo qualitative coding for comprehensive analysis. The results indicate that GenAI positively contributes to the development of higher-order thinking by expanding cognitive boundaries, strengthening knowledge integration, and promoting cognitive iteration. The study proposes a "empowerment rather than substitution" human-machine collaborative teaching model, providing empirical evidence and practical references for fostering higher-order thinking skills in the context of digital transformation in vocational education.
2026, 42(10):68-70.
Abstract:The "database principle and application" course holds an important position in the teaching of computer-related majors in colleges and universities. However, the traditional teaching model, which relies primarily on command-line operations and is relatively abstract, tends to limit students' learning interest and the development of their practical skills. To address this issue, this paper takes the "database principle and application" course as the research object and explores the application strategies and effects of visualization tools in teaching reform. By introducing visual database management tools such as Navicat, DBeaver, and MySQL Workbench, a task-driven teaching model is constructed, with optimizations made in terms of teaching content organization, practical operation guidance, and experiment management. The results show that the introduction of visualization tools can effectively stimulate students' learning interest and achieve a dual improvement in knowledge comprehension and teaching quality.
2026, 42(10):71-73.
Abstract:With the rapid development of information technology, there is an increasing demand for application-oriented and innovative talents in computer-related majors. However, traditional teaching models in universities still face challenges such as a disconnect between theoretical instruction and engineering practice, as well as insufficient practical skills among students. Under the background of industry-education integration, school-enterprise collaborative education has become a crucial pathway to enhance the quality of talent cultivation. Focusing on computer-related majors, this paper analyzes the main implementation models of school-enterprise collaborative education from the perspective of talent cultivation goals, including joint curriculum development, collaborative practical teaching, and enterprise participation in assessment, while also discussing the corresponding operational mechanisms. Based on this, the study examines the effectiveness of collaborative education through practical teaching cases. The findings indicate that this model helps strengthen students' practical abilities and career adaptability, offering valuable insights for reforming the talent cultivation model in computer-related majors.
2026, 42(10):74-76.
Abstract:The application of virtual simulation platforms in the teaching of mechanical control systems can provide students with an intuitive and interactive learning environment. Relying on computer simulation algorithms and human-computer interaction technologies, the platform realizes the visual simulation of mechanical control processes and real-time feedback of operational data, which helps students understand control principles and improve their practical application abilities. This paper focuses on the application of virtual simulation platforms in mechanical control system courses in colleges and universities. It summarizes the basic concepts and technical architecture of the platform, analyzes its core technologies and implementation methods, and discusses the construction of simulation models and the design of teaching modules, aiming to improve the overall teaching quality and practical effectiveness of mechanical control system courses.
REN Jingjing , WANG Yali , CHEN Shulong
2026, 42(10):77-80.
Abstract:Under the background of the New Liberal Arts, how to effectively integrate ideological and political education into the “university computer fundamental” course for foreign language majors in military academies, while improving students? scientific and information literacy and achieving the goal of ideological and political education, has become an important issue that urgently needs to be addressed in the construction of ideological and political education in courses. By analyzing relevant documents and meeting guidelines, and combining the requirements of the new liberal arts construction and the characteristics of computer science, this paper outlines the basic principles for constructing case studies for ideological and political education in courses, and constructs a case study system for “university computer fundamentals” that integrates “New Liberal Arts+military characteristics+information technology”. Based on this, a classroom implementation path of “case introduction— problem exploration—value guidance” is designed, providing practical reference and development direction for the construction of ideological and political education in courses in military academies.
2026, 42(10):81-83.
Abstract:In order to further implement and promote the construction of a new pattern of comprehensive education in the “Great Ideological and Political Course”, accurately grasp the connotation of the times and the development requirements of the “Emerging Engineering Education”, the “university computer fundamental” course should take the cultivation of scientific thinking and the improvement of humanistic literacy as the starting point, and promote the practical implementation of ideological and political education in the course. Based on the educational philosophy of “student-centered and student-centered”, this article analyzes and explores the integration path of curriculum ideology and computational thinking cultivation, focusing on how to effectively support the cultivation of computational thinking through curriculum ideology. By promoting the organic integration of value guidance and knowledge transmission through cultural individuals, the aim is to provide reference ideas and practical paths for the construction of the “university computer fundamental” course in universities.
2026, 42(10):84-86.
Abstract:"Communication electronic circuit" is a fundamental compulsory course for the school of electronic information, aimed at cultivating technical talents in the field of communications. It combines theoretical, engineering, and practical aspects. However, the course content is relatively dry, formula derivations are difficult, concepts are abstract, and circuit structures are complex. As a result, the student pass rate has long been low, and the teaching effectiveness needs improvement. The course mainly consists of three parts: theoretical teaching, in-class experiments, and course design. The in-class experiments present the most significant problems: students rely heavily on the experimental manuals to follow the steps, lacking a deep understanding of the underlying circuit principles. Furthermore, with the course's long-term operation, some experimental equipment has become damaged or even obsolete due to frequent use, further affecting the quality of experimental teaching. Based on the requirement to integrate ideological and political education into university professional courses, and considering the inherent alignment of the course content with the needs of national rejuvenation, an optimized experimental teaching scheme is proposed: integrating the disassembly of experimental box circuit principles, analysis of equipment damage causes, and troubleshooting practice into the experimental teaching process. This approach not only helps students gain a deeper understanding of circuit principles and reduce equipment maintenance costs, but also integrates ideological and political elements such as engineering responsibility and innovative spirit into teaching, promoting the transformation of experimental teaching from "operational training" to "engineering literacy cultivation", and achieving an organic unity of knowledge transmission, ability development, and value guidance.
2026, 42(10):87-88.
