A new approach to smart course construction for electric vehicle design is emerging as universities seek to align engineering education with the demands of new quality productive forces and the rapid development of artificial intelligence. The proposed paradigm combines digital teaching resources, knowledge graphs, an AI-supported smart course platform, SPOC-based blended learning, teacher-student-machine interaction, and diversified process assessment. Its central aim is to strengthen students’ innovation capacity, digital literacy, and ability to work across disciplines in the electric vehicle field. By integrating large language models, modular course design, personalized learning paths, and data-driven teaching improvement, the model offers a structured response to long-standing challenges in electric vehicle design education, including strong interdisciplinary complexity, slow content renewal, limited teaching methods, and incomplete assessment practices.
- Background and Rationale for the Electric Vehicle Design Course Reform
Global economic change and technological revolution have made new quality productive forces a central engine of high-quality social and economic development. The cultivation of these productive forces requires new quality talents who can think creatively, use digital tools competently, and adapt to fast-changing technical environments. Higher education therefore faces a pressing task: to explore how the ideas behind new quality productive forces can be deeply integrated with digital and intelligent technologies in order to improve talent training quality and learning effectiveness. This task is especially urgent in fields such as electric vehicle design, where technological change is continuous and where graduates must be prepared to work at the intersection of mechanical engineering, automation, electrical engineering, computer science, and materials science.
Artificial intelligence has expanded rapidly in education and has created both opportunities and challenges for teaching reform. In course content optimization, AI can analyze existing teaching resources, help teachers plan course content more systematically, support the production of teaching materials, and improve teaching efficiency. In teaching method reform, AI can generate customized learning paths and learning resources based on students’ learning conditions and interests, thereby promoting independent learning and deeper learning. AI can also analyze student learning data to assess learning outcomes more accurately and provide a stronger basis for improving teaching models. These capabilities are particularly relevant to electric vehicle design courses, because such courses require students to connect theory, design methods, system-level thinking, and frontier technology awareness.
Electric vehicle design courses involve multiple disciplinary fields, including machinery, automation, electrical systems, computing, and materials. Teaching in this area commonly faces problems such as strong interdisciplinary complexity, slow updating of teaching content, a single teaching method, and one-sided assessment. These problems can limit the development of students’ engineering judgment and innovation ability. To address them, the proposed paradigm deeply integrates digital resource construction with smart course platform development and implements SPOC blended teaching practice. The goal is to explore a new model for smart course construction in electric vehicle design under the vision of new quality productive forces, so that electric vehicle education can better serve the development of new quality talents.
- Overall Architecture of the Proposed Smart Course Paradigm
The overall architecture of the new paradigm for electric vehicle design smart course construction reflects four connected features: digitalization, intelligence, precision, and personalization. Through digital means, teaching resources are enriched and reorganized. Through intelligent technologies, teaching design is supported by teacher-student-machine interaction. Through precise content delivery, learning resources can be pushed according to student needs and learning conditions. Through personalized learning path design, students can engage with electric vehicle design content in ways that match their interests, progress, and development goals. The paradigm also builds a diversified assessment system and seeks to improve students’ innovation ability and digital literacy, with the larger aim of cultivating innovation-oriented talents with the technical characteristics required in the digital and intelligent era.
The architecture links digital teaching resource construction, intelligent teacher-student-machine teaching practice, and a diversified assessment system. Digital teaching resources include digital textbooks, knowledge graphs, core knowledge point videos, standardized question banks, and frontier technology resources. Intelligent teaching practice includes AI-supported teaching, AI-supported learning, AI-supported practice, AI-supported research, AI-supported examination, and AI-supported evaluation. Personalized and precise functions include group project topic selection, personalized learning paths, and precise delivery of learning content. Together, these elements support the improvement of innovation ability and digital literacy and contribute to the development of innovation-oriented talents for the digital and intelligent age. The model is designed not as a single technological add-on but as an integrated teaching ecosystem for electric vehicle design education.
