In the context of global economic transformation and rapid technological advancement, new quality productivity has emerged as a core engine driving high-quality socio-economic development. Cultivating and developing new quality productivity urgently demands “new quality talents” with innovative thinking and digital literacy. Exploring the deep integration path of new quality productivity concepts and digital-intelligent technologies, and comprehensively enhancing the quality and learning efficiency of higher education talent cultivation, has become a crucial theoretical and practical topic in current educational reform and innovation.
With the rapid development of Artificial Intelligence (AI) technology, its application in the field of education has become increasingly widespread, bringing new opportunities and challenges to teaching reform. In terms of teaching content optimization, AI can deeply analyze existing teaching resources, assist teachers in systematically planning course content, creating teaching materials, and improving teaching efficiency. Regarding teaching methods reform, AI can generate customized learning paths and resources based on students’ learning situations and interest preferences, promoting autonomous and deep learning. Furthermore, AI can accurately assess students’ learning outcomes through the analysis of learning data, providing a strong basis for improving teaching models.
Electric vehicle design courses involve multiple interdisciplinary fields such as mechanics, automation, electrical engineering, computer science, and materials science. These courses often face issues such as strong interdisciplinary nature, slow updates in teaching content,单一 teaching methods, and片面 assessment approaches. Therefore, this article integrates digital resource construction with smart curriculum platform development and implements SPOC (Small Private Online Course) blended teaching model practices to explore a new paradigm for smart curriculum construction in electric vehicle design under the perspective of new quality productivity.
The overall architecture of the new paradigm for smart curriculum construction in electric vehicle design, aimed at cultivating new quality talents, primarily reflects digitalization, intelligence, precision, and personalization. This is achieved by enriching teaching resources through digital means, enabling precise推送 of teaching content and designing personalized learning paths for students. Leveraging intelligent technology for teacher-student-machine triadic interactive teaching design, and constructing a diversified assessment system, comprehensively enhances students’ innovative capabilities and digital literacy, cultivating innovative constant talents with technical characteristics of the digital-intelligent era.
Integration of Digital Teaching Resources Based on Knowledge Graph and Modular Design
Aligning with graduation requirements and course objectives, the teaching content is divided into basic theory modules, design modules, and cutting-edge technology modules. Utilizing large language models to deeply mine and analyze existing teaching resources, extract keywords, and build themes, identifying key information and the relationships between knowledge points in teaching resources. The extracted knowledge points and structured information are integrated as nodes of the knowledge graph, clarifying the logical relationships between knowledge points to form a tree or network structure, ensuring the knowledge graph reflects both the independence and systematic nature of the knowledge points, embodying the前沿, practicality, and systematic nature of the course content.
For instance, in the “Fundamentals of Electric Vehicle Design” course, based on the professional training plan, course objectives are formulated to clarify the competencies students should possess after completing the course. Using large language models to identify key information and construct core modules from teaching resources such as course syllabi, digital textbooks, and past exam questions, and combining them with the development trends of the electric vehicle industry in China, the course content is decomposed into four teaching modules: battery, motor, electronic control, and vehicle. The content of each teaching module is interconnected through inherent relationships, forming a networked knowledge graph. Based on the knowledge graph, core knowledge points such as performance indicators of power batteries, operating characteristics of motors, and design of energy management strategies are extracted, and core knowledge point videos are recorded for students to preview or review before and after class. Each teaching module includes multiple core knowledge point videos, such as basic concept introductions, working principle animation demonstrations, and typical component structure analyses.
| Teaching Module | Core Knowledge Points | Digital Resource Type |
|---|---|---|
| Battery | Performance metrics, thermal management, state of health prediction | Video, Interactive Simulation |
| Motor | Operating characteristics, control strategies, parameter optimization | Animation, Schematic Diagrams |
| Electronic Control | Energy management, vehicle control unit, DC-DC converter efficiency | Case Studies, Algorithm Code |
| Vehicle | Powertrain parameter matching, configuration diagrams, energy recovery | 3D Models, Design Specifications |
The knowledge graph structure can be represented using graph theory, where each knowledge point \( K_i \) is a node, and the relationship \( R_{ij} \) between knowledge points \( K_i \) and \( K_j \) is an edge. The entire graph \( G \) can be defined as:
$$ G = (V, E) $$
where \( V = \{K_1, K_2, \dots, K_n\} \) is the set of knowledge points, and \( E = \{R_{ij} \mid i,j \in 1,\dots,n\} \) is the set of relationships. This facilitates the organization of content for China EV design courses, ensuring logical progression and coverage of essential topics like battery chemistry and motor dynamics.
Construction of Smart Teaching Platform Based on “AI + Wisdom Tree”
Based on the constructed course knowledge graph, the core knowledge point matrix of the course is sorted out, and short videos of core knowledge points are recorded. Relying on the Wisdom Tree online teaching platform, online resource construction for the course is carried out for students’ pre-class preview and post-class review. Using AI to build a diversified exercise library for assessing students’ mastery of core knowledge points in the knowledge graph before or after class. Empowering project-based teaching with large language models to generate personalized project topics meets students’ diverse learning interests and needs, promoting comprehensive student development.