Abstract:Against the backdrop of the in-depth advancement of the “Digital China” strategy, the quality of talent cultivation in cloud computing technology application majors in higher vocational colleges is directly related to the effectiveness of industrial digital transformation. As an important base for cloud computing talent cultivation, higher vocational colleges generally suffer from an overemphasis on skills and a neglect of scientific thinking, resulting in insufficient scientific thinking abilities among students and an inability to meet the actual needs of enterprises for highly skilled personnel. To overcome this predicament, it is necessary to organically integrate five key indicators of scientific thinking into the curriculum design of cloud computing technology application majors: problem analysis and modeling ability, critical evaluation and decision-making ability, systematic thinking and associative ability, hypothesis testing and iterative optimization ability, and evidence-driven communication and expression ability. These indicators should be strengthened during the teaching process to comprehensively improve students' scientific thinking level and better adapt to the new social demand for highly skilled cloud computing talent.
2026, 42(10):89-91.
Abstract:In the new era, higher education emphasizes both knowledge transmission and value guidance. How to organically integrate Outcome-Based Education (OBE) principles with ideological and political education in courses has become a crucial issue in current teaching reforms. Based on the OBE concept and considering the characteristics of the "object-oriented programming" course, a "three-dimensional" ideological and political education model has been constructed. This model uses "knowledge transmission, ability cultivation, and value guidance" as its three core dimensions. Through reverse design of teaching objectives, integration and reconstruction of teaching content, diversified and collaborative teaching methods, and multi-dimensional feedback in teaching evaluation, it achieves a systematic integration of professional education and ideological and political education. The "three-dimensional" model fully embodies the inherent consistency between the OBE concept and ideological and political education in courses, providing a theoretical framework and practical path for the construction of ideological and political education in programming courses.
2026, 42(10):92-94.
Abstract:To address the issues of disconnect between theory and practice, low student participation, and rigid integration of ideological and political education in the "microcontroller principle and interface technology" course, this paper proposes a teaching reform model centered on scenario-based teaching and driven by projects. This model is grounded in real-world application scenarios of the Internet of Things (IoT) and industrial automation, embedding core course knowledge points into typical engineering tasks to construct a chain of operable and immersive micro-projects. Through a teaching process of "role-playing—task execution—system feedback", students are guided to complete the cognitive transformation from abstract principles to concrete operations in simulated engineering practice. During implementation, engineering ethics, safety standards, and sustainable development concepts are naturally integrated, achieving an organic unity of knowledge transmission and value guidance. Practical results show that this model effectively improves students' classroom participation, engineering practice ability, and systems thinking level, while also enhancing teamwork awareness and problem-solving skills. Questionnaire survey results show that students' understanding of abstract concepts and their sense of engineering responsibility have significantly improved. Characterized by "small entry point, deep integration", this model possesses good replicability and promotional value, providing a practical path for the teaching reform of electronic information-related majors under the new engineering background.
2026, 42(10):95-98.
Abstract:Addressing the issues of curriculum disconnect from industry needs, superficial industry-education integration, and application-oriented evaluation systems in the cultivation of application-oriented talents in computer science, this paper analyzes the root causes and proposes a "three-level, four-collaboration, dual-circulation" talent cultivation model. This model constructs a progressive competency development system based on fundamental, professional, and applied skills, it achieves multi-dimensional collaborative education through co-construction of courses, mutual appointment of faculty, platform sharing, and joint evaluation. and it dynamically optimizes the teaching process and content through a university-enterprise "dual-circulation" mechanism. Practical results show that this model effectively improves the engineering practice capabilities of computer science students, with a graduate employment placement rate of 91.67%.
LI Tao , XU Xinyuan , SUN Hongzhi , HUO Yanxia , LIU Tao
2026, 42(10):99-101.
Abstract:Focusing on the 2025 Intelligent Warehousing Control Research Project of the Materials Company of State Grid Shandong Electric Power Company, this paper proposes an e-commerce material warehousing technical solution based on a multi-modal large model. By capturing on-site images of incoming materials and performing multi-modal comparison with standard images and specification information provided by e-commerce platforms, automatic verification of material compliance is achieved. Deployment results show that the solution achieves an identification accuracy of over 98% for material types and specifications, effectively improving the efficiency of material warehousing operations and significantly reducing manual verification costs.
2026, 42(10):102-104.
Abstract:Aiming at the server power consumption problem in high-density data centers, a power consumption modeling method based on performance metrics is proposed, and a reinforcement learning (RL) method is used to provide an energy efficiency optimization strategy for data centers. Experimental results show that this method can significantly reduce the total energy consumption of data centers while ensuring service performance, providing theoretical and practical guidance for the construction of green data centers.
WU Zhongyi , PENG Zhenzhen , WANG Hao , TONG Wei
2026, 42(10):105-107.
Abstract:In the environment of parallel multi-project management in enterprises, resource conflict problems have a significant impact on project progress and execution efficiency. Aiming at the problems of resource conflict identification and optimization in multi-project environments, a resource conflict identification and optimization method based on intelligent algorithms is proposed. By modeling project resource requirements and combining priority analysis with constraint condition detection, rapid conflict identification is achieved. Meanwhile, an optimization strategy combining an improved genetic algorithm with heuristic rules is introduced to achieve intelligent resource allocation and conflict mitigation. Simulation results show that this method can effectively improve multi-project resource utilization, reduce conflict occurrence frequency, and shorten the overall project cycle, providing a feasible intelligent solution for enterprise multi-project management.
ZHANG Ning , FENG Zonghong , XU Kai
2026, 42(10):108-115.
Abstract:Facial expressions are an important way for humans to express emotions and are closely related to daily life. Accurate identification and classification of facial emotions has significant research value in the field of affective computing. To further improve the classification performance of the model, a novel facial emotion classification model—Effi-DictNet is proposed based on EfficientNet and dictionary learning methods. This model effectively integrates the deep feature extraction capabilities of EfficientNet with the sparse representation advantages of dictionary learning, thereby enhancing feature representation capabilities and improving overall recognition performance. During model training, a ReduceLROnPlateau learning rate scheduling strategy is introduced to dynamically adjust the learning rate according to changes in training loss, thereby accelerating model convergence and improving training stability. The proposed model is validated using two publicly available facial emotion datasets. Experimental results show that on the color facial emotion dataset, the Effi-DictNet model performs well, achieving an accuracy of 98.69% on the training set, a stable accuracy of 77.70% on the validation set, and a test set loss of 0.87 with a test accuracy of 81%. On the FER13 dataset, the model's training accuracy is further improved to 81.34%. The comprehensive experimental results show that the Effi-DictNet model has good performance and application potential in facial emotion classification tasks.