In the context of electric vehicle design, this architecture matters because the subject cannot be taught effectively through isolated lectures alone. Electric vehicle systems involve batteries, motors, electronic controls, and whole-vehicle integration, and each area interacts with the others. A smart course environment can help students see these connections, revisit core concepts, explore frontier topics, and receive timely feedback. It can also help teachers observe learning behavior and adjust instruction. The paradigm therefore treats the electric vehicle course as a living system of knowledge, practice, and assessment rather than a fixed sequence of content.
- Digital Teaching Resource Integration Based on Knowledge Graphs and Modular Design
Combining graduation requirements and course objectives, the teaching content is divided into a basic theory module, a design module, and a frontier technology module. Large language models are used to mine and analyze existing teaching resources, extract keywords, identify topics, and recognize relationships among knowledge points. The extracted knowledge points and structured information are integrated as nodes in a course knowledge graph. The logical relationships among knowledge points are clarified, and a tree-like or network-like structure is formed. This structure ensures that the knowledge graph reflects both the independence of individual knowledge points and the systematic nature of the course, thereby demonstrating the frontier, practical, and systematic character of course content. For electric vehicle design education, this approach helps organize a field that might otherwise appear fragmented because of its many disciplines and fast-moving technologies.
In the electric vehicle design course, the digital resource construction process begins with the professional training plan and course objectives. The course objectives define the abilities that students should possess after learning the Electric Vehicle Design Fundamentals course. Large language models are then used to identify key information and construct core modules from teaching resources such as course syllabi, digital textbooks, and past examination questions. Combined with development trends in the electric vehicle industry, the course content is decomposed into four teaching modules: battery, motor, electronic control, and whole vehicle. The contents of these modules are connected through their internal relationships and form a network-like knowledge graph. This structure supports the teaching of electric vehicle design as an integrated engineering field.
Based on the knowledge graph, core knowledge points are refined, including performance indicators of power batteries, operating characteristics of motors, and design of energy management strategies. Core knowledge point videos are recorded for students to preview before class and review after class. Each teaching module includes multiple core knowledge point videos, such as basic concept introductions, working principle animations, and structural analysis of typical components. The digital resources are not merely supplementary materials; they become the shared reference base for classroom teaching, online learning, project work, and assessment. In this way, the electric vehicle design course gains a coherent digital foundation that supports both self-directed learning and teacher-guided instruction.
The use of knowledge graphs also strengthens the relationship between basic theory and design practice. In an electric vehicle design course, students must understand why a battery performance indicator matters for whole-vehicle design, how motor operating characteristics influence control strategies, and how energy management decisions affect system performance. By making these relationships visible, the knowledge graph helps students move from memorizing isolated facts to building an engineering mental model. It also gives teachers a tool for diagnosing gaps in student understanding and for deciding where to add explanations, examples, demonstrations, or project tasks. The graph can be updated as electric vehicle technology develops, which supports the continuous renewal of course content.
- Smart Course Platform Built on AI and the Smart Tree Environment
According to the constructed course knowledge graph, a matrix of core knowledge points is organized, core knowledge point short videos are recorded, and online course resources are built on the Smart Tree online education platform for students to preview before class and review after class. An AI-based diversified question bank is built to assess students’ mastery of core knowledge points in the knowledge graph before or after class. Large language models are used to support project-based teaching, generate personalized project topics, meet students’ diverse learning interests and needs, and promote students’ all-round development. The platform thus becomes an operational environment for electric vehicle design learning, not just a storage space for files.