Teachers rely on the “AI + Wisdom Tree” online education platform to build smart courses, upload teaching PPTs, knowledge graphs, core knowledge point videos for the fundamentals of electric vehicle design (required viewing), cutting-edge technology videos for electric vehicles (optional viewing), post-class extended literature, exercise libraries, project materials, and other digital teaching resources. Simultaneously, teachers need to conduct online teaching and management through this platform, including releasing course tasks (optional and mandatory), initiating group discussions, organizing students to personalized select design topics, and conducting chapter online tests. Students use this platform for online learning, watching video courses, participating in online discussions, completing assignments and tests, etc. Based on the student learning behavior data provided by the smart course platform, teachers can timely understand students’ learning interests and progress, and adjust teaching content at any time.
| Platform Function | Description | Application in EV Design Course |
|---|---|---|
| Resource Upload | Digital materials, videos, knowledge graphs | Battery specs, motor designs, control algorithms |
| Task Management | Release optional/mandatory tasks | Pre-class quizzes on EV components, post-class projects |
| Discussion Forum | Initiate group discussions | Debates on China EV policies, technology trends |
| Assessment Tools | Online tests, personalized projects | Chapter tests on powertrain, custom design reports |
The personalized learning path recommendation can be modeled using an optimization function. Let \( S \) represent a student’s profile, including their learning history, performance, and interests. The goal is to recommend a learning path \( P \) that maximizes learning gain \( G \) while minimizing effort \( E \). This can be formulated as:
$$ \max_P G(P, S) – \lambda E(P, S) $$
where \( \lambda \) is a trade-off parameter. For electric vehicle courses, \( P \) might include modules on battery systems, motor control, and vehicle integration, tailored to the student’s focus areas, such as China EV market needs.
Practice of Personalized Innovative Teaching Model Based on “AI + 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 import-online/offline participation-online detection. Using AI technology to achieve pre-class optional tasks, in-class online detection, and recording of student learning behaviors, real-time mastery of students’ understanding of knowledge points is enabled, providing students with personalized and precise project topic suggestions. Using AI technology to assist teaching promotes communication and feedback between teachers and students, forming a teacher-machine-student triadic interactive scenario, which helps efficiently obtain multidimensional information about students’ learning status.
Based on the improved BOPPPS (Bridge-in, Objective, Pre-assessment, Participatory Learning, Post-assessment, Summary) teaching model, the SPOC blended teaching process of the “Fundamentals of Electric Vehicle Design” course is divided into three parts: import, interactive teaching, and detection. The BOPPPS teaching model emphasizes the flexibility and adaptability of teaching, allowing teachers to adjust activities at each stage according to specific teaching content and student learning situations. For example, during the process of students personalized selecting project topics, teachers can increase corresponding topic content based on AI learning analysis for learning hotspots.
For instance, in the “Fundamentals of Electric Vehicle Design” course, the framework for students’ personalized topic selection is as follows: based on data-driven AI learning analysis, topic direction recommendations and suggestions are provided, leading to students’ autonomous personalized topic selection. Recommended directions include lightweight and structural design of battery packs, thermal management system design for power batteries, research on battery pack balancing control strategies, research on battery health state prediction algorithms, design of energy management systems for electric vehicles, design of vehicle control units for electric vehicles, design of energy recovery systems for electric vehicles, efficiency parameter design of DC-DC converters, parameter design of powertrain systems for hybrid electric vehicles, parameter design of powertrain systems for pure electric vehicles, parameter design of powertrain systems for range-extended electric vehicles, parameter design of powertrain systems for fuel cell electric vehicles, parameter optimization design of permanent magnet synchronous motors, rotor resistance optimization design of induction motors, reluctance torque optimization design of reluctance motors, and control strategies for electric vehicle drive motors.
| Learning Hotspot in Optional Module | Click Rate (%) | Discussion Frequency |
|---|---|---|
| Solid-State Batteries | 95.3 | High |
| Blade Batteries | 94.5 | Medium |
| Flat-Wire Motors | 91.2 | High |
| Fuel Cells | 88.6 | Medium |
| Energy Recovery | 87.9 | Low |
The interactive teaching process can be described using a feedback loop model. Let \( L_t \) represent the learning state at time \( t \), influenced by teaching inputs \( I_t \) and student actions \( A_t \). The evolution can be modeled as:
$$ L_{t+1} = f(L_t, I_t, A_t) + \epsilon_t $$
where \( f \) is a function capturing the learning dynamics, and \( \epsilon_t \) is noise. AI tools monitor \( L_t \) through assessments and participation, adjusting \( I_t \) (e.g., difficulty of EV design tasks) to optimize learning outcomes for topics like China EV energy management.