2026, 42(10):116-118.
Abstract:To address the imbalance among accuracy, complexity, and generalization capability in nonlinear compensation algorithms for optical fiber communication, this paper proposes an improved algorithm that integrates adaptive-step-size Split-Step Fourier Method (SSFM), multi-channel interference modeling, and a lightweight Convolutional Neural Network (CNN). The algorithm performance is validated through mathematical modeling and simulation experiments. The results show that under a transmission distance of 1 000 km, the improved algorithm achieves a bit error rate of 1.2×10-4, reduces the computational complexity by 60% compared to the traditional Backpropagation (BP) algorithm, and attains a GPU latency of 0.9 μs, meeting the requirements of 800 G systems. It also demonstrates good performance across multiple scenarios. The study indicates that the fusion of model-driven and data-driven approaches can effectively enhance nonlinear compensation, providing a feasible solution for high-speed long-haul optical fiber communication.
CHEN Tingting , CAI Jiacan , ZHENG Han , FENG Jie
2026, 42(10):119-121.
Abstract:To address the severe channel contention, large latency fluctuations, and unstable throughput issues in distributed ad hoc networks under highly dynamic topologies and limited spectrum resources, a dynamic time slot allocation algorithm based on a distributed cooperative mechanism is proposed. This algorithm, based on a TDMA framework, introduces local information exchange and contention awareness mechanisms to achieve adaptive allocation of time slot resources and conflict suppression among nodes. Furthermore, by constructing a system model and optimizing the objective function, the time slot allocation process is constrained and modeled, and dynamic scheduling is achieved through a distributed iterative update strategy. Simulation results show that this proposed method outperforms traditional fixed TDMA and random access methods in terms of throughput, latency, and channel conflict rate.
2026, 42(10):122-124.
Abstract:Aiming at the requirements of Ultra-Reliable Low-Latency Communication (URLLC) in Internet of Vehicle application scenarios, the key requirements of Internet of Vehicle in terms of reliability and latency are analyzed. Key technologies such as 5G/6G communication, edge computing, multi-access, and intelligent scheduling are studied, and a low-latency high-reliability communication architecture for Internet of Vehicle is proposed. Simulation results show that this solution achieves significant improvements in both latency and communication reliability, providing a reference for intelligent transportation and autonomous driving application systems.
SUI Yang , LU Lu , BAI Longyu
2026, 42(10):125-127.
Abstract:In the AI vision-based automatic positioning of welding robotic arms, the Unscented Kalman Filter (UKF) model is susceptible to matrix perspective transformation, leading to low positioning and registration accuracy. To address this issue, this paper proposes an AI vision-based automatic positioning algorithm for welding robotic arms in large-diameter pipeline emergency repair. By constructing an AI vision-based automatic positioning model for the welding robotic arm and generating the corresponding positioning strategy, the automatic positioning of the welding robotic arm is achieved. Experimental results show that the proposed algorithm achieves optimal registration under different numbers of point cloud parameters, accurately describes local geometric features, and offers high positioning accuracy and robustness.
WANG Yaohai , CAO Gang , WANG Jiamian , LI Pengfei , WANG Li
2026, 42(10):128-130.
Abstract:Based on the 2025 Scrapped Materials Intelligent Control Research Project of the Materials Company of State Grid Shandong Electric Power Company, this paper proposes an Artificial Intelligence (AI)-driven residual value assessment and decision optimization model for scrapped materials. By conducting a multi-dimensional analysis of key factors affecting residual value—including physical characteristics, material composition, usage scenarios, and market supply and demand—and integrating big data technologies with deep learning algorithms, a complete framework is established encompassing data integration, residual value prediction, and dynamic decision-making. Experimental results show that the model can accurately predict the disposal value of scrapped materials, assist in formulating rational disposal plans, improve resource utilization efficiency and economic benefits while meeting environmental requirements, and provide data-driven support for the development of a circular economy.
2026, 42(10):131-134.
Abstract:To address the issues of high subjectivity, low efficiency, and high cost associated with traditional manual road inspection methods, a deep learning-based automatic road defect identification method is proposed. To meet the high-precision identification requirements for common defects such as cracks, alligator fissures, potholes, and repairs (with an accuracy rate of no less than 80% for each type of defect), YOLOv11s is selected as the base model, and a lightweight improved algorithm, AD-YOLO-F, is proposed. Specifically, firstly, an ADown downsampling module is introduced to replace some conventional convolutional modules, effectively reducing computational load and parameter size while maintaining detection performance; secondly, a Focal Modulation module replaces the SPPF module in the backbone network to optimize multi-scale feature fusion capabilities and improve the detection accuracy of small targets in complex backgrounds. Experiments are conducted on a consolidated dataset of six types of road defects. The results show that the improved AD-YOLO-F model achieves an accuracy rate of 94.1%, significantly higher than the expected 80%. Furthermore, compared to the original YOLOv11s model, the number of parameters is reduced by approximately 20%, computational load by approximately 21%, and model weights are compressed by approximately 19%. This method achieves high-precision identification while maintaining a lightweight model, providing a feasible technical solution for deploying road defect detection systems on resource-constrained edge devices.
CAO Zijian , MING Xinghui , SHI Erfang
2026, 42(10):135-137.