Teachers rely on the AI and Smart Tree online education platform to build the smart course, upload teaching slides, knowledge graphs, core knowledge point videos that are required viewing, frontier electric vehicle technology videos that are optional viewing, post-class extension literature, question banks, project materials, and other digital teaching resources. Teachers also use the platform for online teaching and teaching management, including publishing course tasks that are either elective or required, initiating group discussions, organizing students to choose design topics individually, and conducting chapter online quizzes. Students use the platform for online learning, watching video courses, participating in online discussions, completing assignments, and taking quizzes. Based on student learning behavior data provided by the smart course platform, teachers can understand students’ learning interests and progress in a timely manner and adjust teaching content accordingly.
For electric vehicle design, the platform supports a blended learning experience in which students can prepare for complex classroom activities by reviewing core concepts online. The required videos help ensure that students have a common baseline of knowledge before class, while optional frontier videos allow interested students to go further into electric vehicle technologies. The AI-based question bank provides flexible assessment opportunities. The project-based teaching function supports personalized topics and encourages students to connect course knowledge with real electric vehicle design problems. The platform also creates a record of learning behavior that can be used to improve teaching in subsequent cycles. This data-driven approach is particularly valuable in electric vehicle education, where student backgrounds may vary and where prior knowledge in mechanical, electrical, and computing areas may differ.
The platform also supports teacher development. AI-assisted teaching design encourages teachers to keep learning new knowledge, improve professional quality, strengthen digital literacy, and enhance teaching design ability. At the same time, it can improve classroom teaching efficiency and stimulate students’ innovative thinking and problem-solving ability. In the electric vehicle design context, teachers must stay informed about battery technology, motor technology, electronic control, and whole-vehicle integration. A smart platform does not replace teacher expertise; instead, it helps teachers organize resources, monitor learning, and respond to student needs more effectively. The platform therefore functions as a support for professional judgment, not a substitute for it.
- Personalized Innovative Teaching Practice Based on AI and SPOC
Based on the SPOC teaching model, the advantages of offline classroom teaching and online course teaching are combined to form a teaching model of online introduction, online and offline participation, and online detection. AI technology is used to record data on pre-class optional tasks, in-class online detection, and student learning behavior, so that students’ mastery of knowledge points can be understood in real time. AI technology also supports the provision of personalized and precise project topic suggestions. AI-assisted teaching promotes communication and feedback between teachers and students, forms a teacher-machine-student three-way interaction scenario, and helps obtain multidimensional information about student learning status efficiently. This design is intended to make electric vehicle design learning more responsive, more interactive, and more closely connected to students’ individual development.
The SPOC blended teaching process for the Electric Vehicle Design Fundamentals course is divided into three parts: introduction, interactive teaching, and detection. The teacher-student-machine three-way interactive teaching design framework is built on an improved BOPPPS teaching model, which includes Bridge-in, Objective, Pre-assessment, Participatory Learning, Post-assessment, and Summary. The BOPPPS model emphasizes teaching flexibility and adaptability. Teachers can adjust activities in each stage according to specific teaching content and student learning conditions. For example, when students carry out personalized project topic selection, teachers can use AI learning analysis to add corresponding topic content in response to learning hotspots. This flexibility is important in electric vehicle design education because student interests and industry priorities can change, and because design problems often require iterative exploration.
In the introduction stage, teachers supervise and guide, produce digital teaching resources, publish pre-class learning content, and answer students’ platform questions. Students learn the knowledge graph content independently, complete required pre-class content, and complete optional frontier content. This stage supports pre-class knowledge preparation. In the interactive teaching stage, teachers lead interaction, teach according to the course syllabus, raise questions, guide discussion, and summarize the knowledge graph. Students participate in interaction, answer random classroom questions, join project group discussions, and take part in classroom questionnaires. This stage is delivered through the SPOC model. In the detection stage, teachers supervise and guide, design chapter quiz questions, design post-class assignment questions, and design project task books. Students complete post-class assignments online, complete chapter quizzes within a time limit, and write group project reports. This stage supports post-class independent review. For electric vehicle design, these stages help students move from basic concepts to design participation and then to reflective assessment.