Improvement of Diversified Process Assessment System for New Quality Talent Cultivation
Based on AI technology, a more comprehensive assessment standard is designed, covering aspects such as students’ online course participation, attitude in group project cooperation, problem-solving, and technology application abilities. Using diversified process assessment methods to evaluate students’ innovative abilities and digital literacy, such as learning analysis reports, project reports, usual assignments, final exams, online tests, optional tasks, group discussions, etc., ensures the accuracy and comprehensiveness of evaluation results.
The process assessment for the “Fundamentals of Electric Vehicle Design” course is divided into two parts: online assessment (50%) and offline assessment (50%). Online assessment includes pre-class tasks (optional + mandatory) (5%), post-class assignments (5%), class participation (5%), chapter tests (10%), and personalized project reports (25%). Offline assessment is the final closed-book exam (50%).
| Assessment Component | Weight (%) | Evaluation Criteria |
|---|---|---|
| Pre-class Tasks | 5 | Completion of optional/mandatory readings on EV basics |
| Post-class Assignments | 5 | Accuracy in solving problems related to motor control |
| Class Participation | 5 | Engagement in discussions on China EV trends |
| Chapter Tests | 10 | Scores on quizzes about battery systems and powertrain |
| Personalized Project Reports | 25 | Innovation in design, e.g., for a new China EV model |
| Final Exam | 50 | Comprehensive knowledge of electric vehicle design principles |
The overall score \( S \) for a student can be computed as a weighted sum:
$$ S = \sum_{i=1}^{n} w_i \cdot s_i $$
where \( w_i \) is the weight of the \( i \)-th assessment component, and \( s_i \) is the score achieved. For example, in the context of electric vehicle courses, \( s_i \) might reflect performance in designing a battery thermal management system for a China EV application, ensuring alignment with industry standards.
Implementation Effects of Smart Curriculum and Continuous Improvement Measures
Teachers can collect student behavior data over one teaching cycle through the smart course teaching platform, accurately identifying students’ personalized learning hotspots and error-prone knowledge points. For the “Fundamentals of Electric Vehicle Design” course, the distribution of students’ online learning hotspots is as shown in previous tables, and the distribution of error-prone points in chapter tests is as follows: for power battery-related calculations, the correct rate is relatively low; for motor-related parameter calculations, the correct rate is also not high; for drawing configuration diagrams of hybrid electric vehicles, some students have difficulties; for powertrain parameter matching, the correct rate is moderate; for energy management strategies, the correct rate is relatively high.
| Error-Prone Point in Chapter Tests | Correct Rate (%) |
|---|---|
| Power Battery-Related Calculations | 80.0 |
| Motor-Related Parameter Calculations | 82.0 |
| Hybrid Electric Vehicle Configuration Diagrams | 83.0 |
| Powertrain Parameter Matching | 83.0 |
| Energy Management Strategies | 85.0 |
Based on the statistics from the platform on chapter tests, a certain percentage of students find it difficult to accurately draw configuration diagrams of hybrid electric vehicles. Therefore, teachers can increase demonstrations of various hybrid electric vehicle configurations and discussions on their advantages and disadvantages during classroom discussion sessions, push customized learning materials to students, and improve the assessment score rate for this knowledge point. For students’ personalized learning hotspots, teachers can conduct targeted teaching designs in the next teaching cycle, such as adding corresponding cutting-edge knowledge videos on electric vehicles, design topics related to learning hotspots, and group discussions on learning hotspots to enhance students’ innovative thinking and mastery of cutting-edge knowledge.
In summary, by constructing course knowledge graphs and smart course platforms, the traditional electric vehicle design teaching model can transition to digitalization and intelligence, creating a personalized and precise learning ecosystem for students. AI-assisted teaching design encourages teachers to continuously learn new knowledge, enhance professional literacy, digital literacy, and teaching design capabilities, effectively improving classroom teaching efficiency while continuously stimulating students’ innovative thinking and problem-solving abilities.
Conclusion
By constructing a digital teaching resource system, an intelligent teaching platform, and adopting efficient teacher-student-machine triadic interactive teaching design, precise推送 of teaching content has been achieved, meeting students’ personalized learning needs. Simultaneously, multi-dimensional resource integration based on large language models has been completed, significantly enhancing teaching effectiveness. AI-empowered knowledge graphs inject innovative momentum into teaching content and methods, facilitating rapid updates of course content according to industry development trends and technological innovations, ensuring the前沿 and contemporaneity of teaching content. The smart course platform can stimulate students’ learning interest through personalized learning path recommendations and interactive learning tools, helping students enhance their innovative capabilities and digital literacy. The new paradigm for smart curriculum construction in electric vehicle design, aimed at new quality talents, is worthy of widespread application and promotion in the digital-intelligent transformation of higher education, particularly in programs focused on China EV development and global electric vehicle advancements.
The integration of AI and digital tools in electric vehicle education not only addresses current challenges but also prepares students for future innovations in the China EV sector. As the industry evolves, continuous refinement of this paradigm will be essential, leveraging data analytics and adaptive learning technologies to maintain relevance and effectiveness. This approach underscores the critical role of education in fostering the new quality productivity needed for sustainable growth in the electric vehicle market, both in China and worldwide.