Abstract:With the expansion of distribution network scales and the increasing complexity of their structures, traditional fault detection and location methods face challenges in terms of real-time performance and accuracy. Intelligent algorithms, owing to their strong capabilities in data processing and pattern recognition, have been widely applied in fault detection and location for distribution networks. This paper reviews the research progress and application approaches of typical intelligent algorithms such as traveling wave method, genetic algorithm, neural network, and deep learning, and analyzes their roles in improving the efficiency and accuracy of fault location. The study shows that algorithmic methods based on multi-source data fusion and intelligent optimization can effectively enhance the fault handling capability of distribution networks.
LIANG Yulin , SHI Erfang , TIAN Jiawang
2026, 42(10):138-139.
Abstract:This paper comprehensively applies Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and their hybrid strategies to optimize power grid dispatch schemes. By constructing typical simulation cases, the performance of the optimization methods in reducing dispatch costs, achieving load balancing, and improving system operational stability is verified. The results show that the proposed method can effectively enhance the economy and security of power grid dispatch, providing theoretical basis and technical support for the high penetration of renewable energy and the development of smart grids.
2026, 42(10):140-142.
Abstract:With the continuous development of information technology, GIS technology is increasingly widely used in detailed planning projects. However, the handling of road intersections in detailed planning still faces problems such as heavy workload and low efficiency. Based on this, a systematic study of the intersection handling process was conducted, classifying road intersections into four types and designing corresponding intersection handling algorithms. Finally, the proposed algorithms were implemented and verified using the ArcGIS platform, aiming to improve the efficiency and standardization of detailed planning.
ZHANG Jun , XU Xinyuan , HUO Yanxia , LI Haibo , GAO Zekai
2026, 42(10):143-145.
Abstract:Based on the 2025 Warehouse Digital Intelligence Control Research Project of the Materials Company of State Grid Shandong Electric Power Company, this paper proposes a visual inventory software solution for power storage facilities. First, the warehouse business processes are analyzed, and the core requirements for visual inventory are identified. A deep learning-based object detection model architecture is designed, integrating preprocessing, feature extraction, and attention mechanism modules, while optimizing anchor box strategies and loss functions to build a dedicated recognition model. Second, an interface adaptation layer is developed to achieve seamless integration between the model and the warehouse system. After embedding the model into the software, a closed-loop “acquisition—inference—display” processing flow is established. Model quantization compression and edge deployment are adopted to reduce response latency. Finally, a sample library module is introduced to support incremental iterative updates of the model. Experimental results validate the application value of this method in improving inventory accuracy, shortening inventory time, and reducing error rates.
2026, 42(10):146-148.
Abstract:With the rapid development of autonomous driving technology, L3-level and above autonomous driving vehicles have put forward higher requirements for the reliability and accuracy of control systems. As a core technology to replace traditional mechanical steering, the Steer-by-Wire (SbW) system can achieve high-precision steering control and provide flexible control interfaces for Advanced Driver Assistance Systems (ADAS) and autonomous driving systems. Based on the actual requirements of L3-level and above autonomous driving, a highly reliable, low-latency and scalable SbW system architecture is designed, covering four aspects: sensor redundancy design, actuator control strategy, control logic hierarchy and safety guarantee mechanism. The simulation and experimental results show that the system can stably maintain high precision and high reliability in complex driving scenarios.
XU Lujie , ZHANG Hanyun , QIN Wei , ZHANG Yaozhong , ZHANG Yi
2026, 42(10):149-151.
Abstract:Aiming at the problems of insufficient positioning accuracy, limited power consumption, and high real-time requirements faced by micro UAVs in indoor or complex environments, a low-power visual and UWB fusion navigation hardware accelerator based on an embedded platform is designed from an engineering practice perspective. The system uses STM32 as the core control unit, integrates a CMOS image sensor and UWB positioning module, and achieves multi-source information fusion through PCB-level hardware design and lightweight algorithms. Experimental results show that this significantly improves positioning stability and navigation accuracy while maintaining low power consumption, demonstrating good engineering application value.
2026, 42(10):152-154.
Abstract:Aiming at the problems of insufficient recommendation accuracy and limited personalization level in university learning platforms, a learning resource recommendation model based on behavior sequence analysis and deep learning is proposed. This model constructs student learning behavior sequences, combines Long Short-Term Memory (LSTM) networks to model learning paths, and introduces an attention mechanism to enhance the expressive power of key behavioral features. On this basis, a multi-feature fusion recommendation framework is constructed to achieve accurate recommendation of learning resources.
HUO Xiaoqian , ZHOU Hao , JIA Bingrun
2026, 42(10):155-158.
Abstract:A YOLOv8 based intelligent detection and analysis platform has been developed to address the issues of strong subjectivity, low detection efficiency, and difficulty in recognizing complex cell morphology in the detection of astrocytes. The platform is based on high-precision object detection algorithms and combines Vue3 front-end and Spring Boot back-end architecture to achieve automatic recognition and precise localization of astrocytes. The system supports real-time image processing, visualization of detection results, and data sharing functions, significantly improving detection efficiency and accuracy. At the same time, the platform has established a unified data management system and standardized annotation process, providing efficient and reliable analysis tools for researchers. The research results indicate that the platform has good stability and universality, which can provide strong technical support for astrocyte image analysis and intelligent medical image processing, and lay the foundation for the deep integration of artificial intelligence in neuroscience research and clinical applications.
ZHOU Liang , ZHOU Huan , ZHOU Dongxin
2026, 42(10):159-163.
Abstract:In 5G and 4G hybrid network environments, the Telecom Operation Support System (OSS) faces the challenge of massive heterogeneous alarms and complex fault location caused by the interweaving of multiple technology domains. Traditional diagnostic methods suffer from problems such as lagging knowledge updates and insufficient generalization ability. While large language models possess rich general knowledge, their application in critical operation and maintenance scenarios is limited by the "illusion" phenomenon and insufficient knowledge timeliness. This paper proposes an intelligent fault diagnosis framework that integrates Retrieval Augmented Generation (RAG) and Graph Neural Network (GNN). By constructing a multi-source knowledge base (covering 3GPP/ITU-T standards, historical fault cases, and network topology information) and introducing a proxy retrieval mechanism, the reasoning ability of the large language model in cross-domain fault root cause analysis is enhanced. Experimental results show that compared with random forest and traditional GNN baseline models, the diagnostic accuracy of this method is improved by 28.0% and 15.3%, respectively, and it exhibits better generalization ability in small sample scenarios. This research provides a feasible technical path for the intelligent upgrade of operator OSS and has important practical value for improving the operation and maintenance efficiency and automation level of complex heterogeneous networks.