Personalized topic selection is a key part of the teaching practice. In the Electric Vehicle Design Fundamentals course, the organization of student personalized topic selection is supported by data-driven AI learning analysis, topic direction recommendations, and student independent personalized topic selection. Topic directions cover battery, motor, electronic control, and whole vehicle areas. Battery-related topics include battery pack lightweight and structural design, power battery thermal management system design, battery pack balancing control strategy research, and battery state of health prediction algorithm research. Whole-vehicle and electronic control topics include electric vehicle energy management system design, electric vehicle vehicle controller design, electric vehicle energy recovery system design, and DC-DC converter efficiency parameter design. Powertrain topics include hybrid electric vehicle powertrain parameter design, pure electric vehicle powertrain parameter design, extended-range electric vehicle powertrain parameter design, and fuel cell electric vehicle powertrain parameter design. Motor topics include permanent magnet synchronous motor parameter optimization design, induction motor rotor resistance optimization design, reluctance motor reluctance torque optimization design, and electric vehicle drive motor control strategy.
These topic directions show how an electric vehicle design course can connect foundational knowledge with design exploration. Students can select topics according to their interests, while teachers can use AI learning analysis to recommend directions and provide guidance. The teacher-machine-student interaction scenario allows teachers to monitor progress, answer questions, and provide feedback. It also allows students to receive support that is more precisely matched to their project needs. Because electric vehicle design requires both system-level thinking and component-level analysis, personalized topics can help students develop depth in a particular area while still understanding how that area relates to the whole vehicle. The approach also supports the development of digital literacy, since students use online resources, data, simulation tools, and AI-supported learning environments as part of their design work.
- Diversified Process Assessment for New Quality Talent Development
Process assessment encourages students to participate continuously in learning activities throughout the teaching process. AI technology can obtain records of student online learning behavior, which helps stimulate students’ learning motivation. The process assessment of the Electric Vehicle Design Fundamentals course is divided into online assessment at 50 percent and offline assessment at 50 percent. Online assessment includes five parts: pre-class tasks, both elective and required, at 5 percent; post-class assignments at 5 percent; classroom participation at 5 percent; chapter quizzes at 10 percent; and personalized project reports at 25 percent. Offline assessment is the final closed-book examination at 50 percent. This structure combines continuous learning evidence with a summative examination and reflects a more comprehensive view of student development in electric vehicle design education.
| Assessment Type | Component | Weight (%) |
|---|---|---|
| Online assessment | Pre-class tasks, elective and required | 5 |
| Online assessment | Post-class assignments | 5 |
| Online assessment | Classroom participation | 5 |
| Online assessment | Chapter quizzes | 10 |
| Online assessment | Personalized project report | 25 |
| Online assessment | Online subtotal | 50 |
| Offline assessment | Final closed-book examination | 50 |
Based on AI technology, the assessment standards are designed to be more comprehensive and cover students’ online course participation, group project cooperation attitude, problem-solving ability, and technology application ability. Diversified process assessment methods are used to evaluate students’ innovation ability and digital literacy, such as learning analysis reports, project reports, regular assignments, final examinations, online quizzes, optional tasks, and group discussions. These methods help ensure the accuracy and comprehensiveness of evaluation results. In electric vehicle design, where design ability and system thinking are difficult to capture through a single examination, a diversified assessment system can provide a fuller picture of student learning. It can also give teachers more actionable information for improving instruction.
The assessment system also reinforces the goals of the smart course. Online learning behavior data can reveal whether students are engaging with core knowledge point videos, optional frontier content, quizzes, and project materials. Project reports and group discussions can show how students apply electric vehicle design knowledge to open-ended problems. Classroom participation can capture the quality of student interaction during SPOC sessions. The final closed-book examination provides a summative check of foundational knowledge. Together, these assessment components support the development of new quality talents by valuing not only final answers but also the process of learning, collaboration, application, and reflection. The assessment design therefore aligns with the broader paradigm of digitalization, intelligence, precision, and personalization.