2026, 42(10):164-165.
Abstract:A precise fault location method for distribution networks based on intelligent algorithms is proposed, and a corresponding automated fault handling platform is designed and implemented. This method collects data such as current, voltage, relay protection action information, and switch status, and employs improved fuzzy reasoning and graph neural network algorithms to achieve accurate identification of fault types and fault locations. The platform integrates automated control functions and can perform fault isolation, reclosing, and backup power switching operations based on the positioning results. Experimental results show that the system can achieve rapid location, accurate judgment, and intelligent handling after fault occurrence, significantly improving the safety and reliability of distribution network operation.
2026, 42(10):166-168.
Abstract:As data center scale continues to expand, rack thermal environment management has become a key factor in ensuring equipment stable operation and energy saving. To address this, a dynamic simulation and intelligent control system for data center rack thermal environment based on digital twin is proposed. By constructing digital twin models of racks and cooling equipment, dynamic simulation and prediction of the temperature field inside racks are achieved. Based on the simulation results, intelligent control of air conditioning and cooling equipment is performed to achieve rack-level temperature regulation. Experimental results show that the system can significantly reduce the fluctuation amplitude of hot spots, improve Power Usage Effectiveness (PUE), and demonstrate good practicality and promotion value.
LI Lei , HE Tianqi , LI Youkang
2026, 42(10):169-171.
Abstract:To study the transmission patterns of emerging infectious diseases, this paper establishes a discrete-time SIRD epidemic transmission model based on the theory of numerical series and Cauchy product. Building upon the classical SIR model, the population is divided into four categories: susceptible, infected, recovered, and deceased. A convolution series is employed to describe the relationship between the number of newly infected cases and historical infection data, thereby formulating the dynamic equations of epidemic transmission. Furthermore, the infectious force function, mortality risk function, and recovery probability function are introduced to fit and analyze the model parameters. Finally, the model is validated using simulated data. The results indicate that the proposed model can effectively capture the fundamental dynamical characteristics of epidemic transmission and may serve as a reference for epidemic analysis and prediction.
2026, 42(10):172-174.
Abstract:Taking Tongcheng Social Insurance Management Center as the research object, combined with its internal control, audit management, fund financial management, and first-question responsibility system, this paper analyzes the shortcomings of the existing social security information system in multi-permission management and business risk control. An architecture optimization scheme based on hierarchical permission control, separation of duties, operation auditing, and intelligent risk early warning is proposed. Combined with business process optimization, clear job responsibilities, and audit monitoring mechanisms, system security, business efficiency, and risk controllability are achieved. Research shows that the optimized multi-permission information system can effectively prevent the illegal use of social security funds and business operation risks, improve business processing efficiency and insured person satisfaction, providing a reference for the information construction of local social security institutions.
2026, 42(10):175-176.
Abstract:Against the backdrop of the rapid development of the Internet of Things (IoT) and Artificial Intelligence (AI), multimodal human-computer interaction is gradually becoming an important direction in the design of smart devices. This paper proposes a smart desktop ornament system that combines independent operation and collaborative work. The system integrates technologies such as microcontrollers, wireless communication, gesture sensing, and dot matrix display to achieve a natural and intuitive interactive experience. The system uses an STM32 as the core control unit, combined with an ESP8266 to achieve wireless communication between multiple devices, and uses an MPU6050 to capture and recognize user gestures to complete interactive operations. Each ornament is equipped with an 8×8 dot matrix display module, integrating nine functional modes including clock display, music spectrum, mini-games, temperature monitoring, and weather display, and can realize multimodal feedback such as graphics, facial expressions, and lighting. Furthermore, the system supports both single-device independent operation and multi-device collaborative operation modes. In collaborative mode, the master node can uniformly control multiple slave devices through the communication interface to achieve linked display and synchronous response. Experimental results show that the system performs well in terms of interactive sensitivity, operational stability, and functional scalability, providing a feasible implementation approach for the design of emotional and multimodal interactive products in smart home scenarios.
LIANG Yanxia , SHENG Yibo , HUANG Weijie , HE Yuwen , GAN Panxi
2026, 42(10):177-179.
Abstract:With the rapid development of mobile communication technology, mobile communication networks generate massive amounts of data, which contain rich information on social behavior, user preferences, and network operation. How to efficiently utilize these big data resources has become a hot topic of concern for communication enterprises and research institutions. Based on the characteristics and application requirements of mobile communication big data, this paper analyzes its main utilization modes, covering operational optimization, precision marketing, urban management, and research analysis. Combined with case studies, key aspects such as data collection, processing, analysis, and application are discussed, and corresponding technical and management recommendations are proposed, providing a reference for the efficient utilization of mobile communication big data resources.
CHEN Weisheng , DING Changwu , ZHANG Xirui
2026, 42(10):180-182.
Abstract:With the continuous expansion of electrical engineering scale and the increasing level of intelligent application, the requirements for efficiency, safety, and reliability in system operation have been simultaneously raised. Traditional line installation methods can no longer meet the demands of current engineering. Research is conducted on electrical engineering line installation technology and its supporting intelligent monitoring applications, the basic processes, core technologies, and common problems of line installation, and analyzing the application paths of intelligent monitoring technology in both construction and operation stages. Research results show that the application of intelligent monitoring technology significantly improves line construction quality, construction efficiency, and later-stage operation and maintenance levels, providing scientific reference for electrical engineering construction.
YIN Xiangling , WANG Haibo , CHEN Jianxue
2026, 42(10):183-185.
Abstract:Addressing the practical needs of quality management in chemical enterprises, this paper proposes a method for managing and optimizing decision-making of chemical product testing data based on intelligent analysis technology. By constructing core modules such as data acquisition, feature extraction, anomaly identification, and predictive analysis, automated processing of testing data and forward-looking early warning of quality risks are achieved. Experimental results show that this method improves the efficiency of testing data processing by approximately 48%, achieves an anomaly identification accuracy of 96.7%, and significantly enhances the stability and controllability of product quality.
2026, 42(10):186-193.
Abstract:In today's wave of enterprise digital transformation, information system construction is showing a diversified and professional development trend. However, due to differences in suppliers, technology platforms, and application goals, the problem of "information silos" is prevalent within enterprises, severely restricting the mining of data value and the collaborative efficiency of business processes. This article, based on enterprise informatization practices, explores how to achieve deep integration of multiple systems, including digital Office Automation (OA), Electronic Human Resource (eHR) systems, Enterprise Resource Planning (ERP) systems, and Manufacturing Execution Systems (MES), through systematic software interface development. Starting with requirements analysis, the article systematically elaborates on interface design methods (including function call patterns and intermediate database patterns) and specific system integration paths, such as single sign-on and integration solutions between OA and eHR systems. Practical results show that this interface-based data fusion strategy can effectively break down information barriers, achieve seamless data flow and sharing, promote the collaborative operation of cross-system business processes, and thus significantly improve enterprise operational efficiency, decision support capabilities, and overall competitiveness, providing a referable path and method for similar enterprises to promote digital transformation.
2026, 42(10):194-196.
Abstract:In industrial automation systems, instrument selection and control system integration are directly related to the stability, safety, and economy of the production process. Aiming at the problems existing in traditional industrial systems, such as unreasonable instrument matching, low system integration, and insufficient data utilization, this paper systematically analyzes the principles and methods of automation instrument selection from an engineering application perspective, and proposes a control system integration scheme based on layered architecture. Through application verification in typical industrial scenarios, this scheme demonstrates significant advantages in system stability, data real-time performance, and operation and maintenance efficiency, providing a reference for the construction of industrial automation systems.
2026, 42(10):197-199.
Abstract:With the acceleration of urbanization, the scale of urban municipal drainage pipe networks continues to expand, and the pressure on their operation and maintenance is increasingly growing. Traditional manual inspection methods suffer from problems such as time consumption, labor intensity, and high false detection rates. To address this, an intelligent inspection method for municipal drainage pipeline defects based on computer vision is proposed. This method covers image acquisition, preprocessing, deep learning model training, and defect image recognition, enabling intelligent detection and classification of common defects such as cracks, blockages, and corrosion in pipelines. Experimental results show that the defect recognition accuracy of this method exceeds 92%, significantly improving inspection efficiency and providing effective technical support for smart city pipe network maintenance and monitoring.
2026, 42(10):200-202.
Abstract:With the rapid growth of electric vehicle ownership, the uncertainty of charging load imposes significant pressure on power grid operation. Aiming at the problems of dynamic scheduling of energy storage systems and load balancing in charging pile clusters, a scheduling strategy based on improved prediction models and optimization algorithms is proposed. This strategy achieves intelligent charge and discharge control of energy storage systems through real-time monitoring of electric vehicle charging demand and grid load variations, thereby smoothing load curves, reducing peak-to-valley differences, and improving the operational efficiency of charging pile clusters. Simulation experimental results show that this method outperforms traditional static scheduling strategies in both load balancing effect and energy storage utilization rate, demonstrating good practical application value.
2026, 42(10):203-205.
Abstract:Continuous Fiber Reinforced Composites (CFRC) have been widely used in aerospace, automotive, and mechanical manufacturing industries due to their high specific strength and high specific modulus. Mechanical 3D printing technology provides a new approach for rapid prototyping of CFRC, but limited by the printing process and fiber-matrix interfacial performance, the mechanical properties of the printed parts are generally lower than those produced by traditional processes. Based on the forming mechanism of mechanically 3D printed CFRC, the effects of fiber laying path, matrix melt flow, and interfacial bonding behavior on the mechanical properties of the composites are analyzed, and optimization strategies for interfacial performance are proposed.
2026, 42(10):206-208.
Abstract:Leveraging its advantages of ultra-large bandwidth, low loss, and high reliability, optical fiber communication has become the core supporting technology for Data Center Interconnect (DCI). This paper systematically analyzes the application status, key technologies, and development trends of optical fiber communication in DCI. By comparing different transmission schemes, the performance and economy of optical modules, wavelength division multiplexing, coherent transmission, and passive optical network technologies in DCI scenarios are discussed, providing a reference basis for the selection of optical interconnection schemes for data centers.
WANG Manli , LIANG Shunqiang , XU Hongshan , DOU Wenchan
2026, 42(10):209-211.
Abstract:The four western hub nodes in the "East Data, West Computing" project face the challenge of homogeneous competition, especially given differences in resource endowments. Taking Gansu Province as the research object and through comparative analysis with other western hubs, this paper reveals Gansu's core competitiveness (including its wind and solar resource potential, geographical centrality as a geometric center, and industrial foundation) as well as its shortcomings, such as lagging energy infrastructure, low-end lock-in of the industrial chain, and insufficient talent pool. It proposes optimization pathways including breaking energy bottlenecks, promoting regional coordinated development, and activating local demand, aiming to achieve a transformation from a "computing power receiver" to a "benchmark for green computing in western China." The research findings provide a reference for the differentiated layout of western hub nodes and the upgrading of Gansu's digital economy.
2026, 42(10):212-214.