- Implementation Effects and Continuous Improvement in Electric Vehicle Education
Through the smart course teaching platform, teachers can collect student behavior data within a teaching cycle and precisely identify students’ personalized learning hotspots and error-prone knowledge points. The online learning hotspot distribution for the Electric Vehicle Design Fundamentals course shows that, in the frontier elective module, students’ online learning hotspots are mainly reflected in frontier research on electric vehicle batteries and motors. For example, the click rate for solid-state batteries is 95.3 percent, and the question volume for flat wire motors reaches 87 questions. In response to students’ personalized learning hotspots, teachers can make targeted teaching designs in the next teaching cycle, such as adding corresponding electric vehicle frontier knowledge videos, design topics related to learning hotspots, and group discussions on learning hotspots, so as to improve students’ innovative thinking and mastery of frontier knowledge.
| Learning Hotspot | Click Rate (%) |
|---|---|
| Solid-state battery | 95.3 |
| Blade battery | 94.5 |
| Flat wire motor | 91.2 |
| Fuel cell | 88.6 |
| Energy recovery | 87.9 |
The statistics on error-prone points in chapter quizzes show that the accuracy rate for calculations related to electric vehicle power batteries is relatively low. Teachers can refine the teaching videos for related knowledge points and present them to students in smaller video units, helping students quickly locate error-prone points and engage in repeated learning. At the same time, teachers can train the Smart Tree AI teaching assistant on students’ error-prone knowledge points and provide carefully selected supporting materials. By using the AI teaching assistant, students can quickly and accurately learn error-prone knowledge points. According to platform statistics on chapter quizzes, 17.0 percent of students had difficulty accurately drawing the configuration diagram of hybrid electric vehicles. Therefore, teachers can add demonstrations of various hybrid electric vehicle configurations and discussions of their advantages and disadvantages in the classroom discussion session, push customized learning materials to students, and improve the score rate for this knowledge point.
| Error-Prone Point | Score Rate (%) |
|---|---|
| Calculations related to power batteries | 80.0 |
| Calculations related to motor parameters | 82.0 |
| Configuration diagram of hybrid electric vehicles | 83.0 |
| Powertrain parameter matching | 83.0 |
| Energy management strategy | 85.0 |
These findings illustrate how data from the smart course platform can support continuous improvement in electric vehicle design teaching. When students show strong interest in solid-state batteries, blade batteries, flat wire motors, fuel cells, and energy recovery, teachers can respond by updating frontier content and designing project topics that connect these hotspots with electric vehicle design fundamentals. When students struggle with power battery calculations, motor parameter calculations, hybrid electric vehicle configuration diagrams, powertrain parameter matching, and energy management strategies, teachers can adjust video granularity, add demonstrations, provide customized materials, and improve AI teaching assistant support. The improvement process is therefore based on observed learning behavior and quiz performance rather than on assumption alone.
Through the construction of course knowledge graphs and the smart course platform, the traditional automobile design teaching model can be transformed toward digitalization and intelligence, creating a personalized and precise learning ecology for students. AI-assisted teaching design encourages teachers to continuously learn new knowledge and improve professional quality, digital literacy, and teaching design ability. While effectively improving classroom teaching efficiency, it also continuously stimulates students’ innovative thinking and problem-solving ability. For electric vehicle education, this means that the course can remain connected to industry trends and technological innovation while still providing students with a structured path through foundational concepts and design methods.
- Implications for Electric Vehicle Design Curricula and Smart Course Development
The significance of the proposed paradigm extends beyond a single course. Electric vehicle design is a field in which knowledge from mechanical engineering, automation, electrical engineering, computer science, and materials science must be integrated. A smart course that uses knowledge graphs can make the structure of this interdisciplinary field more visible to students and teachers. A platform that supports AI and SPOC can provide flexible learning opportunities and timely feedback. A diversified assessment system can recognize the many forms of competence that electric vehicle design requires. Together, these elements can help cultivate new quality talents who are capable of innovation, digital literacy, and continuous learning in a rapidly changing technical environment.