Abstract:This paper proposes a 3D real-scene reconstruction method integrating UAV imagery acquisition, laser scanning, and multi-source data fusion for fine modeling of river channels and water conservancy engineering structures. By applying Digital Surface Model (DSM), point cloud processing, and texture mapping techniques, high-precision and interactive 3D visualization is achieved. A case study of a reservoir and its river channel verifies the effectiveness of this method in improving flood risk analysis, dam inspection, and dam break simulation. Experimental results show that the method can control the modeling accuracy within 5 cm while significantly enhancing the efficiency of disaster prevention decision-making.
WANG Wei , WANG Tuo , WANG Chenyu , SUN Jianyu , PENG Fuxiang
2026, 42(10):215-219.
Abstract:The ventilation system in a nuclear auxiliary building is used to maintain the temperature conditions required for equipment operation and personnel health during normal power plant operation, and to provide necessary airflow and filtration for the containment ventilation system during reactor cold shutdown. This article systematically analyzes and summarizes the composition, operating conditions, operating specifications, relevant design changes during the commissioning phase, and lessons learned by various power plants during production and operation of the nuclear auxiliary building ventilation system.
2026, 42(10):220-222.
Abstract:With the introduction of technologies such as big data, the Internet of Things (IoT) , Building Information Modeling (BIM) , cloud computing, and artificial intelligence, construction engineering management is accelerating its transformation towards digitalization and intelligence. This paper analyzes the specific applications of digital technology in project schedule management, quality control, safety management, and cost control, discusses its implementation effects and existing problems, and proposes corresponding optimization paths, aiming to provide a reference for improving construction engineering management levels.
2026, 42(10):223-225.
Abstract:In Border Gateway Protocol (BGP) networks, as data traffic continues to grow, the buffering pressure on routing nodes keeps increasing, leading to increasingly prominent network congestion issues, which in turn result in higher packet loss rates and increased transmission delays. To address the above problems, this paper first establishes a BGP anomaly congestion model. Then, a BGP network anomaly detection method based on MODE-LSTM is proposed, and the effectiveness and stability of the method are verified through experiments. The experimental results show that the proposed method achieves a detection rate of 94.12%, reduces the feature dimension from 26 to 15, and maintains good detection performance and stability while realizing feature dimensionality reduction.
2026, 42(10):226-228.
Abstract:In the industrial control, capacitive sensors are widely used to measure physical quantities such as pressure, displacement, and humidity. This paper analyzes the operating characteristics of capacitive sensors using Proteus simulation software, with a focus on simulating bridge circuits, operational amplifier measurement circuits, and resonant circuits. The experimental results verify the response characteristics of capacitive sensors in different circuits, providing data support for their application in industrial inspection.
ZHENG Huilin , YANG Qingshan , LI Haibo , LI Jianghua , ZHANG Hongyuan
2026, 42(10):229-231.
Abstract:Based on the whole-process dynamic visualization project of Materials Company of State Grid Shandong Electric Power Company, this paper proposes a "Smart Fast Lane" model for power emergency material logistics, achieving the goals of "resource lock-in within 30 minutes and cross-city arrival within 7 hours". First, a region-wide resource pool covering 460 suppliers and 131 warehouses was constructed. Second, leveraging the Shandong Electric Power MAP platform, intelligent matching and one-click dispatch were realized. Combined with the SGI Map and intelligent logistics terminals, visualized monitoring of the entire transportation process was achieved. Through a comparative analysis of outbound efficiency and transportation loss rates before and after a typical emergency event, the effectiveness of this model in improving dispatch efficiency and transportation reliability is verified. The research shows that this model can significantly enhance the support capability for power emergency materials and provide a replicable technical pathway for emergency logistics dispatch in power systems.
CHENG Yunxiang , LI Haibo , LI Pengfei , ZHANG Yuyang , ZHANG Baifu
2026, 42(10):232-234.
Abstract:With the continuous expansion of power engineering projects, material management is facing problems such as large quantities, complex types, low efficiency of arrival registration, and difficulty in ensuring data accuracy. This paper is based on the research project of automatic registration of material arrival at the project site of Materials Company of State Grid Shandong Electric Power Company in 2025, and proposes an automatic registration scheme based on edge computing of the Internet of Things (IoT). This solution integrates RFID, QR code recognition, and intelligent receiving devices to achieve rapid identification and automatic registration of on-site materials, and focuses on the research of data accuracy verification methods for incoming materials. Through deployment experiments on actual engineering sites, the results show that the system can significantly improve the efficiency of material arrival registration, with a data accuracy rate of over 98%, providing feasible references for the informatization and intelligence of power engineering material management.
2026, 42(10):235-237.
Abstract:In the digital economy era, data has become a key production factor driving economic and social development, and human resource management is gradually shifting from traditional experience-based management to data-driven management. This paper analyzes the development trend of digital transformation in human resource management against the backdrop of the digital economy, explains the basic connotation and core components of data literacy, and discusses the current problems faced by human resource managers in terms of data awareness, data analysis, and data application. It proposes countermeasures such as improving training systems, strengthening technology application, and enhancing management mechanisms, aiming to help organizations improve their HR management capabilities and digital governance capacity.
2026, 42(10):238-240.
Abstract:Based on digital twin technology, an early warning and control model for construction project cost overrun is constructed. Through real-time data acquisition, simulation analysis, and risk prediction, dynamic cost management is achieved. Research results show that the digital twin-driven early warning system can significantly improve cost monitoring accuracy, identify potential risks in advance, and provide scientific basis for engineering decision-making. The proposed control measures provide a feasible path for construction enterprises to reduce costs and optimize resource allocation.
2026, 42(10):241-243.
Abstract:Aiming at the problems of "information silos", easy data tampering, and difficulties in sharing maintenance records among departments in the procurement, use, maintenance, and disposal stages of medical equipment, a blockchain-based trusted traceability and maintenance record sharing mechanism for the full lifecycle of medical equipment is proposed. By constructing a consortium blockchain and utilizing smart contracts to achieve automatic recording of equipment status and permission control, combined with the Internet of Things and information systems, the fusion of multi-source heterogeneous data is accomplished.
2026, 42(10):244-246.
Abstract:This paper analyzes the operational mechanisms of integrated technologies combining Industrial Ethernet, 5G wireless communication, and OPC UA. Addressing the "information silo" problem prevalent in traditional automation systems, it proposes a cloud-edge-device collaborative industrial network architecture and constructs a multi-protocol fusion communication solution. Based on this, a simulated industrial environment is designed and built to compare the real-time performance of three typical network architectures. Experimental results show that the network architecture integrating TSN and 5G achieves a control command transmission latency of less than 3.2 ms, a system reliability of 99.998%, and a data acquisition integrity rate exceeding 99.5% for OPC UA.
ZHANG Yanwei , SI Songsong , MENG Binbin
2026, 42(10):247-249.
Abstract:With the continuous improvement of construction industrialization and intelligence levels, the complexity and system coupling degree of mechanical and electrical installation engineering have significantly increased. Safety risks exhibit dynamic, concealed, and multi-source characteristics. Traditional safety management methods are no longer able to meet modern engineering requirements. Based on the concept of intelligent management, a safety risk prevention and control system for mechanical and electrical installation engineering is constructed, and systematic research is conducted around risk identification, dynamic monitoring, intelligent early warning, and closed-loop management. Combined with engineering practice, the effectiveness of this method in improving safety management efficiency and reducing accident rates is verified.
2026, 42(10):250-252.
Abstract:With the advancement of new power system construction, the scale of multi-source heterogeneous data generated by power grid operations continues to grow. Traditional centralized cloud data governance models face challenges such as bandwidth constraints, high latency, and "data silos" in highly real-time and distributed scenarios. To address these issues, this paper proposes an efficient data governance method for power grid data based on a cloud-edge collaborative architecture. Data preprocessing and quality control are performed at the edge side, while global modeling and policy optimization are carried out at the cloud side, establishing a governance mechanism characterized by "edge autonomy and cloud coordination". Experimental results demonstrate that the proposed method offers significant advantages in reducing transmission latency, improving processing efficiency, and enhancing data quality.
2026, 42(10):253-255.
Abstract:Bridges are key components of urban road networks, and their safety is directly related to urban public safety. Traditional bridge crack detection relies on manual inspection, which suffers from problems such as low efficiency, strong subjectivity, and high safety risks. To address these issues, an intelligent bridge crack detection and recognition method based on the YOLO (You Only Look Once) deep learning algorithm is proposed. By constructing a bridge crack image dataset and training and improving the YOLO model, accurate and real-time recognition of bridge cracks is achieved. Experimental results show that this method can effectively identify various types of cracks in complex bridge environments, with high detection efficiency and recognition accuracy, providing important guidance for bridge structural health monitoring.
2026, 42(10):256-258.
Abstract:Aiming at the inadequate adaptation of existing systems on domestic software and hardware platforms, this paper proposes a systematic optimization scheme. By analyzing the architecture and operation mechanism of typical configuration software (e.g., ForceSCADA), key system call conflicts and performance bottlenecks are identified. On this basis, technical improvements are carried out in aspects such as kernel call mapping, driver interfacing, protocol stack middleware construction, graphical component replacement, and thread management optimization. Combined with static scanning, dynamic tracing, and event-driven mechanisms, the stable operation and efficient adaptation of the configuration software on domestic operating systems (e.g., Kylin V10 and UnionTech UOS) are achieved. The research results indicate that this method can effectively enhance system compatibility and operational efficiency, providing a feasible technical path and implementation reference for the domestic upgrade of oilfield informatization systems under the Information technology application innovation system.
WANG Ranran , LI Shanshan , WANG Qingyuan
2026, 42(10):259-261.
Abstract:With the acceleration of urbanization, urban transportation systems are facing increasingly complex travel demands. How to efficiently understand residents' travel behavior and provide personalized recommendations has become a research hotspot. A method integrating trajectory mining and knowledge graph is proposed, combining urban transportation big data to construct a travel behavior model and achieve travel pattern analysis and intelligent recommendation. Firstly, GPS trajectory data is used to mine users' regular travel patterns and potential points of interest. Secondly, a knowledge graph is utilized to traffic facilities, geographic information, and user behaviors, achieving semantically enhanced travel pattern representation. Finally, personalized travel recommendations are provided to users based on the graph and trajectory features. Experimental results show that this method outperforms traditional methods in both travel pattern recognition accuracy and recommendation satisfaction, providing a reference for smart city traffic management and travel service optimization.
2026, 42(10):262-263.
Abstract:A dynamic construction progress management method for electromechanical equipment integrating Building Information Modeling (BIM) and Real-Time Locating System (RTLS) is proposed. BIM is used to construct a construction information database, combined with RFID, UWB and other technologies to dynamically collect equipment positioning and construction status, and relevant information is presented in manner. Research shows that this method can achieve real-time monitoring of construction progress, predict potential delay risks, assist managers in rapid decision-making, and provide technical support for refined management of electromechanical installation engineering.
LU Mengzhu , ZHANG Feihong , SU Haotian , SHI Ruihang
2026, 42(10):264-266.
Abstract:An illegal mining identification method for open-pit mines based on image segmentation is proposed. Surface information of mining areas is obtained through high-resolution remote sensing imagery, and deep learning-based image segmentation technology is used to automatically identify abnormal mining areas. The research employs the U-Net network for remote sensing image segmentation and incorporates post-processing methods to optimize recognition accuracy. Experimental results show that this method achieves high accuracy and recall in detecting illegal mining areas, providing data support and technical means for mining area supervision, environmental protection, and law enforcement.
Address:重庆市渝北区洪湖西路 18 号西南信息楼Zip Code:401121
Phone:023-63507395 Email:jsjyywzbjb@163.com
® 2026 All Rights Reserved
渝公网安备50019002505578号
京ICP备09084417号-24
Supported by:Beijing E-Tiller Technology Development Co., Ltd.