The paradigm also responds to the need for content renewal. Electric vehicle technology is developing quickly, and course content can become outdated if it relies only on static textbooks and traditional lectures. By using large language models to analyze teaching resources, by building and updating knowledge graphs, and by incorporating frontier technology videos and optional learning materials, the course can refresh its content more efficiently. The smart course platform can support the rapid addition of new electric vehicle topics, the design of new project tasks, and the updating of question banks. This capacity for renewal is essential if electric vehicle design education is to remain relevant to industry development and technological innovation.
Another implication concerns personalized learning. Students in electric vehicle design courses may have different strengths and interests. Some may be drawn to battery thermal management, others to motor optimization, electronic control, energy management, or whole-vehicle integration. A smart course environment can offer personalized learning paths, precise content delivery, and project topic recommendations. It can also help students identify areas where they need further study. The teacher-student-machine interaction scenario does not remove the teacher from the process; rather, it gives teachers more information and more tools for guiding students. In this sense, the paradigm supports a more responsive form of teaching that can address both common learning goals and individual development needs.
The paradigm further highlights the importance of digital literacy for both students and teachers. Students interact with online resources, knowledge graphs, AI teaching assistants, digital question banks, and project materials. Teachers use platform data, AI learning analysis, and digital resource development tools. These activities help build the digital competence that new quality talents need. At the same time, the paradigm emphasizes that technology should serve teaching and learning goals. The purpose of AI is not to replace professional judgment but to support the organization of electric vehicle design knowledge, the delivery of personalized learning, and the improvement of assessment and feedback.
For electric vehicle design programs, the new paradigm offers a practical route for integrating digital resources, intelligent platforms, blended teaching, and diversified assessment. It can be adapted to different course contexts and different student groups. The core elements remain consistent: a knowledge graph that organizes electric vehicle design content, a smart platform that supports online and offline learning, AI tools that assist teaching and learning, SPOC-based interaction, and assessment methods that value process and innovation. When these elements are aligned, the course can better support the development of students’ engineering thinking, design ability, and capacity for lifelong learning in the electric vehicle field.
- Conclusion
The construction of a digital teaching resource system, an intelligent teaching platform, and an efficient teacher-student-machine three-way interactive teaching design has enabled precise delivery of teaching content and met students’ personalized learning needs. It has also completed multidimensional resource integration based on large language models and significantly improved teaching effectiveness. The AI-enabled knowledge graph injects innovative momentum into teaching content and methods, helps update course content quickly according to industry development trends and technological innovation, and ensures the frontier and contemporary character of teaching content. The smart course platform can stimulate students’ learning interest through personalized learning path recommendations and interactive learning tools, helping students improve innovation ability and digital literacy. The new paradigm for smart course construction in electric vehicle design, oriented toward new quality talent development, is worthy of broad application and promotion in the digital and intelligent transformation of higher education.
The proposed paradigm demonstrates how electric vehicle design education can move toward digitalization, intelligence, precision, and personalization. It addresses the interdisciplinary complexity of electric vehicle design, the need for continuous content renewal, the limitations of single teaching methods, and the shortcomings of one-sided assessment. By combining knowledge graphs, AI-supported platforms, SPOC blended learning, BOPPPS-inspired teaching design, personalized project topics, and diversified process assessment, the model creates a coherent learning ecology for electric vehicle design. It supports teachers in improving professional quality, digital literacy, and teaching design ability, and it supports students in developing innovative thinking, problem-solving ability, and digital competence. As electric vehicle technology continues to evolve, such smart course paradigms can help higher education produce new quality talents who are ready to contribute to the advancement of electric vehicle design and the broader transportation transition.